Skip to main content

Full text of "Civil war and rebellion in the Roman empire A.D. 69-70; companion to the "Histories" of Tacitus"

See other formats


Digitized  by  the  Internet  Archive 

in  2008  with  funding  from 

Microsoft  Corporation 


http://www.archive.org/details/civilwarrebellioOOhenduoft 


CIVIL  WAR  AND   REBELLION 


ROMAN    EMPIRE 


MACMILLAN   AND  CO.,    Limited 

LONDON  •  BOMBAY  •  CALCUTTA 

MELBOURNE 

THE    MACMILLAN    COMPANY 

NEW  YORK  •  BOSTON  •  CHICAGO 
ATLANTA  •  SAN   FRANCISCO 

THE  MACMILLAN    CO.   OF  CANADA,   Ltd. 

TORONTO 


GALBA. 
From  a  bust  in  the  Capitoline  Museum,  Rome. 


w 
Civil  War  and  Rebellion 

in  the 

Roman  Empire 

A.D.    69-70 

A  COMPANION  TO  THE  'HISTORIES'  OF  TACITUS 

BY 

BERNARD   W.    HENDERSON,    M.A. 

SUB-RECTOR    AND    TUTOR    OF    EXETER    COLLEGE,    OXFORD 

SOMETIME    FELLOW    OF    MERTON    COLLEGE,    OXFORD 

AND    FELLOW    OF    UNIVERSITY    COLLEGE,     LONDON 

AUTHOR    OF 

'THE    LIFE    AND    PRINCIPATE    OF    THE    EMPEROR     NERO  ' 

'  HISTORY    OF    MERTON    COLLEGE,    OXFORD,'    ETC. 


WITH  MAPS   AND   ILLUSTRATIONS 


%0 


MACMILLAN   AND   CO.,   LIMITED 
ST.   MARTIN'S   STREET,   LONDON 

1908 


Ho 

Cop  S 


TO 

MY    "GREATS"    PUPILS    AT    EXETER    COLLEGE 

AND    IN    PARTICULAR 

TO   THOSE 

MY    COMPANIONS    ON    THE    ROAD    IN    ITALY 


PREFACE 

From  the  days  of  the  elder  Pliny  to  the  present 
there  have  been  many  who  have  written  concerning 
the  history  of  the  Civil  Wars  of  a.d.  69  and  70.  Of 
the  writers  whose  works  are  extant,  Tacitus  stands 
easily  first.  Without  his  "Histories"  we  should 
indeed  have  but  an  inferior  story  of  the  struggle 
between  the  Emperors  Otho  and  Vitellius,  and  a 
still  poorer  story  of  that  between  Vitellius  and 
Vespasian.  And  yet  always  from  the  very  first 
the  strategic  and  military  aspect  of  the  three  cam- 
paigns narrated  by  Tacitus  has  been  neglected. 
To  write  the  history  of  those  campaigns  by  the 
aid  of,  and  as  illustrative  of,  modern  strategical 
principles  is  the  main  purpose  of  this  book. 

Two  recent  writers,  it  is  true,  have  to  some 
extent  recognised  the  interest  and  value  of  such  a 
treatment  of  Tacitus'  story.  Gerstenecker  in  1882 
contributed  some  suggestions  on  the  military  history 
of  the  war  between   Otho  and  Vitellius.1     These, 

1  Der  Krieg  dcs  Otho  und  Vitellius  in  Italien  im  J.  6g,  von  Joh. 
Gerstenecker.  Programm  des  Kdniglichen  Maximilians-Gymnasiums  fur  das 
Schuljahr  1881-82,  Svo,  Miinchen,  1882,  pp.  Si. 


Vlll 


PREFACE 


however,  seem  to  me  of  peculiarly  little  value  in 
spite  of  their  considerable  length,  lacking  alike  in 
military  knowledge  and  in  insight.  Mommsen's 
short  paper,  on  the  other  hand,  published  in  1871, 
is  full  of  valuable  suggestions.1  It  will  always  be 
almost  impertinent  for  any  student  of  Ancient 
History  to  commend  any  paper  by  the  German 
master.  This  article,  however,  is  very  brief,  deals 
scarcely  at  all  with  the  strategy  of  the  campaigns 
before  the  actual  contact  of  the  opposing  armies 
upon  the  field  of  battle,  and  is  still,  I  think,  in  one 
or  two  respects  unduly  captive  to  Tacitus,  whom 
Mommsen  himself  has  called,  once  and  for  all  time, 
"  the  most  unmilitary  of  historians." 

For  in  very  truth  the  inadequate  and  short- 
sighted treatment  of  the  military  problems  and 
history  of  these  two  years  has  to  be  referred  back 
to  the  Roman  historian.  The  more  often  I  read 
Tacitus,  the  more  convinced  I  become  that  in 
matters  military  his  information  represents  little  but 
the  common  gossip  of  the  camp,  the  talk  of  the 
private  soldier  or  subordinate  officer,  reproduced  at 
second-hand  with  all  the  literary  power  of  a  great 
writer  who  possessed  the  most  vivid  visualising 
power  (if  I  may  so  call  it).  The  troops  on  the 
blood-stained  plain  outside  the  red  walls  of  Cremona 
battled,  as   it  were,  before   Tacitus'  very  eyes,  as 

1  Die  zivci  Schlachtcn  von  Betriacum  imjahre  6g  n.  C/ir.}  "  Hermes,"  Band 
v.  (1871),  pp.  161-173  ;  recently  republished  in  Mommsen's  Gesammelte 
Schriften,  Band  iv.  pp.  354-365,  8vo,  Berlin,  1906. 


PREFACE  ix 

he  sate  writing  in  his  study.  But  the  historian  was 
but  a  pleader  at  the  Roman  bar  who  had  taken  to 
history.  How  should  such  a  rhetorician  care  to 
inquire  very  deeply  into  the  strategical  causes  which 
led  to  that  battling  in  that  precise  position?  He 
seems  to  have  felt  no  interest  in  any  such  inquiry, 
and  distance  of  time  did  not  increase  for  him  clear- 
ness of  vision.  Generals  are  criticised  hastily ; 
impossible  plans  are  ascribed  to  them  ;  strategies 
are  ignored  or  misrepresented  ;  events  strategically 
connected  are  treated  as  isolated  movements  ; 
success  or  apparent  failure  is  the  one  criterion  of 
judgment.  With  all  this,  the  troops'  endurance  and 
pluck  are  rightly  recognised  ;  brilliance  and  "  dash  " 
are  duly  appraised.  But  the  result  of  such  an 
attitude  to  events  is  but  an  unsatisfactory  military 
history,  as  we  in  this  country  have  had  recent  cause 
to  know.  Yet  it  is  surely  the  military  history  of 
these  campaigns  which  is  of  great,  perhaps  chief, 
interest.  "  Nothing,"  remarks  a  modern  writer  of 
military  history,  "  is  so  misleading  as  the  camp 
gossip  which  is  reproduced  in  many  memoirs." ' 
Tacitus'  "camp  gossip"  has  been  too  faithfully 
repeated  as  the  whole  sound  sense  of  the  matter 
by  historians  who  have  had  to  rely  almost  entirely 
upon  his  narrative  for  their  facts. 

During  these  last   twelve  years  it  has  been  my 
good  fortune  to  roam  on  foot  many  times  in  different 

1  Maj.-Gen.  Sir  J.  F.  Maurice,  The  Diary  of  Sir  John  Moore,  ii.  p.  373. 


x  PREFACE 

parts  of  Italy.  If  in  this  book  I  now  attempt,  after 
two  recent  visits  of  my  own  to  the  actual  theatre  of 
war  in  the  Lombard  plain,  to  trace  again  the  history 
of  these  campaigns,  it  is  with  the  hope  chiefly  of 
calling  attention  to  a  somewhat  neglected  part  of 
them,  namely,  the  strategical  and  geographical 
questions  which  they  involve.  For  this  more 
prosaic  purpose  such  fineries  of  language  as,  for 
example,  adorn  Merivale's  record  of  these  wars 
cannot  be  allowed  to  me.  For  me  the  Vitellian 
columns  of  invasion  cannot  be  seen  "  beetling  on 
the  summits  of  the  Alps,"  nor  can  Otho  be  found 
"  bounding  from  his  voluptuous  couch  at  the  first 
sound  of  the  trumpet."  An  insistence  on  a  different 
method  of  treatment  of  these  wars  must  be,  if  it  so 
happen,  my  justification  for  yet  another  handling  of 
an  old  theme. 

BERNARD  W.  HENDERSON. 

Oxford,  March  1908. 


~\ 


CONTENTS 

CHAPTER    I 

The  Campaign  of  Otho  and  the  Vitellians 
January-April  a.d.  69 

§  1.  The  origins  of  the  civil  war  ;  the  fall  of  Nero  ;  the  rule 
of  Galba  ;  the  revolt  of  Vitellius  and  the  "  Army  of 
Germany  "  ;  the  accession  of  Otho    ....  1 

§  2.  The  troops  engaged  ;  the  military  system  as  cause  of 
the  civil  war  ;  strength  and  position  of  the  opposing 
armies ;  the  strategic  initiative  .  .  .  .  16 

§  3.   The  strategical   aspect  of  the  opening  campaign,   and 

opportunities  of  the  two  sides  .....  38 

§  4.   The  march  of  the  two  Vitellian  advance  columns  under 

Valens  and  Caecina  .  .  .  .  .  .  57 

§  5.  The  Othonjan  measures  of  defence  ;  use  of  the  fleet  ; 
the  mobilisation  of  the  "  Army  of  the  Danube  "  :  the 
loss  of  Cremona      .......  70 

§  6.   The  first   encounters  ;  assault   on    Placentia  ;  battle   of 

Locus  G&Stprum      .  .  .  .  .  .  .  81 

§  7.  The  strategies  of  the  final  struggle  ;  Vitellian  "  strategy 
of  penetration  "  ;  Otho's  Council  of  War  ;  Othonian 
"  strategy  of  envelopment "       .  .  .  .  .  92 

§  8.   The  "  Battle  of  Bedriacum  " 114 

§9.   The  death  of  Otho /124 

xi 


PAGE 


xii  CONTENTS 


CHAPTER    II 

The  Flavian  Invasion  of  Italy 
July-December  a.d.  69 

§t  ia  Vitellius  and  his  army  in  Rome  .  .  .  .  .128 

§  2.  The  gathering  of  the  storm  .  .  .  .  .132 

§  3.    Flavian  plans  of  war  : — 

The  muster  of  the  Eastern  army  ;  the  Council  of 
War  at  Berytus  ;  the  "  strategy  of  exhaustion  "  ; 
the  Council  of  War  at  Poetovio  ;  the  "  strategy 
of  annihilation "        .  .  .  .  .  .141 

§  4.   The  strategy  of  the  defence  .  .  .  .  .        165 

§  5.   The  strategies  compared      .  .  .  .  .  .173 

§  6.  The  second  "  Battle  of  Bedriacum  "  : — 

The  Flavian  advance  to  Verona ;  the  race  for 
Cremona  ;  the  battle  of  Cremona ;  the  sack  of 
Cremona  .  .  .  .  .  .  .185 

§  7.  The  advance  to  Rome  : — 

The   halt    at    Fano ;    movements    of    the   Vitellian 

forces  ;  the  capture  of  Rome     .  .  .  .206 

§  8.   The  death  of  Vitellius 228 


CHAPTER     III 

The  Rebellion  on  the  Rhine 

a.d.  69,  70 

§    1.   The  tribes  of  the  "  Low  Countries "      .  .  .  .        231 

§  2.   The  cause  of  the  rebellion  : — 

The  "  armed  plea  for  liberty "  ;  the  peoples  of  the 
revolt  ;  the  leaders :  Classicus,  Tutor,  Sabinus, 
Civilis     ........        237 


CONTENTS  xiii 


PAGE 


§  3.   The  Roman  army  on  the  Rhine  : — 

Its  composition  and  numbers  :    ( 1 )    legionaries,   (2) 

auxiliaries;  its  weakness  .  .  .  .        250 

§  4.   The  war,  up  to  the  relief  of  Vetera  : — 

The  clearing  of  the  "  Island"  ;  the  siege  of  Vetera  ; 
the  advance  of  the  relieving  army  ;  the  relief  of 
Vetera    .  .  .  .  .  .  .  .261 

§   5.   Flood  tide  :  the  success  of  the  Mutiny  : — 

The  retention  of  Vetera  ;  the  death  of  Vocula  ;  the 

loss  of  Germany       .  .  .  .  .  .276 

§  6.   The  Ebb  :  reduction  of  the  Gallic  revolt  : — 

The   gathering  of  the   Romans  ;    the  struggle  with 

the  Treveri  ;   the  advance  to  Cologne  .  .        290 

§  7.   The  submission  of  the  Germans  .....        306 

§  8.   The  lessons  of  the  Mutiny  : — 

The  results  in  Gaul  and  Germany  ;  the  results  in 
the  Roman  army  :  (1)  legionaries,  (2)  auxiliaries  ; 
the  Flavian  Army  of  the  Rhine ;  the  victory  and 
strength  of  Rome     .  .  .  .  .  .318 


NOTES 

To  Chapter  I. 


A.  The  Movements  of  the  Danube  Legions  . 

B.  The  Capture  of  Cremona  by  the  Vitellians 

C.  The  Site  of  Bedriacum  .... 

D.  The  "  Distances  "  in  Tacitus,  ii.  39,  40  . 

E.  Tacitus  as  Military  Historian 


337 
338 
339 
34o 
346 


To  Chapter  II. 

F.  Valens'  March  to  the  North    .....        348 

To  Chapter  III. 

G.  Vetera  and  Harper's  Ferry     .  .  .  .  350 
H.   The  Flavian  Army  of  the  Lower  Rhine  .          .          .        352 


ILLUSTRATIONS 

Galba  (from  a  Bust  in  the  Capitoline  Museum,  Rome) 
Otho  (from  a  Bust  in  the  Capitoline  Museum,  Rome) 
Vitellius  (from  a  Bust  in  Vienna)      .... 

Vespasian    (from    a    Bust   in   the   Museo   Nazionale, 
Naples)  ....... 

Coins  of  Galba,  Otho,  Vitellius,  Vespasian 


PAGE 

Frontispiece 
To  face        I 

,,  i  _u 

„  138 

„  26l 


Note. — The  busts  of  Galba,  Otho,   Vitellius,  and  Vespasian  are  reproduced 
from  Rom.  Ikonographie  ii.,  by  kind  permission  of  Prof.  J.  J.  Bernoulli. 


MAPS    AND    PLANS 


Diagram  . 

Diagram  . 

Diagram  . 

Diagram  . 

The  Theatre  of  War, 

Plan 

Diagram  . 

Diagram  . 

The  War  on  the  Rhi 

The  Via  Postumia 

Diagram  . 

Diagram  . 


68-69 


le,  A 


D.  69- 


70 


41 

48 

no 
199 

To  face    230 

233 
299 
302 
To  face  336 
339 

•  345 

•  35i 


OTHO. 
From  a  bust  in  the  Capitoline  Museum,  Rome. 


CHAPTER    I 

THE    CAMPAIGN    OF    OTHO    AND    THE    VITELLIANS 

After  a  war  one  ought  to  write  not  only  the  history  of  what  has  happened, 
but  also  the  history  of  what  was  intended  ;  the  narrative  would  then  be 
instructive. — Von  der  Goltz. 

§   i .    The  Origins  of  the  Civil  War 

Soon  after  daybreak  on  the  9th  of  June  a.d.  68 
the  Roman  Emperor  Nero  died  by  his  own  hand. 
He  who  had  been  for  thirteen  years  the  master  of 
the  Roman  world  ended  his  life  in  squalor  and  in 
misery,  with  only  three  freed  slaves  and  a  treacherous 
centurion  present  to  watch  his  death.  He  who  was 
the  last  descendant  of  Julius  Caesar,  the  last  Prince 
of  the  Julian  line,  enjoyed  for  resting-place  on  the 
last  evening  of  his  life  the  gloomy  underground 
cellar  of  a  villa  in  the  suburbs  of  his  capital ;  for  the 
furniture  of  his  death-chamber  a  scanty  mattress 
and  a  ragged  quilt ;  for  the  final  banquet  a  little 
lukewarm  water  and  old  crusts  of  bread.  Thanks 
to  others'  falseness  and  his  own  faint-heartedness 
he  had  to  die.  His  cruelty  and  lust  had  cost  him 
many  friends  ;  his  passion  for  art  and  music  had 
cost  him  more.      But  the  chief  cause  of  his  ruin  was r 

B 


2  THE  CAMPAIGN  OF  OTHO         ch.i 

the  indifference  shown  by  him  towards  his  troops, 
towards  the  art  of  war,  towards  the  practice  of  the 
camps.  The  nobles,  who  had  found  a  ruthless  perse- 
cutor in  him  ;  the  philosophers,  who  wrote  him  down 
a  frenzied  tyrant ;  the  Christians,  who  supposed  him 
to  be  the  Antichrist,  lord  of  a  world  abandoned  by 
God, — these  all  rejoiced  at  his  miserable  end  and 
defamed  his  memory.  But  the  lower  classes  in  Rome 
mourned  for  him.  Unknown  hands  yearly  decked  his 
tomb  in  the  gardens  of  the  Pincian  Hill  with  spring 
and  summer  flowers.  The  countless  inhabitants  of 
Italy  and  the  provinces  of  the  Roman  Empire  had 
no  reason  to  welcome  his  overthrow.  Not  a  few 
of  these  in  the  past  had  enjoyed  his  care  for  them, 
and  might  in  gratitude  sorrow  for  his  fall.  Neither 
had  the  Imperialist  any  reason  to  denounce  this  the 
last  Julian  Emperor.  Britain  had  been  well-nigh  lost, 
but  the  triumphant  courage  of  Nero's  legionaries  had 
saved  it  to  the  Empire.  The  war  upon  the  eastern 
frontier  with  Rome's  old  and  bitter  Parthian  enemy 
had  at  last  been  ended,  not  without  glory  to  the 
Roman  arms,  and  now,  after  a  century  of  hostility, 
there  was  a  fair  promise  that  the  agreement 
reached  would  be  an  enduring  peace  with  honour. 
But  all  such  blame  and  all  such  praise  availed 
Nero  little  when  his  soldiers  felt  no  love  for  him, 
and  had  no  reason  to  admire  him  or  fear  him  as 
their  General.  When  the  standard  of  rebellion  was 
raised  in  distant  Spain,  his  Guards  at  Rome,  piqued 
and  deceived,  deserted  their  Prince.  Nero, 
abandoned,  treacherously  betrayed,  slew  himself. 
The  whole   Empire,   if  it  had  good   cause   for  joy 


// 


sec.  i  AND  THE  VITELLIANS  3 

at  the  death  of  the  man,  had   speedily   reason   to 
regret  the  downfall  of  the  Emperor. 

For  now,  to  use  the  words  of  the  Roman  historian 
Tacitus,  the  secret  of  the  Empire  was  revealed. 
"A  prince  could  be  appointed  elsewhere  than  in 
the  city  of  Rome."  Hitherto,  under  Tiberius,  I 
Caligula,  Claudius,  and  Nero,  the  Imperial  power 
had  in  practice  been  but  the  heirloom  of  the  Julian 
family.  Now  the  last  of  the  family  was  dead.  Yet 
some  Emperor  there  must  be.  The  vast  body  of 
the  Empire  could  not  "  stande  without  governour."  * 
But  there  was  no  heir  to  the  throne.  The  Prince  1 
must  now  in  actual  fact  be  "elected,"  and  thus  the 
theory  of  election  which,  as  a  theory,  had  persisted 
from  the  beginning  must  be  realised  in  practice. 
Men  flattered  themselves  that  such  an  election  was 
a  sign  of  liberty  restored.  It  was  in  reality  no  gain 
to  liberty  that  the  might  of  armed  force  now  took 
the  place  of  such  a  right  as  inheritance  might  give. 
It  was  no  gain  to  liberty  that  "  two  common  soldiers 
of  the  line  took  upon  themselves  the  task  of  trans- 
ferring the  Empire  over  the  Roman  people  from 
one  Prince  to  another,  and  transferred  it." 2 

At  this  time,  in  fact,  the  army  of  the  Roman 
world  was  not  at  unity  with  itself.  Upon  the  death 
of  Nero  different  armies  in  different  quarters  of 
the  Empire  set  up  their  own  popular  leaders  and 
generals  as  claimants  to  the  Imperial  power.  Why 
should    the    legions    of    Germany,     or    the    proud 

1  Sir  H.  Savile's  translation  of  Tacitus,  Histories,  i.  16  (1591). 

2  Tacitus,  Histories,  i.  25.     All  references  henceforward  to  Tacitus  in  the 
notes  which  give  a  number  only  are  references  to  the  Histories. 


4  THE  CAMPAIGN  OF  OTHO         ch.i 

Praetorians  of  Rome,  submit  to  an  Imperator 
appointed  by  the  troops  in  Spain  ?  Why  should 
the  veteran  and  victorious  army  of  the  East  or  the 
hardy  garrison  of  the  Danube  frontier  tamely 
accept  an  Emperor  at  the  hands  of  the  rebel 
soldiery  of  the  Rhine  ?  The  miserable  death  of 
Nero  was  ominous  of  the  greater  misery  to  come, 
of  the  terrors  of  a  year  of  savage  civil  strife.  The 
Empire  was  the  prize  for  which  the  armies  battled  ; 
Italy  was  the  battle-ground.  Twice  within  eight 
months  armies  of  invasion  swept  down  over  the 
Alps  upon  the  unhappy  land.  "  Ah,  would  that 
Italy  had  never  been  dowered  at  Fortune's  hands 
with  the  luckless  gift  of  beauty ! "  cried  the 
Florentine  poet  of  the  seventeenth  century  :— 

Ch'  or  giu  dall'  Alpi  non  vedrei  torrenti 

Scender  d'  armati,  e  del  tuo  sangue  tinta 
Bever  1'  onda  del  Po  gallici  armenti. 

But  now  it  was  the  very  Empire  of  the  Roman 
world  which  called  the  rivals  down  to  Italy. 

Nero  had  been  indifferent  to  war  and  its  pursuits. 
Such  interests  were  unworthy  of  an  artist,  if  not 
of  an  Emperor.  This  indifference  on  his  part 
revenged  itself  upon  the  fairest  of  all  beautiful 
lands.  Four  Roman  Emperors  perished  within 
twenty  months.  Two  of  these,  Nero  and  Otho, 
fell  by  their  own  hand.  Two,  Galba  and  Vitellius, 
were  murdered  in  open  daylight  by  order  of  their 
conquerors.  The  death  of  each  of  these  selfish  and 
ambitious  princes  might  have  seemed  a  gain  to  the 
Roman  world,  had  not  each  been  followed  by  such 


AND  THE  VITELLIANS  5 

a  successor.  Then  at  the  last  Vespasian  came,  and 
the  land  had  peace.  It  was  always  Italy  which  paid 
the  chief  part  of  the  price  of  this,  the  contending  of 
the  Emperors.  Those  who  have  ever  seen  her 
dancing  sunlight  and  luxuriant  plains,  her  rushing 
rivers  and  her  sombre  mountains,  know  that  this 
land  alone  might  seem  worth  all  the  striving. 

Servius  Sulpicius  Galba1  had  already  been  in~ 
arms  against  his  Emperor  for  some  two  months  when 
the  Roman  Senate  elected  him  "  Princeps  "  on  the 
day  of  Nero's  death.  He  was  a  man  of  high  birth, J- 
descended  on  his  mother's  side  from  Lucius 
Mummius,  the  destroyer  of  Corinth  in  146  B.C. 
After  a  long  and  honourable  civil  and  military 
career  in  other  parts  of  the  Empire,  Galba  had  been 
sent  by  Nero  to  govern  the  province  of  Hispania 
Tarraconensis  (North -East  Spain)  eight  years 
before,  and  there  as  governor  he  had  stayed  ever 
since.  There,  too,  increasing  years  and  familiarity 
with  his  duties  had  presently  changed  him  from  a 
vigorous  and  efficient  governor  to  one  careless  and 
indolent.  No  one,  he  said,  had  to  render  an  account 
of  his  idle  hours.  But  an  alarming  rebellion  in  the 
neighbouring  land  of  Gaul  broke  out  in  the  spring 
of  a.d.  68,  and  compelled  him  to  take  action  either 
with  or  against  the  rebels.  Impelled  by  the  offers 
of  the  rebel  leader  Vindex  and  by  his  own  personal 
ambition,  he  chose  the  former  course  and  renounced 
his  allegiance  to  Nero.  The  Gallic  rebellion 
indeed  was  promptly  crushed  by  the  Roman  army 

1  Galba,  cf.  Suetonius,  Galba,  1-9;  Plutarch,  Galba,  3,  4,  8.      His  prae- 
nomen  Lucius  he  changed  to  that  of  Servius  in  A.D.  68. 


6  THE  CAMPAIGN  OF  OTHO 

in  the  district  of  Upper  Germany  under  its  famous 
general  Verginius  Rufus.  But  the  infection  of  dis- 
loyalty was  in  the  air,  and  even  Verginius'  victorious 
troops  were  eager  to  march  to  Rome  and  set  up 
their  general  there  as  Emperor.  But  Verginius 
was  well  content  with  his  achievement.  He  had 
saved  the  integrity  of  the  Roman  Empire  and  now 
would  preserve  his  own.  It  was  still  possible  to 
find  in  the  Roman  Empire  a  general  of  repute  who 
was  untainted  by  any  ambition  save  by  that  of 
serving  his  country.  He  declined  the  offered  gift 
of  Empire,  and  his  troops  had  sullenly  to  acquiesce. 
Galba  therefore,  despite  his  great  miscalculation, 
reaped  the  fruits  of  Verginius'  refusal,  and  had  his 
short-lived  joy  of  them.  He  was  now  an  old  man  of 
seventy-three  years  of  age,  but  the  crisis  called  out 
his  better  military  qualities.  On  receiving  from 
Rome  the  tidings  of  Nero's  death  and  of  his  own 
election  as  Princeps  he  marched  for  Rome  at  the 
head  of  a  newly-raised  legion,  the  Seventh  Galbiana,1 
and  found  his  progress  unopposed.  Towards  the 
middle  of  October  in  the  year  a.d.  68  he  entered 
the  city,  and  though  his  entry  had  been  marred 
by  scenes  of  needless  bloodshed  and  panic,  no 
rival  yet  disputed  with  him  his  possession  of  the 
Imperial  power.  For  some  three  months  after  his 
entry  Galba  remained  Emperor  of  Rome.  But 
then  the  end  came.  It  took  but  these  three  months 
for  him  to  lose  the  popularity  which,  by  remaining 

1  This  legion  later  was  known  as  the  Seventh  Gemina,  but  this  title  seems 
to  have  been  given  it  first  by  Vespasian,  when  he  disbanded  the  First  legion 
(so  Heraeus,  note  to  Tac.  i.  6).  Galba  evidently  sent  it  almost  at  once  to 
Pannonia,  as  it  is  found  here  in  January  A.D.  69. 


sec,  AND  THE  VITELLIANS  7 

unknown,  he  had  gained.  His  Ministers  and 
dependents  justly  earned  dislike  by  their  venality 
and  greed,  and  this  dislike  was  extended  to  the  old 
Emperor,  who  made  no  attempt  to  check  their 
rapacity.  His  own  severity,  amounting  in  cases  to 
cruelty,  his  age,  his  ugliness,  above  all  his  fatal 
parsimony,  cost  him  the  support  of  all  classes  in  the 
city,  who  were  quick  to  contrast  him  in  all  these 
respects  with  the  Nero  whom  he  had  supplanted. 
He  was,  it  is  true,  a  brave  disciplinarian,  and 
scorned  to  secure  by  purchase  the  doubtful  fidelity 
of  his  wavering  Guards.  The  exhausted  state  of 
the  Roman  Treasury  would  indeed  have  amply 
justified  the  greatest  thrift  and  the  most  careful 
financial  administration  on  the  part  of  any  ruler  save 
one  who,  like  Galba,  could  only  buy  the  goodwill 
of  the  soldiery  by  donatives,  the  affection  of  the 
unruly  populace  by  extravagance.  Tacitus'  biting 
epigram  has  characterised  Galba  for  all  time : 
"  Omnium  consensu  capax  imperii  nisi  imperasset."1 
The  disaffection  of  the  troops  in  Germany  and  the 
treachery  of  one  of  his  disappointed  adherents  in 
Rome  showed  how  shifting  and  unstable  was  the 
foundation  of  honesty  upon  which  Galba  had  striven 
to  build  his  rule.  It  was  not  for  the  enjoyment  of" 
such  an  Emperor  that  Nero  had  been  overthrown. 

The  trouble  began  in  "  Germany."  This  was 
the  name  given  by  the  Romans  of  this  time  to  the 
districts  lying  on  the  left  bank  of  the  river  Rhine 
from  Lake  Constance  to  the  sea.  Augustus  had 
renounced    the    attempt    to    add    to    the    Empire 

1  Tac.  i.  49. 


8  THE  CAMPAIGN  OF  OTHO         ch.  i 

territory  over  the  river,  and  the  German  savages 
between  the  Rhine  and  the  Elbe  remained  inde- 
pendent of  Roman  government  henceforward. 
Those  tribes  who  lay  immediately  opposite  the 
Roman  settlements  and  garrisons  on  the  left  bank 
were  to  a  certain  degree  civilised  by  their  acquaint- 
ance with  their  Roman  neighbours  and  Romanised 
kinsmen,  and  Roman  traders  ventured  in  their 
pursuit  of  wealth  to  penetrate  districts  which  were 
to  the  Roman  legionary  forbidden  land.  But  the 
venturesome  traders  took  their  lives  in  their  hands, 
as  they  had  done  among  the  independent  Gallic 
tribes  in  the  days  of  Julius  Caesar,  and  the  farther 
east  they  travelled  among  the  black  forests  and 
mountains  of  the  land  which  is  modern  Germany,  the 
more  barbaric  and  terrible  they  found  the  German 
tribes.  Migrations  of  whole  peoples  were  not 
uncommon,  and  each  tribe  lived  by  plundering  its 
neighbours  when  the  whim  seized  it.  Restless 
savagery  and  lust  for  bloodshed,  precarious  peace 
and  internecine  war,  such  were  the  pursuits  and 
characteristics  of  the  hordes  who  roamed  the  lands 
east  of  the  Rhine.  The  more  restless  cast  greedy 
eyes  on  the  fields  lying  west  of  the  river ;  the  more 
peaceable  were  driven  by  the  irresistible  pressure 
of  wild  tribesmen  from  the  unknown  forests  of  the 
interior  to  strive  to  put  the  barrier  of  the  river 
between  themselves  and  their  assailants. 

The  Roman  Empire  was  therefore  compelled  to 
"  police  "  its  side  of  the  Rhine  by  a  strong  stand- 
ing army.  For  this  purpose  the  left  bank  was 
marked  out  into  two  districts,   each  of  which  was 


AND  THE  VITELLIANS  9 

garrisoned  by  four  legions  with  auxiliaries  to  help 
them,  and  was  under  the  military  control  of  a 
governor,  the  Legatus  Augusti  pro  praetore. 
"  Upper  Germany  "  stretched  from  Lake  Constance 
to  a  point  midway  between  Coblenz  and  Bonn 
(now  Brohl,  between  Andernach  and  Remagen) ; 
"  Lower  Germany  "  reached  from  this  point  to  the 
sea.  For  civil  administration  "  Germany  "  belonged 
to  the  province  of  Gallia  Belgica  down  to  the  days 
of  Domitian  ;  for  financial,  at  least  half  a  century 
longer.1  But  the  governor  of  Belgica  had  no 
regular  troops  at  his  command,  so  pacified  by  now 
seemed  the  Gauls ;  and  the  two  governors  of  Upper 
and  Lower  Germany,  commanding,  as  they  did, 
powerful  armies  on  the  frontiers,  were  the  men 
on  whose  sagacity  depended  the  security  of  the 
Empire,  on  whose  fidelity  that  of  the  Emperor  at 
Rome. 

Galba  shortly  after  his  accession  had  recalled 
the  governor  of  Upper  Germany,  Verginius  Rufus, 
and  executed  the  governor  of  Lower  Germany, 
Fonteius  Capito.  To  take  their  places  he  had 
appointed  to  Upper  Germany  an  old  and  infirm 
man,  Hordeonius  Flaccus,  who  proved  utterly 
unable  to  control  turbulence  or  mutiny  among  his 
troops.  To  Lower  Germany  he  sent  Aulus 
Vitellius. 

Vitellius  was  then  fifty-five  years  of  age.  His 
career  up  to  that  time  had  been  a  curious  mixture 
of  good  and  evil.  As  a  boy  he  had  been  in 
attendance  upon  the  morose  old  Emperor  Tiberius 

1  See  below,  Chap.  III.  §  3. 


io  THE  CAMPAIGN  OF  OTHO 


CII.   I 


in  his  retreat  on  the  island  of  Capri,  and  men  were 
therefore  but  too  ready  to  speak  ill  of  him.  In 
Rome  he  had  won  the  young  Caligula's  favour  by 
his  skill  in  chariot-driving,  and  the  goodwill  of  the 
next  Emperor,  Claudius,  by  his  love  of  dicing.  But 
when  sent  out  as  governor  of  Africa  by  Nero  he 
too,  like  other  Roman  nobles  of  the  time,  left  his 
worst  qualities  behind  him  in  Rome,  and  displayed 
integrity  and  justice  in  his  administration,  so  that 
at  the  last  crisis  of  his  life  only  Africa  showed  any 
zeal  on  his  behalf.  He  had  returned  from  Africa 
in  a.d.  61,  and  lived  the  next  seven  years,  it  seems, 
in  obscure  retirement  at  Rome.  Either  his 
integrity  as  governor  or  his  gluttony,  which  was 
notorious,  reduced  him  to  such  straits  of  poverty 
that  when  Galba  commanded  him  to  proceed  as 
governor  to  Lower  Germany  in  the  autumn  of  a.d. 
68  he  left  his  family  behind  him  living  in  a  hired 
garret,  and  pawned  his  mother's  earrings  to  obtain 
the  money  necessary  for  his  travelling  expenses. 
By  such  means  he  was  able  to  reach  his  province 
on  the  ist  of  December  of  this  year. 

Both  new  governors  found  their  troops  sullen 
and  disloyal  to  Galba.  The  attempt  of  the  army 
of  Upper  Germany  to  proclaim  Verginius  Emperor 
had  recently  been  baffled  first  by  his  refusal,  and, 
soon  after,  by  his  recall  to  Rome.  But  they  loved 
Galba  none  the  better  for  that.  Galba  had  recently 
been  lavishing  favour  on  the  Gauls,  rebels  to  the 
Empire,  whom  they,  true  soldiers  of  the  Empire, 
had  lately  crushed.  Galba  was  but  the  nominee 
of  the  troops  in  Spain,  troops  whom  they,  the  proud 


sec.  i  AND  THE  VITELLIANS  u 

and  warlike  frontier  army  of  Germany,  could  have 
annihilated  with  ease.  Neither  governor  was  a 
disciplinarian  ;  neither  was  attached  to  Galba  by 
any  ties  of  affection  or  loyalty.  The  troops'  dis- 
content was  not  long  in  coming  to  a  head.  The 
legions  of  Upper  Germany  refused  the  military 
oath  of  allegiance  to  Galba  on  the  ist  of  January 
a.d.  69,  and  in  default  for  the  moment  of  a  rival  - 
Emperor  they  proclaimed  as  rulers  of  the  State  the 
Senate  and  People  of  Rome.  But  Republicanism 
had  never  any  real  influence  in  the  Roman  army 
after  the  days  of  Sulla  a  century  and  a  half  ago. 
The  legions  of  Upper  Germany  had  not  long  to 
wait  before  they  found  a  new  Emperor.  Next  day 
their  comrades  in  Lower  Germany,  who  the  day 
before  had  taken  the  oath  of  allegiance  to  Galba 
with  very  bad  grace,  renounced  it,  and  proclaimed 
their  governor,  Vitellius,  Emperor  at  Cologne. 
The  army  of  Upper  Germany  at  once  accepted  him, 
and  followed  the  example  on  January  3.  Vitellius 
for  his  part  was  far  too  slothful  and  too  flattered 
to  resist  the  dangerous  honour.  Two  men,  each 
of  them  in  command  of  a  legion,  both  of  great 
influence  with  the  armies,  found  it  an  easy  task  to 
persuade  him.  Fabius  Valens,  of  Anagni,  legate  Anagnia. 
of  the  First  legion  in  Lower  Germany,  was  an  able 
general  who  had  won  Nero's  favour  by  doubtful 
means  and  his  troops'  admiration  by  soldierly 
qualities.1  Aulus  Caecina  Alienus  of  Vicenza,  also  Vicetia. 
legionary  legate  in  Upper  Germany,  was  a  younger 
man   and   the   darling   of  the   troops.      Handsome, 

1  Valens,  cf.  Tac.  i.  52,  iii.  62  ;   Plutarch,  Galba,  10. 


12  THE  CAMPAIGN  OF  OTHO 

tall,  and  energetic,  he  was  also  to  show  true 
military  qualities  of  daring  and  resource.  He  had 
at  first,  when  quaestor  of  the  province  of  Baetica  in 
Spain,  been  a  partisan  of  Galba,  until  his  friendship 
was  changed  to  enmity  when  Galba  ordered  his 
prosecution  for  embezzlement.1  These  two  men, 
Valens  in  the  Lower  Province,  Caecina  in  the 
Upper,  worked  hard  to  secure  the  proclamation  of 
Vitellius  by  the  troops.  By  the  3rd  of  January 
their  object  was  won.  The  army  of  Germany  was 
united  in  its  declaration.  Vitellius  was  named 
Emperor,  and  open  defiance  hurled  in  Galba's 
face. 

When  in  a  few  days  news  of  this  reached  Rome 
the  old  Emperor  affected  to  make  light  of  it.  But 
it  finally  determined  him  to  take  a  step  which  he 
had  for  some  time  past  been  meditating,  and  to 
associate  with  himself  a  younger  man  as  colleague 
in  the  Empire.  There  was  both  good  precedent 
for  the  plan  and  also  every  hope  of  strengthening 
his  own  position  thereby,  had  he  chosen  his 
colleague  wisely.  Unhappily  for  himself,  Galba 
made  a  foolish  choice,  and  paid  for  it  in  a  week 
with  his  life. 

The  man  whom  he  presented  to  the  troops  and 
to  the  Senate  as  his  comrade  henceforward  in  the 
Iburdens  of  Empire  came  of  an  honourable  but 
unlucky  family.  Lucius  Calpurnius  Piso  was  by 
now  thirty  years  of  age.  Two  of  his  elder  brothers 
he  had  already  seen  slain — the  one    by  Claudius, . 

1  Caecina,  cf.  Tac.  i.  53,  iii.  S  :  "privala  mala  reipublicae  malis  operire 
statuit." 


AND  THE  VITELLIANS  13 

the  other  by  Nero.  He  himself  had  lived  long  in 
exile,  and  was  equally  without  experience  of  civil 
administration  or  military  service.  Staid,  sedate, 
melancholy,  he  was  a  man  on  whose  honour  the  old 
Emperor  could  rely  for  sober  counsel  and  loyal 
support.  But  he  was  not  a  man  to  gain  the 
devotion  of  the  Guards  or  fascinate  the  populace. 
And  even  on  the  very  day  of  his  adoption  by  the 
Emperor,  when  the  greedy  Praetorians  might  not 
unreasonably  have  received  the  donative  customary 
on  any  such  occasion,  Galba's  old-fashioned  thrift 
conceded  nothing.  His  maxim,  that  it  was  his 
wont  to  choose  his  soldiers  and  not  to  purchase 
them,  was  worthy  of  an  ancient  Roman,  but  won 
small  sympathy  from  the  Praetorians  of  his  day. 
Piso's  adoption  by  Galba  on  the  10th  of  January 
a.d.  69  was  received  sullenly  by  the  troops  in 
Rome — men  soon  so  resolute  to  fight  and  quick  to 
follow  a  general  whom  they  knew  and  loved,  but 
impatient  of  control  and  resentful  of  what  they 
deemed  neglect.  Civil  war  was  already  threatening,!  1 
and  military  discipline  is  the  first  virtue  to  fly  at  its- 
approach. 

The  discontent  of  the  Guards  was  all  the  more 
dangerous  because  it  quickly  found  a  leader,  in 
whose  heart  anger  at  Galba's  choice  of  Piso  burned 
all  the  more  deeply  because  he  himself  had  expected 
to  be  chosen.  And  indeed  Marcus  Salvius  Otho, 
of  Ferento  in  South  Etruria,1  had  some  reason  to 
indulge  in  his  hopes,  now  disappointed. 

1   In  Tac.  ii.  50,  Ferentio  must  be  read  instead  of  Ferentino.     Cf.  Suetonius, 
Otho,  I. 


14  THE  CAMPAIGN  OF  OTHO 

Otho  is  one  of  those  perplexing  figures  in 
history  whom  it  is  very  easy  to  condemn  and  very 
hard  to  dislike.  His  wayward  brilliance  and  calm 
courage,  his  strong  affections  and  the  gentleness 
and  mercy  which  he  showed  when  Emperor  even 
to  his  enemies,  were  qualities  which  endear  him  to 
the  memory  of  following  ages  as  they  won  for  him 
the  praise  and  the  love  of  the  Romans  of  his  own 
day.  Yet  his  youth  had  been  stained  by  vice, 
luxury,  and  immodesty,  and  he  gained  his  power  by 
base  treachery  and  murder.  But  the  men  of  his 
own  day  judged  these  faults  of  character  the  more 
leniently  as  they  were  the  more  familiar  with  them 
in  men  who  had  none  of  Otho's  charm  to  com- 
pensate. As  Nero  had  won  men's  approval,  so  did 
Otho  also,  and  when  the  careless  Roman  mob  nick- 
named him  Nero,  Otho  gladly  accepted  the  name  at 
their  hands. 

Now  in  these  early  days  of  January  Otho  had 
counted  on  Galba's  choice  falling  on  himself.  He 
had  done  good  service  to  the  Emperor  in  Spain. 
For  Nero  had  determined  to  take  Otho's  beautiful 
wife  Poppaea  for  his  own,  and  to  secure  this  end 
had  banished  the  husband  to  honourable  yet  real 
exile  as  governor  of  Lusitania,  the  modern  Portugal, 
-  in  a.d.  58.  Here  he  had  of  necessity  stayed  ten 
years,  surprising  all  who  had  known  his  dissolute 
life  in  Rome  by  his  suavity  and  uprightness,  when 
once  removed  from  the  accursed  atmosphere  of  the 
Court  at  Rome.  But  he  never  forgave  Nero  for 
Poppaea's  loss,  and  it  was  one  of  his  earliest  acts  as 
Emperor  to  set  up  again  the  statues  of  her  which 


AND  THE  VITELLIANS  15 

the  mob  had  overthrown.  Hence  when  Galba  had 
meditated  treason,  Otho  had  urged  him  on.  At  his 
side  he  had  come  to  Rome.  Presently  in  his  place 
he  had  hoped  to  reign.  Now  he  suddenly  found  a 
younger,  untried,  and  unpopular  man  preferred 
before  him.1 

It  was  an  age  when  few  men  in  high  places  - 
acted  on  any  principles  save  those  of  personal 
ambition ;  when  safety  was  sought  in  treachery ; 
when  treason  was  the  speediest  refuge  in  distress. 
Five  days'  plotting  followed.  Then  on  the  morning 
of  January  15,  Otho  left  the  side  of  the  old 
Emperor  Galba  as  he  stood  sacrificing — "  impor- 
tuning the  Gods  now  of  another  man's  Empire"2 — 
and,  muttering  some  lying  excuse,  hurried  to  the 
Praetorians'  Camp,  which  lay  by  the  city  wall  a 
short  distance  away. 

A  handful  of  troops  acclaimed  him  Emperor.- 
Galba  and  Piso,  lured  down  to  the  forum  from  the 
height  of  the  Palatine,  were  abandoned  by  an 
indifferent  mob  and  treacherous  soldiers  to  their  I 
fate,  and  Otho  reigned  sole  Emperor  of  Rome. 
Fourteen  days  before,  the  army  of  Germany  had 
proclaimed  Vitellius  Emperor.  The  rivals  must 
meet  in  open  war.  All  embassies  passing  between 
the  two  were  useless,  for  neither  would  yield  place 
to  the  other.  Galba  had  been  treacherously  slain. 
But  open  war  should  decide  between  Otho  and 
Vitellius. 

1  For  the  Nero-Otho-Poppaea  story  and  its  different  versions  see  my  Life 
and  Principate  of  the  Emperor  Nero,  pp.   1 1 6- 1 17,  467. 

2  Savile's  translation  of  Tac.  i.  29. 


16  THE  CAMPAIGN  OF  OTHO         ch.i 

§  2.    The  Troops  Engaged 

Civil  war  between  Otho  and  Vitellius,  the  first 
of  the  three  great  wars  of  these  years  a.d.  69  and 
70,  was  thus  imminent  in  the  month  of  January  in 
the  former  year.  The  various  parts  of  the  whole 
Roman  Empire  would  have  to  choose  sides.  Some 
of  the  provinces,  however,  were  "  unarmed,"  that 
is,  possessed  no  regular  troops  in  them,  and  their 
goodwill  or  hostility  therefore  counted  for  little  in  a 
struggle  which  only  the  sword  could  decide.  For 
at  this  time  the  Roman  army,  apart  from  the  garrison 
of  Rome,  was  for  the  most  part  distributed  along 
the  frontiers  of  the  Roman  Empire,  and  the  provinces 
within  those  frontiers  enjoyed  security  without  the 
presence  of  troops.  Even  of  the  frontier  provinces 
some  were  garrisoned  only  by  local  auxiliary  troops, 
and  their  contribution  to  the  military  strength  of 
either  side  could  be  but  trifling,  while  their 
sympathies  were  determined  by  the  wishes  of  a 
neighbouring  province  of  which  Roman  legionaries 
formed  the  garrison. 

The  Roman  army  at  this  time  consisted  of  thirty 
legions,  and  a  force  of  "  auxiliaries  "  which  probably 
equalled  in  strength  that  of  the  legions.1  The 
legionaries,  all  of  whom  were  Roman  citizens,  may 
have  numbered  upward  of  a  hundred  and  fifty 
thousand  men.  All  of  them  were  men  who  had 
made    the   practice    of  arms    their    profession ;    all 

1  This  is  generally  assumed,  and  is  a  conclusion  based  on  Tacitus  [Annals, 
iv.  5,  and  Histories,  v.  I)  ;  but  though  the  numbers  and  names  of  a  vast 
number  of  auxiliary  alae  and  cohortes  are  now  known,  it  is  quite  impossible 
to  supply  any  more  precise  data  of  their  total  strength. 


sec.  n  AND  THE  VITELLIANS  17 

of  them  were  heavy-armed ;  most  of  them  were 
disciplined  and  efficient.  Each  legion  bore  a 
number,  and  almost  always  a  distinctive  title  ;  and 
in  some  of  the  legions  regimental  pride  and 
loyalty  were  strong  inducements  to  valour.  The 
legionary  cavalry,  however,  were  few  in  number, 
and  the  bulk  of  the  horse,  as  well  as  considerable 
numbers  of  infantry,  mostly  light  -  armed,  were 
supplied  by  the  auxiliaries.  These  were  organised 
corps,  known  as  alae  (of  cavalry)  and  cohortes 
(both  infantry  and  cavalry,  or  infantry  only),  usually 
marked  by  a  number  and  a  special  name.  The 
name  was  sometimes  derived  from  the  man  who 
first  enrolled  the  corps,  sometimes  from  the  nation- 
ality of  the  troops  who  composed  it,  sometimes 
from  the  particular  equipment  which  distinguished 
it  from  other  troops.  These  special  corps  were 
either  five  hundred  or  a  thousand  strong.  The 
auxiliaries  for  the  most  part  were  at  this  time  not 
Roman  citizens,  but  earned  the  citizenship  by 
twenty-five  or  more  years'  service,  and  were  granted 
it  by  the  Emperor  on  their  discharge.  Legionaries 
served  twenty  years  with  the  colours,  but  after  that 
term  of  service  many  continued  in  the  army,  being 
formed  into  special  cohortes  veteranorum. 

The  Roman  military  system  was  thus  a  long- 
service  system.  And  although  a  legion  or  auxiliary 
corps  was  always  liable  for  service  in  any  part  of 
the  known  world,  there  had  been  developing  since 
the  beginnings  of  the  Empire  a  tendency  to  keep 
the  same  troops  in  the  same  province  for  years 
together,  and    to  recruit  the  legions  on  the  spot. 

c 


18  THE  CAMPAIGN  OF  OTHO 

The  legions  were  established  in  more  or  less  per- 
manent camps,  and  while  these  "  castra  stativa " 
served  as  headquarters  for  the  troops,  in  course  of 
time  civil  settlements  of  veterans,  with  their  wives 
and  families,  and  of  traders,  began  to  cluster  round 
the  military  lines.  Thus  the  children  of  the  legion- 
aries grew  up  in  close  touch  with  the  legions,  and 
the  children  of  the  auxiliaries  in  like  manner  would 
be  able  and  inclined  to  take  service  in  the  legions, 
for  which  service  they  were  duly  qualified  as  soon 
as  their  fathers  had  received  the  Roman  citizenship 
on  their  discharge.  The  problem  of  recruiting 
became  an  easy  one,  and  the  Roman  army  was  in 
truth  a  voluntary  army,  although  the  old  civic 
liability  on  every  citizen  to  be  called  out  to  war  was 
never  formally  abolished.  Always  every  citizen 
must  be  ready  and  able  to  fight  for  his  country 
if  need  arose.  But  since  the  days  of  Marius  the 
Roman  army  was  never  the  "  Nation  in  arms," 
except  in  theory.  In  compensation  for  this,  the 
experience  and  courage  of  the  legionaries  were  alike 
notable,  and  the  numbers  of  the  army,  though 
small,  were  adequate  for  all  the  work,  defensive  and 
offensive,  which  it  was  from  time  to  time  called  on 
to  perform.  Military  service  on  the  whole  was 
popular.  The  troops  were  well  cared  for  during 
service,  and  a  system  of  pensions  provided  comfort 
for  them  in  their  old  age.  The  permanent  camps 
upon  the  frontiers  were  centres  of  Romanisation 
and  civilisation  just  where  such  were  most  needed, 
namely,  on  the  outskirts  of  Empire,  where  Rome 
came  into  contact  with  still  uncivilised  and  savage 


sec  ii  AND  THE  VITELLIANS  19 

tribes.  The  camps  guarded  the  frontiers,  proved 
the  beginnings  of  towns  later  to  be  famous,  and 
were  places  of  refuge  when  the  unquiet  natives 
threatened  war.  And  the  steady  growth  in  the 
number  of  Roman  citizens  during  the  first  two 
centuries  of  the  Empire,  with  all  that  this  implied 
in  the  feeling  of  pride,  responsibility,  and  dignity, 
on  part  of  the  individual,  was  due  chiefly  to  the 
Roman  military  system  as  established  by  the  first 
and  greatest  of  the  Emperors,  Augustus. 

Upon  this  system,  now  comfortably  practised 
for  half  a  century,  and  upon  this  Roman  army  dis- 
tributed for  the  most  part  in  cantonments  along  the 
frontiers  of  the  Empire,  there  broke  the  storm  of 
civil  war.  Then  was  shown  the  one  great  blemish 
of  the  system ;  for  it  could  not  but  stimulate  the 
growth  of  local  sympathy  in  the  various  frontier 
armies  at  the  expense  of  their  loyalty  to  the  Empire 
as  a  whole  and  to  the  Emperor  at  Rome.  This 
danger  was  less  ominous  so  long  as  the  Emperor 
was  known  through  the  Roman  world  either  to  be, 
like  Tiberius,  a  soldier  himself,  of  tried  military 
capacity,  or  to  be  one  who,  like  Claudius,  would 
always  put  himself  at. the  head  of  his  troops — at  least 
at  the  end  of  a  difficult  or  dangerous  campaign. 
The  danger  was  also  less  ominous  if  the  governors 
of  the  frontier  provinces  were  changed  from  time  to 
time  and  not  allowed  protracted  periods  of  com- 
mand. Nero  had  been  the  first  Emperor  to 
disregard  both  principles  together.  He  had  in 
consequence  been  deserted  by  the  troops,  and 
perished.      The    danger    of  local    feeling,   of  local 


20  THE  CAMPAIGN  OF  OTHO         ch.i 

rivalries,  in  the  frontier  armies,  became  at  once 
pronounced,  and  the  length  and  bitterness  of  the 
civil  wars  of  a.d.  69-70  were  directly  the  result.  And 
hence,  when  finally  Vespasian  won  the  victory,  the 
interest  taken  both  by  him  and  by  all  his  successors 
in  the  army  and  its  welfare  is  very  marked.  Whether 
the  Emperor  were  a  man  of  war,  like  Domitian  or 
the  great  Trajan,  or  a  cultured  gentleman  and  man 
of  peace,  like  Hadrian  or  Antoninus  Pius,  or  a 
veritable  philosopher,  like  Marcus  Aurelius,  made 
no  difference.  He  was  bound  to  know  his  troops 
and  to  be  known  by  them. 

The  civil  wars  enforced  this  lesson  of  the  Roman 
military  system.  They  also  emphasised  another 
danger  of  the  system  which  becomes  clear  in  the 
great  native  rebellion  on  the  Rhine  towards  their 
close,  and  will  then  be  explained.  But  at  the  outset, 
when  the  Roman  legionaries  were  called  on  to 
choose  sides  between  Otho  and  Vitellius,  there 
seemed  no  reason  why  they  should  hesitate  to  take 
up  arms  for  the  one  or  the  other,  according  as  their 
private  interests  or  affections  or  passions  should 
command.  For  eighteen  months  selfishness  was 
lord  paramount  of  most  men  in  the  Roman  Empire. 

The  Roman  army  was  agreed  on  but  two  things: 
firstly,  that  it  would  not  restore  the  Republican 
form  of  government ;  secondly,  that  so  splendid  an 
opportunity  for  fighting  and  for  plunder  as  had  now 
arisen  was  not  to  be  let  go.  In  the  course  of  the 
struggle  the  troops  from  time  to  time  displayed 
courage  to  the  point  of  heroism,  and  loyal  affection 
for  at  least  one  of  their  generals  to  the  degree  of 


sec.  n  AND  THE  VITELLIANS  21 

the  very  passion  of  love  itself.  Yet  the  main 
interests  of  the  campaigns  are  strategical  and 
military.  They  are  no  battle  for  Country  or  for 
Liberty  when  war  is  glorious,  and  to  refrain  from 
arms  is  contemptible. 

The  Eastern  provinces  and  their  armies,  Italy 
and  the  garrison  of  Rome,  and  the  regular  troops 
of  the  "Danube"  frontier,  were  for  the  most  part 
in  sympathy  with  Otho ;  the  Western  half  of  the 
Empire  was  with  Vitellius.  But  some  of  the 
adherents  on  both  sides  were  too  far  removed 
from  the  scene  of  conflict  to  take  an  active  part 
in  it. 

The  army  of  the  Eastern  frontier  (including 
Egypt)  consisted  of  eight  legions.  Three  of  these 
— the  Fourth  Scythica,  Sixth  Ferrata,  and  Twelfth 
Fulminata — were  stationed  in  Syria.  The  governor 
of  Syria  at  this  time  was  Caius  Licinius  Mucianus, 
an  able  soldier  and  statesman,  who  had  been  ap- 
pointed to  this  duty  by  Nero  in  a.d.  67.  Three 
more  legions  were  still  engaged  in  quelling  the 
fierce  rebellion  of  the  Jews,  which  was  to  be 
ended  by  the  fall  of  Jerusalem  on  September  2, 
a.d.  70.1  These  legions  were  the  Fifth  Mace- 
donia, the  Tenth  Fretensis,  and  the  Fifteenth 
Apollinaris.  Their  general  was  Titus  Flavius 
Vespasianus,  at  this  time  a  man  of  fifty-nine  years 
of  age.  Vespasian  was  of  humble  origin,  from  a 
small  hamlet  near  Rieti  in  the  highlands  of  the  Reate. 
Abruzzi  in  Central  Italy,  but  of  long  and  honour- 
able service  and  of  proved  military  ability.     In  the 

1  For  this  Jewish  war,  see  my  Life  and  Principatt  of  Nero,  chap.  x.  §  5. 


22  THE  CAMPAIGN  OF  OTHO         ch.i 

first  conquest  of  Britain  under  Claudius  he  had 
subdued  the  Isle  of  Wight  after  many  battles,  and 
since  then,  after  peaceful  duty  as  governor  of  Africa, 
had  been  chosen  by  Nero  to  command  in  the  Jewish 
war.  A  keen  and  active  soldier,  blunt,  outspoken, 
hardy,  thrifty,  and  temperate,  a  man  possessing 
alike  common  sense  and  dry  humour,  Vespasian  was 
reserved  by  the  Fates  to  heal  the  wounds  of  the 
Roman  Empire  after  the  sore  months  of  civil  war 
were  ended.  But  in  January  a.d.  69  he  was  still 
busily  engaged  with  the  war  in  Judaea,  and  not 
ready  to  make  his  bid  for  Empire.  He  had  sent 
the  elder  of  his  two  sons  Titus  and  Domitian  to 
carry  his  homage  to  Galba  from  the  seat  of  war. 
But  the  news  of  Galba's  death  reached  Titus  when 
he  arrived  at  Corinth  on  his  voyage  to  Italy.  Titus 
therefore  returned  from  Greece  to  Syria,  and  both 
Vespasian  and  Mucianus  with  their  respective 
armies  swore  fidelity  to  Otho.  Finally,  in  Egypt 
there  were  two  legions,  the  Third  Cyrenaica  and 
the  Twenty-second  Deiotariana.  This  restless,  rich, 
and  turbulent  country  was  at  this  time  happily  con- 
trolled by  a  governor  of  striking  ability,  Tiberius 
Julius  Alexander.  He  was  a  Jew  by  birth  who 
had  renounced  Judaism,  and  after  serving  with 
Corbulo  in  the  Armenian  war  had  been  made 
Prefect  of  Egypt  by  Nero  in  a.d.  63,  and  continued 
in  that  office  under  Galba.  A  long  edict  by  him  is 
still  preserved,  checking  the  extortion  of  officials 
and  the  greedy  activity  of  professional  informers. 
Under  his  direction  Egypt  and  its  army  were  well 
disposed  to  Otho.     Subsequently  he  acted  as  chief 


sec.  ii  AND  THE  VITELLIANS  23 

of  staff  to  Titus  in  the  Jewish  war,  with  zeal  and 
ability. 

The  eight  legions  in  the  East  took  no  part  in 
the  civil  war  between  Otho  and  Vitellius,  but  their 
open  sympathy  with  the  former  quickly  bore  fruit 
after  the  triumph  of  his  rival.  For  the  time  being, 
however,  "  the  East  was  undisturbed."  l 

The  Roman  province  of  Africa  had  lately  been 
greatly  disturbed  by  the  foolish  ambition  of  the 
legate  of  the  one  legion,  the  Third  Augusta,  which 
at  this  time  occupied  the  military  district  of  Numidia 
(which  for  administration  counted  as  part  of  Africa). 
This  man,  Lucius  Clodius  Macer,  had  revolted 
against  Nero  and  posed  falsely  as  a  Republican 
enthusiast,  when  in  reality  he  was  seeking  his  own 
selfish  ends.  Galba  had  secured  his  death  without 
difficulty,  and  thereupon  the  province  was  only  too 
happy  to  be  quiet,  "  being  content  with  any  kind  of 
a  Prince  after  its  experience  of  a  petty  master." 
Following  the  lead  of  its  chief  town  Carthage,  it 
professed  mild  interest  in  Otho's  cause.  But  its 
legion  took  no  part  in  the  war.2 

If  Otho  was  thus  unable  to  use  nine  friendly 
legions  in  the  East  and  South,  Vitellius  in  like 
manner,  though  to  a  less  degree,  could  not  employ 
all  the  troops  who  wished  well  to  his  cause  in  the 
West. 

1  Position,  etc.,  of  the  legions:  Tac.  i.  10;  v.  I.  Mucianus  :  Pliny, 
Hist.  Nat.  xii.  9  ;  xiii.  88  ;  Tac.  i.  76  ;  ii.  5.  Vespasian  :  Sueton.  Vit. 
Vesp.;  cf.  Tac.  i.  50;  ii.  5,  6,  78;  iv.  v.  passim.  Titus:  Tac.  i.  10;  ii.  I, 
2.  Egypt  and  Tib.  Julius  Alexander :  Tac.  i.  1 1  ;  Ann.  xv.  28  ;  Josephus, 
Antiq.  xx.  5.  2  ;  Bell.  Jud.  ii.  15.  1  ;  18.  7,  8  ;  Corpus  Inscr.  Graec.  No. 
4957- 

2  Africa  :  Tac.  i.  11,  76,  78;  Macer:  Tac.  i.  7;  ii.  97;  Plutarch, 
Galba,  6. 


24  THE  CAMPAIGN  OF  OTHO         ch.i 

In  Britain  there  were  now  left  three  of  the 
famous  four  legions  which  had  "conquered"  the 
island  under  Claudius  and  kept  it  for  the  Empire 
at  the  time  of  the  furious  rising  of  the  natives  led 
by  Queen  Boudicca  (Boadicea)  against  Nero.  These 
legions  were  the  Second  Augusta,  Ninth  Hispana, 
and  Twentieth  Valeria  Victrix.  Their  old  comrades 
of  the  Fourteenth  Gemina  had  lately  been  trans- 
ferred to  Dalmatia.  The  governor  of  Britain  at 
the  time  of  Galba's  accession  was  Marcus  Trebellius 
Maximus.  But  he  incurred  the  displeasure  of  his 
troops,  and  the  men  of  the  Twentieth  legion  took 
upon  themselves  to  turn  him  out  of  the  country 
without  more  ado.  In  his  absence  the  legates  of  the 
three  legions  administered  the  province,  sharing  the 
duties  between  them,  until  Vitellius  after  his  victory 
sent  out  a  new  governor  in  the  person  of  Marcus 
Vettius  Bolanus.  Separated  by  the  sea  from  the 
rest  of  the  Empire,  and  with  trouble  threatening 
from  the  tribesmen  of  Yorkshire  and  the  north,  the 
army  of  Britain  displayed  no  lively  interest  in  the 
opening  stages  of  the  civil  wars.1 

Spain  was  garrisoned  by  the  two  legions,  the 
Sixth  Victrix  and  the  Tenth  Gemina.  Galba  had 
been  followed  as  governor  of  the  district  Tarracon- 
ensis  by  the  historian  Cluvius  Rufus2;  and  Otho, 
familiar  with  the  land,  and  a  fellow-courtier  with 
Cluvius  under  Nero,  had  hoped  to  gain  the  support 
of  the  Spanish  army  and  strengthen  his  position  in 
Spain  by  favours  bestowed  on  the  province.     New 

1   Britain  :  Tac.  i.  9,  60  ;  iii.  22  ;  cf.  my  Life  of  Nero,  chap.  vi. 
2  Clu%'ius  Rufus  :  cf.  my  Life  of  Nero,  p.  429. 


sec.  ii  AND  THE  VITELLIANS  25 

settlers  were  sent  by  him  to  the  two  colonies  of 
Hispalis  (Seville),  in  the  province  of  Baetica  in 
South  Spain,  and  Augusta  Emerita  (Merida)  in 
Lusitania  (Portugal),  and  the  Roman  citizenship 
was  probably  bestowed  by  him  on  the  Lusones,  a 
Celtiberian  tribe  round  the  sources  of  the  Tagus.1 
Certain  "  towns  of  the  Mauri "  also,  lying  in  the 
district  of  Mauretania,  the  other  side  of  the  Straits 
of  Gibraltar,  such  as  Tingi  (Tangiers),  were  added 
by  him  to  Baetica  for  administrative  and  revenue 
purposes.  But  neither  the  Spanish  provinces  nor 
the  Spanish  army  nor  Cluvius  Rufus  gave  Otho 
any  support.  It  may  well  be  that  the  soldiers 
resented  his  murder  of  their  former  governor  Galba. 
Indeed,  when  the  procurator  of  Mauretania,  one 
Lucceius  Albinus,  threatened  Spain  in  Otho's 
interests,  Rufus  guarded  the  Straits  and  persuaded 
Albinus'  army  to  murder  their  procurator.  This, 
however,  was  the  only  service  which  the  Spanish 
army  rendered  to  Vitellius  ;  and  later  they  aban- 
doned his  cause  when  his  doom  was  coming  close.2 

Of  the  thirty  legions  of  the  Roman  army,  four- 
teen, therefore,  were  not  concerned  with  the  war 
between  Otho  and  Vitellius.  The  remaining  sixteen 
were  divided  in  allegiance.  Seven  legions  in 
Germany  and  one  in  Gaul,  on  the  one  hand,  formed 
Vitellius'  army  of  invasion.      On   the  other  hand, 

1  Tac.  i.  78,  accepting  the  emendation  "  Lusonibus "  for  the  text 
"  Lingonibus  " — as  the  sentence  is  sandwiched  between  two  others,  both  of 
which  refer  to  Otho's  Spanish  measures.  Other  suggestions  are  Ilurconibus 
(Ilurco  =  Pinos  Puente  in  Baetica)  and  Lanciensibus,  a  Lusitanian  folk 
mentioned  in  an  inscription  of  A.D.  5.     Cf.  C.I.L.  ii.  460. 

2  Otho's  gifts  to  Spain  :  Tac.  i.  7S.  The  legions,  etc. :  Tac.  ii.  58  ; 
iii.  44  ;  Albinus,  ii.  58,  59. 


26 


THE  CAMPAIGN  OF  OTHO 


seven  stalwart  legions  in  the  "Danube  provinces" 
and  one  at  Rome  declared  for  Otho,  who  besides 
commanded  the  support  of  the  garrison  of  Rome, 
the  most  famous  regiments  of  "  Guards."  Four  of 
the  seven  legions  of  the  German  army  were  sta- 
tioned in  the  Lower  Province,  three  in  the  Upper ; 
all  seven  being  on  the  left  bank  of  the  Rhine  and 
in  the  immediate  neighbourhood  of  the  river. 
Nearest  the  sea  the  "  Old  Camp,"  Castra  Vetera, 
served  as  the  usual  headquarters  for  two  legions, 
the  Fifth  Alaudae  and  the  Fifteenth  Primigenia. 
The  First  legion  was  probably  encamped  at  Bonn, 
near  the  southern  frontier  of  the  province,  and  be- 
tween these  two  camps  the  Sixteenth  legion  lay 
at  Novaesium.  In  Upper  Germany  Mainz  supplied 
a  double  camp  for  the  two  legions,  the  Fourth 
Macedonica  and  Twenty-second  Primigenia,  while 
the  Twenty-first  Rapax  was  stationed  at  Vindonissa, 
east  of  the  great  northward  bend  of  the  Rhine.1 
The  four  Gallic  provinces — Belgica  on  the  north, 
Lugdunensis  in  the  centre,  Aquitania  on  the  south- 
west, and  Narbonensis  on  the  south-east — were 
sufficiently  guarded  against  the  barbarians  by  the 
powerful  garrison  in  the  two  Germanies  on  the 
Rhine,  and  only  the  city  of  Lugdunum  itself  at  the 
meeting-place  of  the  Rhone  and  Saone  was  guarded 


1  It  must  be  remembered  that  some  of  the  evidence  for  these  places  as 
headquarters  for  these  legions  belongs  to  the  year  70.  But  in  the  absence  of 
contrary  evidence  we  may  suppose  it  to  be  true  of  the  year  before.  The 
evidence  is,  for  Leg.  I. :  Tac.  i.  57 ;  iv.  25  ;  V.  and  XV. :  ibid.  iv.  22,  35  ; 
XVI. :  ibid.  iv.  26,  61;  IV.  and  XXII.:  ibid.  i.  55;  iv.  24,  25;  XXL: 
ibid.  i.  61  ;  iv.  61,  70.  Throughout  I  use  the  more  familiar  name  of  the 
place,  whether  ancient  or  modern,  placing  its  modern  or  ancient  equivalent 
in  the  margin. 


sec.  i.  AND  THE  VITELLIANS  27 

by  troops.  Here  in  the  most  important  city  of  the 
whole  land  were  stationed  the  First  Italica  legion  ; 
an  auxiliary  cavalry  regiment,  the  ala  Tauriana, 
so  called  from  the  name  of  Statilius  Taurus,  who 
first  enlisted  it ;  and  the  eighteenth  cohort  of  the 
Guards.1  The  town  and  garrison  of  Lugdunum 
embraced  eagerly  the  cause  of  Vitellius.  Twenty 
miles  down  the  Rhone,  nestling,  like  its  enemy, 
under  precipitous  heights,  lay  the  hated  city  of 
Vienne,  and  never  did  bitterness  felt  by  one  little 
Greek  city-state  for  a  neighbour  exceed  that  anger 
which  Lugdunum  cherished  against  Vienne,  both 
towns  though  they  were  of  the  Empire.  This  hos- 
tility was  doubtless  partly  due  to  tribal  feeling, 
partly  perhaps  to  the  very  Greek  element  persisting 
in  the  valley  of  the  Rhone.  And  it  is  curious  to 
reflect  that  just  as  Lugdunum  was  the  new  proud 
centre  for  the  great  political  institution  of  Caesar- 
worship,  so  it  was  at  Vienne  that  Christianity,  the 
foe  of,  and  at  last  the  victor  over,  the  Imperial  cult, 
had  its  chief  beginnings  in  the  West.  During  the 
recent  revolt  of  Vindex  and  his  Gauls,  Vienne  had 
been  enthusiastic  for  the  national  cause ;  Lugdunum 
had  triumphed  over  her  when  the  Roman  legions 
of  Germany  crushed  the  national  rising.2  Yet 
Galba,  Emperor  of  Rome,  had  shown  favour  to 
Vienne ;  and  Otho  had  actually  named  as  consul 
one  of  her  citizens.  Lugdunum  would  never  for- 
give this  injustice,  nor  forget  the  slight  put  upon 

1  See  below  for  this  cohort.     Evidence  for  the  garrison  of  Lugdunum, 
Tac.  i.  59,  64. 

2  For  Lugdunum  v.  Vienne  in  the  revolt  of  Vindex  see  my  Life  of  Nero, 
chap.  xi.  §  5  ;  Otho  and  Gaul,  Tac.  i.  76,  77. 


28  THE  CAMPAIGN  OF  OTHO 

her  loyalty.  Old  ally  of  the  Roman  legions  of 
Germany,  she  welcomed  the  day  when  Vitellius' 
troops  marched  down  stream  to  take  vengeance, 
she  hoped,  upon  her  enemies  and  those  of  Rome, 
as  well  as  on  his  own. 

The  military  camp  of  Vindonissa  was  but  a  few 
miles  from  the  eastern  border  of  Roman  Germany. 
The  point  where  the  Rhine  flowed  out  of  Lake 
Constance  marked  the  beginning  of  that  district, 
and  the  river  formed  its  frontier  from  the  lake  to 
the  sea.  A  few  miles  north  of  that  camp  there  rose 
in  the  recesses  of  the  Black  Forest  a  far  mightier 
and  more  wonderful  river,  the  course  of  which  from 
its  source  to  its  outflow  into  the  Black  Sea  traced  for 
many  years  the  northern  limit  of  the  Roman  advance 
into  the  heart  of  Europe.  Along  its  southern  bank 
the  Romans  made  four  districts.  Raetia,  the  most 
westerly,  stretched  from  the  frontier  of  Upper 
Germany  at  Lake  Constance,  and  south  of  this  from 
the  land  of  the  Helvetii  and  the  Lake  of  Geneva,  to 
the  valley  of  the  lower  Inn  and  the  point  where  this 
muddy,  rushing  river  joins  its  waters  to  the  cool, 
clear,  beautiful  stream  of  the  Danube  at  Passau. 
From  this  point  Noricum  reached  well-nigh  as  far 
as  Vienna,  to  be  succeeded  by  the  province  of 
Pannonia,  whose  northern  and  eastern  boundaries 
alike  were  formed  by  the  Danube.  The  issue  of 
the  river  Save  from  the  west  into  this  river  at 
Belgrade  was  the  beginning  of  the  province  of 
Moesia,  which  stretched  all  the  many  hundred 
miles  from  this  point  along  the  southern  bank  to 
the  sea. 


sec.  it  AND  THE  VITELLIANS  29 

As  the  Danube  far  excels  the  Rhine  in  power, 
so  did  the  native  tribes  north  of  the  river  surpass 
those  beyond  the  Rhine  in  restlessness  and  terror. 
If  the  "  Army  of  the  Danube  "  in  the  middle  of  the 
century  was  slightly  inferior  in  numbers  to  the 
troops  who  guarded  the  line  of  the  Rhine,  this  was 
due  to  the  fact  that  the  greater  danger  which 
threatened  the  Roman  peace  from  the  trans-Dan- 
ubian  peoples  seemed  less  imminent  during  the 
first  half  of  the  first  century,  and  came  to  be  realised 
more  and  more  vividly  only  towards  its  close.  None 
the  less  there  were  already  seven  legions  appointed 
to  form  this  army.  In  Raetia  and  Noricum  no 
legionaries  were  stationed.  Just  as  in  earlier  days 
the  master  of  an  army  in  Cisalpine  Gaul  between 
the  Alps  and  the  Rubicon  had  held  the  key  to  Italy 
in  his  grasp,  so  now  when  Italy  reached  as  one 
country  to  the  Alpine  chain  the  governor  of  the 
district  of  Raetia,  which  commanded  all  the  northern 
passes  over  the  mountains,  would  have  had  Italy 
and  the  Emperor  of  Rome  too  directly  and  immedi- 
ately at  his  mercy,  had  he  been  placed  in  control  of 
a  legionary  army.  Raetia,  therefore,  was  but  a 
minor  command  under  control  of  a  procurator,  and 
his  only  troops  were  such  native  levies  as  he  could 
raise  in  the  case  of  any  sudden  peril.  If  a  serious 
danger  threatened  the  Raetian  frontier,  the  governor 
of  Upper  Germany  must  see  to  it.  In  the  same 
way,  Noricum  was  administered  by  a  procurator 
only,  who  depended  for  his  protection  ultimately 
on  the  legions  in  the  province  of  Pannonia  on 
his  eastern  border.     The  great  frontier  commands 


3o  THE  CAMPAIGN  OF  OTHO 

were  those  of  Pannonia  and  Moesia.  In  Pannonia, 
in  January  a.d.  69,  were  two  legions — the  Seventh 
Galbiana  and  Thirteenth  Gemina;  in  Moesia  three 
— the  Third  Gallica,  which  had  newly  come  to  the 
province  from  Syria,  the  Seventh  Claudia,  and  the 
Eighth  Augusta ;  and  as  a  great  reserve  force  to 
the  army  of  the  Danube,  two  legions  kept  the 
province  of  Dalmatia — the  Eleventh  Claudia,  and 
the  most  famous  of  the  legions  of  the  war  in  Britain, 
the  Fourteenth  Gemina.  This  province  of  Dal- 
matia extended  all  down  the  eastern  shore  of  the 
Adriatic  Sea,  from  the  promontory  of  I  stria  on  the 
north  to  the  Macedonian  frontier  by  Lissus  on  the 
south.  Thus  its  army  acted  as  a  great  rearguard 
alike  to  the  troops  in  Pannonia  on  the  north  and  to 
those  in  Moesia  on  the  east.  And  its  two  legions 
are  justly  counted  as  part  of  the  Army  of  the 
Danube.1 

These  legions  of  Dalmatia,  Pannonia,  and  Moesia 
duly  swore  allegiance  to  Otho  in  the  early  part  of 
the  year  a.d.  69.  They  were  too  widely  separated 
from  one  another  to  be  likely  eagerly  to  act  in 
concert  on  behalf  of  some  nominee  of  their  own.2 
Moreover,  Otho  was  the  emperor  accepted  by  the 
Senate  and  People  of  Rome,  ruling  from  the  city  of 
Rome,  which  was  the  very  "head  of  the  Empire 
and  glory  of  all  the  provinces,"  3  and  there  was  no 

1  Legions  in  Pannonia:  Tac.  ii.  II,  86;  in  Moesia,  i.  79;  ii.  85;  iii. 
24;  C.I.L.  iii.  2715;  in  Dalmatia,  ii.  11.  For  the  Fourteenth  legion 
in  Britain  see  my  Life  of  Nero,  pp.  200,  211  so.  For  VII.  Galbiana  (later 
Gemina)  see  note  to  p.  6,  above. 

2  Cf.  Tac.  i.  9  :  "  Longis  spatiis  discreti  exercitus,  quod  saluberrimum 
est  ad  continendam  militarem  fidem." 

3  "Caput  imperii  et  decora  omnium  provinciarum,"  i.  84. 


sec.  ii  AND  THE  VITELLIANS  31 

reason  why  they  should  love  the  rival  set  up  by  the 
rebellious  army  of  Germany.  The  appreciation 
of  Otho,  however,  was  of  a  slightly  passive  nature, 
save  in  the  case  of  the  legion  from  Britain,  the 
Fourteenth,  which  was  enthusiastic  on  his  behalf.1 
Moreover,  the  governor  of  Moesia,  Marcus  Aponius, 
and  his  troops  had  suddenly  in  the  spring  of  this 
year  a  task  imposed  upon  them  which  left  them 
small  leisure  for  pondering  over  Otho's  virtues. 
Already,  in  the  winter,  the  Roxolani,  a  tribe 
belonging  to  the  wild  Sarmatian  hordes  of  Eastern 
Europe,  had  raided  across  the  frontier  and  cut  to 
pieces  two  cohorts  of  auxiliaries.  And  now  in  the 
early  spring  they  repeated  their  attack,  encouraged 
by  the  rumours  of  civil  war  among  their  foes  which 
had  rapidly  spread  to  them.  Nine  thousand  horse- 
men clad  in  chain  armour  or  leathern  jerkins  crossed 
the  Danube  to  plunder.  Then  the  Third  legion, 
with  auxiliaries  to  help  them,  did  good  work  for 
Rome.  On  a  February  day,  when  the  deep  snows 
of  winter  were  melting  and  rain  was  falling,  when 
the  rude  tracks  were  well-nigh  impassable,  and 
horses  could  scarcely  keep  their  feet,  the  Roman 
infantry  fell  unexpectedly  upon  the  straggling  and 
unsuspecting  foe,  and  had  them  at  their  mercy. 
They  were  unable  to  ride  away ;  they  were  hurled 
from  their  horses  or  pierced  by  the  Roman  javelin  ; 
they  lay  prostrate  in  the  watery  snow,  and  were 
unable  to  struggle  to  their  feet  for  the  weight  of 
their  armour ;  their  pikes,  their  long  swords,  need- 
ing two  hands  to  wield,  were  useless ;  shields  they 

1  Tac.  ii.  11. 


32  THE  CAMPAIGN  OF  OTHO         ch.  i 

had  none,  nor  any  courage  to  defend  themselves 
on  foot  against  the  exultant  legionary  and  his 
native  ally.  They  put  their  trust  in  their  horses, 
and  they  were  destroyed  with  an  utter  destruc- 
tion by  the  short  stabbing-blade  of  the  Roman. 
Only  a  handful  of  the  raiders  escaped  to  lurk 
in  the  marshes,  there  miserably  to  die  of  cold  or 
wounds.1 

Otho  gladly  seized  the  chance  given  him  by  the 
annihilation  of  the  Sarmatians  to  reward  the  governor 
of  Moesia  and  the  legates  of  all  three  legions, 
though  only  one  of  the  three  had  won  the  victory. 
But  it  was  no  time  for  making  distinctions  and 
exciting  jealousies.  It  was  for  him  to  gain  the 
goodwill  of  all  the  officers.  And  the  Danube  army 
might  reasonably  now  be  proud  of,  and  loyal  to,  an 
Emperor  on  whose  brief  annals  they  had  been  the 
first  to  inscribe  deeds  worthy  of  remembrance. 
With  this  intent,  Otho  celebrated  their  prowess  and 
published  it  abroad  in  Rome.  And  the  seven 
legions  of  the  Army  of  the  Danube  were  true  to 
him  to  the  day  of  his  death. 

There  remained  the  garrison  of  Rome  itself, 
which  was  devoted  to  his  cause,  and  such  other 
troops  as  were  to  be  found  in  Italy.  In  Rome  itself 
at  this  time  a  large  body  of  troops  was  gathered 
together.  Foremost  among  these  were  the  soldiers 
who  formed  the  regular  garrison  of  the  city — nine 
cohorts  of  Praetorian  Guards,  and  seven  cohorts  of 


1  Tac.  i.  79.  Cf.  the  "Ob  Laurum,"  ap.  Acta  fratrum  Arvalium  for 
March  1.  For  earlier  relations  with  the  Roxolani  under  Nero,  cf.  C.I.L. 
xiv.  360S,  and  my  Life  of  Nero,  p.  225. 


sec.  ii  AND  THE  VITELLIANS  33 

Urban  Guards.  The  strength  of  a  cohort  was  one 
thousand  men.  The  Praetorians,  the  only  troops 
whom  Italy  contributed  regularly  to  the  Imperial 
forces,  were  the  very  flower  of  the  Roman  army. 
All  had  volunteered  for  the  service,  which  lasted 
for  them  only  sixteen  years,  and  was  rewarded  by 
higher  pay  than  that  which  the  legionary  received. 
They  were  commanded  by  two  Prefects,  men  speci- 
ally chosen  by  the  Emperor  out  of  the  staff  of  his 
own  Civil  Service,  and  this  Prefecture  formed  the 
crown  of  the  Service.  The  seven  Urban  cohorts 
were  men  well  trained  and  fought  well,  but  they 
were  held  in  less  repute  than  the  "  Guards  "  proper. 
In  these  sixteen  cohorts  Otho  had  a  force  of  sixteen 
thousand  men,  the  most  famous  regiments  in  the 
army,  and  all  enthusiastic  for  a  Prince  to  whom 
they  had  given  the  power. 

Besides  these,  there  was  then  at  Rome  the  First 
Adjutrix  legion,  newly  levied  by  Nero  from  the 
sailors  of  the  fleet,  and  recently  given  its  legionary 
"  eagle,"  the  ensign  and  emblem  of  due  enrolment, 
by  Galba.  But  on  their  entry  into  Rome  Galba's 
troops  had  hewn  down  many  of  these  legionaries, 
who  had  poured  out  of  the  city  somewhat  turbu- 
lently  to  greet  the  new  Emperor  and  clamour  for 
their  "  eagle."  Hence  he  had  been  frightened, 
and  had  given  orders  to  slay.  The  sole  tradition, 
therefore,  of  the  new  regiment  was  one  of  hatred 
for  the  dead  Prince,  and  it  could  be  trusted  to  serve 
his  slayer  well.  Moreover,  its  spurs  were  yet  to 
win,  and  the  men  were  not  unmindful  of  this  when 
a  few  months  later  they  stood  face  to  face  in  their 

D 


34  THE  CAMPAIGN  OF  OTHO         ch.i 

first  battle  with  a  veteran  regiment  of  the  German 
army. 

Besides  the  Guards  and  this  legion  there  also 
chanced  to  be  in  Rome  detachments  of  troops 
properly  belonging  to  the  armies  of  Britain,  Ger- 
many, and  the  Danube.  For  just  before  his  death 
Nero  had  summoned  these  "  vexilla  "  and  "  numeri  " 
to  Italy  when  he  was  making  preparations  for  an 
expedition  against  the  Alans  in  the  Caucasus.  The 
revolt  of  Vindex  had  caused  him  to  call  these  forces 
hurriedly  to  Rome,  where  they  remained,  it  seems, 
after  his  death  and  during  the  short  Principate  of 
Galba. 

Finally,  there  was  one  cohort,  the  Seventeenth, 
in  garrison  at  Ostia 1 ;  and  upon  the  river  Po, 
whither  the  storm-clouds  of  invasion  were  sweeping, 
one  auxiliary  squadron  of  horse,  nine  hundred  and 
sixty  strong,  the  ala  Siliana,  was  stationed.  Nero 
had  called  it  to  Italy  from  Egypt  on  the  news  of 
Vindex's  revolt,  and  sent  it  north  to  guard  the  line 
of  the  river.2 

These  then  were  the  troops  which  might  be 
counted  on  by  both  sides  as  able  to  take  part  in  the 
coming  civil  war  : — 

1  The  Cohortes  XVII.  and  XVIII.  are  counted  on  from  the  Urban 
Cohorts  (Coh.  X.-XVI.),  just  as  these  are  counted  on  from  the  Praetorian 
(Coh.  I. -IX.).  Under  Tiberius,  Cohors  XVII.  is  in  garrison  at  Lugdunum 
"ad  monetam"  (Tac.  Ann.  iii.  41  ;  C.I.L.  xiii.  1499;  Tac.  Hist.  i.  80). 
But  as  in  a.d.  69  it  is  found  at  Ostia,  it  is  probable  that  Nero  recalled  it 
from  Lugdunum  and  sent  it  there,  placing  Cohors  XVIII.  at  Lugdunum 
instead.      Cf.  Hirschfeld,  ap.  C.I.L.  xiii.  p.  250. 

For  Leg.  I.  Adjutrix  and  the  garrison  of  Rome,  cf.  Tac.  i.  6,  31  ;  ii.  n. 
The  Caspian  expeditionary  troops,  cf.  my  Life  of 'Nero,  p.  227,  and  references 
in  note. 

2  The  ala  Siliana,  so  called  from  C.  Silius,  legate  of  Upper  Germany  under 
Tiberius.      Tac.  Ann.  i.  31  ;  Hist.  i.  70. 


sec.  ii  AND  THE  VITELLIANS  35 

(1)  On  the  Vitellian  side: 

Legio  I.  in  Lower  Germany,  at  Bonn 

Legio  V.  Alaudae  ,,  ,,  at  Castra  Vetera 

Legio  XV.   Primigenia  „  „  „ 

Legio  XVI.  „  „  at  Novaesium 

Legio  IV.   Macedonica,  in  Upper  Germany,  at  Mainz 
Legio  XXII.  Primigenia  ,,  „  „ 

Legio  XXI.  Rapax  „  „  at  Vindonissa 

Legio  I.  Italica  in  Gaul  at  Lugdunum 

Ala  Tauriana  ,,  „ 

Cohors  XVIII. 
and  an  indeterminate  but  large  number  of  auxiliaries,  horse  and 
foot,  besides   such   irregulars,    native  levies,   etc.,   as   could    be 
enlisted. 

(2)  On  the  Othonian  side: 

Legio  VII.  Galbiana,  in  Pannonia 

Legio  XIII.  Gemina  „ 

Legio  III.  Gallica,  in  Moesia 

Legio  VII.  Claudia         „ 

Legio  VIII.  Augusta       „ 

Legio  XL  Claudia,  in  Dalmatia 

Legio  XIV.  Gemina  „ 

Legio  I.  Adjutrix  in  Rome 

Cohortes  I.-IX.  (Praetorian)  „ 

Cohortes  X.-XVI.  (Urban) 

Cohors  XVII.  at  Ostia 

Ala  Siliana  in  Upper  Italy 
and  an  indeterminate  but  large  number  of  auxiliaries,  horse  and 
foot,    besides   such   irregulars,   native   levies,  etc.,   or  gladiators 
from  the  schools  at  Rome,  as  could  be  enlisted. 

The  struggle,  therefore,  seemed  likely  to  be  one 
between  the  troops  in  Germany  and  Gaul  on  the 
l'"t>he'side,  and  those  of  Italy  and  the  Danube  pro- 
vinces on  the  other.      In  this  reckoning  each  of  the 
rivals  could  employ  eight  legions. 


36  THE  CAMPAIGN  OF  OTHO         ch.j 

Vitellius'  army  numbered  over  a  hundred  thou- 
sand men  of  all  arms.  When  he  moved  on  Italy, 
his  two  advance  columns  consisted,  the  one  of  forty 
thousand,  the  other  of  thirty  thousand  men,  when 
they  left  the  Rhine,  and  the  column  which  marched 
through  Gaul  continually  gathered  in  fresh  troops 
from  the  country.  Vitellius  himself  followed  later 
with  the  rest  of  his  available  strength,  and  he  too 
received  reinforcements  on  the  march,  as  by  this 
time  the  army  in  Britain  found  itself  able  to  con- 
tribute to  his  forces.  Only  a  few  men  were  left 
behind  under  Hordeonius  Flaccus  to  garrison  the 
bank  of  the  Rhine  against  the  peril  from  the  natives 
over  the  river,  who  happily  for  the  time  remained 
quiet.1 

To  resist  this  attack,  Otho  had  two  armies  to 
put  into  the  field.  At  Rome  his  army  cannot  have 
largely  exceeded  twenty  -  five  thousand  men  in 
number,  though  the  majority  of  these  were  excellent 
soldiers.  But  in  the  Danube  provinces  his  troops, 
when  and  if  concentrated  into  a  single  striking 
force,  would  scarcely  be  inferior  to  the  German 
army  in  number.  The  Emperor  could  reckon  them 
as  at  least  upwards  of  seventy  thousand  men  of  all 
arms. 

In  actual  numbers,  as  also  in  the  probable  quality 
of  the  soldiers,  Otho  was  thus  not  inferior  to 
Vitellius.     But  one  great  difference  in  his  situation 

1  Tac.  ii.  57-  The  numbers  of  the  columns  of  Caecina  and  Valens  are 
given  precisely  by  Tacitus  ;  but  Vitellius  himself  is  described  as  to  follow 
"tota  mole  belli,"  i.  6i.  If  we  insisted  on  this  very  vague  and  worthless 
phrase,  we  might  increase  Vitellius'  available  numbers  to  150,000  men.  But 
the  lower  total  seems  to  me  the  more  probable. 


nc.ii  AND  THE  VITELLIANS  37 

became  clear  at  once.  The  Army  of  Germany  was 
more  easily  concentrated,  more  easily  set  in  motion 
under  one  command.  His  own  troops  consisted  of 
two  widely  separated  armies — the  smaller  Army  of 
Italy,  the  larger  Army  of  the  Danube.  But  this  last 
army  also  was  far  from  being  concentrated.  The 
obvious  base  for  its  military  operations,  and  there- 
fore its  place  of  muster,  was  Aquileia,  the  town  look-j 
ing  due  southwards  over  the  Adriatic.  Aquileia  is 
to-day  a  petty  village  in  the  marshes,  some  fifty-five 
miles  north-east  of  Venice.  But  in  Roman  days, 
when  Venice  did  not  exist,  it  was  the  most 
important  military  stronghold  on  the  Italian  north- 
eastern frontier,  and  the  great  military  roads  from 
the  "  Danube  provinces,"  Dalmatia,  Pannonia,  and 
Moesia,  first  converged  upon  it.  The  chief  military 
centre  of  Pannonia,  the  town  of  Poetovio,  lay  a 
hundred  and  fifty  miles  to  the  east  of  it ;  the  chief 
town  of  Moesia,  Naissus,  some  four  hundred  miles 
beyond  Poetovio,  and  from  Scodra,  chief  town  of 
Dalmatia,  to  Aquileia  direct  was  also  four  hundred 
miles.  There  is  no  evidence  of  the  actual  position 
of  the  legions  of  the  three  provinces  in  January 
a.d.  69.  But  it  was  evident  that  to  muster  the 
whole,  or  even  a  considerable  part  of,  the  Army 
of  the  Danube  at  Aquileia  would  require  much, 
organisation,  and  take  probably  a  longer  time  than 
was  needed  for  the  mobilisation  of  the  German 
army  within  striking  distance  of  Italy.  Although) 
Otho  sent  orders  at  once  to  the  nearest  legions, 
those  of  Pannonia  and  Dalmatia,  to  march  on  Italy, 
yet  the  enemy  possessed  the  great  initial  advantage  j 


38  THE  CAMPAIGN  OF  OTHO         ch.  i 

of  a  concentrated  over  a  dispersed  army.  The  very- 
force  of  circumstances  therefore  dictated  the  initial 

.  strategy  of  the  coming  campaign.      The  strategic 

— >  initiative   rested   with    the  Vitellians.      It   was   not 

only  because  Otho  was  actual  Emperor  of  Rome 

land  Vitellius  challenged  his  right  to  rule,  that  the 
Army  of  Germany  had  to  attack.  Before  a  soldier 
left  camp,  the  strategy  of  offence  and  invasion  was 
clearly  marked  out  for  the  Vitellians  by  the  position 
of  the  opposed  armies  and  their  numbers.     When 

—  Otho's  scattered  armies  were  united,  it  might  well 
be  that  they  would  prove  more  than  equal  to 
Vitellius'  troops.     But  at  the  outset  there  existed 

—  great  gaps  between  the  dispersed  fragments  of  the 
Othonians.  While  they  were  mustering,  Vitellius 
must  strike.  Possibly  even  he  might  have  the 
opportunity  to  penetrate  between  the  foe's  divided 

f  forces.  This  "  strategy  of  penetration  "  gives  the 
chance  of  dividing  up  and  defeating  the  enemy  in 
detail.  It  has  risks  and  obvious  perils  of  its  own, 
and  all  its  success  depends  upon  careful  timing. 
But  at  the  outset  it  was  clear  to  all  concerned  that 
circumstances  and  numbers  dictated  to  the  Vitellians 
the  strategy  of  offence  and  the  invasion  of  Italy. 
They  possessed  the  strategical  initiative. 


/ 


§  3.    The  Strategical  Aspect  of  the   Opening 
Campaign 

The  Army  of  Invasion  had  two  ultimate  bases  of 
operation.  These  may  be  taken  to  be  Cologne  for 
the  force   in  Lower  Germany,  and  Vindonissa   for 


sec. in  AND  THE  VITELLIANS  39 

that    in  Upper  Germany.     The  objective  of  both 
forces    was    the    enemy's    army,    which    must    be  j 
destroyed.     That   army   was  not   likely  to  be  met 
north  of  the  Alps,  nor  indeed  north  of  the  Po,  for 
reasons    partly  of  time,   partly  of  strategy,   which 
were  obvious  to  both  sides.1     The  theatre  of  war 
was  likely  to  be  the  great  plain  of  this  river,  that  i 
plain   which  has  been  the  scene  of  more  fighting  I 
in  the  course  of  history  than  even  have  the  Low 
Countries.     The  immediate  geographical  objective, 
therefore,  of  the  Army  of  Invasion  was  the  section  i 
of   the    Po    between    Placentia    on    the    west    and 
Hostilia  on  the  east.     At  the  former  place  was  the 
crossing  of  the  river  by  the  great  highway  which 
led  south-east,  skirting  the  Apennines,  to  Ariminum 
and  so  to  Rome  ;  and  this  road  would  have  to  be 
pursued    by    a    force    crossing    any  of  the    Alpine 
passes  on  the  west  and  north-west  of  Italy.     And 
at   Hostilia  was  the  second   chief  crossing   of  the 
river  by  the   road  which  ran  from  Verona  on  the 
north  to  join  the  great  highway  at  Bologna,  fifty  Bononia. 
miles    to    the    south ;   and    by  this    road    an    army 
marching  by  any  of  the   northern   passes  down  on 
Italy  would  have  to  come. 

It    was    therefore    necessary    for    the    Army    of  I 
Germany  to  cross  the  Alps  as  speedily  as  possible.  I 
The  natural  difficulties  of  the  passage  of  the  moun- 
tains in  early  spring  by  a  large  force,  as  well  as  the 
problem    of  supplies,    made    it    expedient    for   the 
Vitellians  to  divide  their  army.      Moreover,  it  was 

1  I  postpone  the  explanation  of  these  to  the  paragraphs  dealing  with  the 
strategical  position  of  the  Othonians. 


\ 


40  THE  CAMPAIGN  OF  OTHO 

important  to  secure  Gaul,  as  this  country  lay  on 
the  flank  and  in  rear  of  the  advance,  and  further  to 
increase  the  numbers  of  the  invading  army  by 
Gallic  reinforcements  swept  in  during  the  forward 
movement.  But  if  the  whole  army  marched 
through  Gaul  and  over  one  of  the  western  passes, 
the  delay  caused  by  the  long  detour  might  well 
imperil  the  success  of  the  whole  campaign.  There- 
fore it  was  decided  that  the  Army  of  Germany 
should  remain  divided,  and  that  two  columns  of 
invasion  should  march  at  once.  The  Vindonissa 
column  was  to  proceed  direct  from  Upper  Germany 
to  Italy;  the  Cologne  column,  as  it  may  be  called, 
was  to  march  through  Gaul,  and  strike  thence  east- 
wards over  one  of  the  western  passes  of  the  Alps. 
The  distance  to  be  marched  by  the  Cologne  column 
was  nearly  three  times  as  great  as  tKaTof  the  other. 
It  would  arrive  later  at  the  objective,  and  there  join 
the  Vindonissa  army,  should  the  latter  need  help. 

The  Arnry_jof  Defence  had  also  two  ultimate 
bases  of  operation — Rome  for  the  Army  of  Italy,  and 
Aquileia  for  the  Army  of  the  Danube.  Both  of 
these  were  similarly  many  miles  away  from  the 
river  Po,  and,  besides  this,  the  concentration  of  the 
whole  or,  at  least,  part  of  the  Army  of  the  Danube 
at  Aquileia  must  first  be  effected.  A  diagram  may 
serve  to  illustrate  the  strategical  position  at  the 
beginning  of  the  campaign.1 

Although  the  distance   from   Rome   to   the  ob- 


1  The  diagram  is  drawn  roughly  to  scale  according  to  the  distances  which 
the  troops  marched  by  the  ways  they  actually  pursued,  and  these  are  the 
distances  given  in  round  numbers. 


sec,  in  AND  THE  VITELLIANS  41 

jective  was  longer  than  that  from  Vindonissa,  the 
time  taken  by  an  army  marching  from  Rome  would 
be  much  shorter,  as  the  natural  difficulties  which 
hindered  the  pace  of  the  Vindonissa  column  were 

1?  Cologne 


IN 


// 


1*0 

vo 

\»  jo  Poetovio 

»  rf\Ue?-"     (Pannonia) 


25  mjJ£HAquileia 
Pi3r,»nf;^Objective-r^^-' 
-zrt. „ —       7-6^miles  \HostUia  \ 


Rome 


o  Scodra 
(Oalmatia) 


far  greater,  and,  as  it  proved,  these  troops  indulged 
in  some  petty  fighting  with  the  tribes  north  of  the 
Alps  before  they  set  out  resolutely  on  the  road. 

A.  Strategical  Opportunities  of  the  Othonians. — 
Until  the  Danube  army  arrived  in  North  Italy  to 
co-operate  with  them,  Otho's  troops  in  that  country 


42  THE  CAMPAIGN  OF  OTHO 

were  so  greatly  inferior  in  numbers  to  the  approach- 
ing invaders  that  their  only  possible  strategy  at 
first  was  a  defensive  one.  It  is  true  that  such  a 
strategy,  unless  it  were  unexpectedly  crowned  by  a 
decisive  victory  on  the  field  of  battle,  could  never 
be  expected  to  end  the  war. 

The  records  of  warfare  contain  no  instance,  when  two 
armies  were  of  much  the  same  quality,  of  the  smaller  army 
bringing  the  campaign  to  a  decisive  issue  by  defensive  tactics. 
Wellington  and  Lee  both  fought  many  defensive  battles  with 
inferior  forces.  But  neither  of  them  under  such  conditions 
ever  achieved  the  destruction  of  the  enemy.  They  fought  such 
battles  to  gain  time,  and  their  hopes  soared  no  higher. l 

Defence  was  forced  upon  the  Army  of  Italy  until 
their  comrades  should  arrive,  but  only  for  so  long. 
For  defence  pure  and  simple  sometimes  wins 
battles,  but  wars  scarcely  ever. 

It  was  therefore,  above  all,  important  to  retard 
the  advance  of  the  Vitellians  into  Italy  by  every 
possible  means.  All  Othonian  efforts  in  Italy  had 
to  be  directed  at  first  to  secure  this  end,  and  to 
give  time  for  the  Army  of  the  Danube  to  arrive. 
The  questions,  therefore,  which  arose  were  two. 
Firstly,  what  precise  line  of  defence  should  be 
chosen  ?  Secondly,  what  means  of  delaying  the 
enemy's  march  could  be  employed  ? 
(i)  Line  of       (i)  Two  possible  lines  of  defence  suggest  them- 

Defence. 

selves  at  once  to  a  general  who  wishes  to  defend 
(a)  The      North  Italy,  namely,  the  Alps  and  the  Po. 

But  for  Otho  the  blocking  of  the  Alpine  passes 
J    was   impossible.     In   the   first   place,   had   he  even 

1  Stonewall  Jackson,  by  Lieut. -Col.  Henderson,  vol.  ii.  p.  228. 


sec.  in  AND  THE  VITELLIANS  43 

wished  to  block  them,  time,  distance,  and  numbers  i 
forbade  this.  Actually  he  had  in  North  Italy  in 
January  a.d.  69  but  one  small  regiment  of  horse, 
the  ala  Siliana,  and  this  quickly  turned  traitor  to 
his  cause.  The  troops  in  Rome  could  scarcely 
reach  the  Alpine  passes  on  the  north  and  north- 
west before  the  troops  of  Upper  Germany  had/ 
seized  them.  And  it  would  be  madness  for  them 
to  block  the  western  passes,  whither  they  might 
have  arrived  in  time,  when  the  foe  advancing  from 
the  north  would  already  be  down  in  the  plain  of 
the  Po.  But  even  if  the  Vitellians  delayed  their 
approach,  and  thus  gave  Otho  time  to  block  the 
passes  (and  he  could  not  count  upon  this  for  a 
moment),  the  Emperor  was  quite  uncertain  which 
route  or  routes  his  foe  would  choose.  The  Army 
of  Italy,  scarcely  twenty-five  thousand  strong,  would 
have  been  distributed  along  the  chain  of  mountains 
in  isolated,  widely  separated  fragments.  A  reverse 
suffered  in  any  single  pass  would  snap  at  once 
the  chain  of  resistance.  The  whole  scheme  of 
defence  would  have  been  destroyed,  and  the  entire 
army  would  have  been  in  danger  of  piecemeal 
annihilation. 

In  the  next  place,  the  proper  method  of  defend- 
ing a  mountain  ridge  is  not  the  blocking  of  the 
passes,  when  several  such  passes  over  the  ridge 
exist.  To  place  a  division  sitting  on  top  of  each 
pass  in  entrenchments,  however  strong,  is  but  to 
court  disaster.  No  mountain  barrier,  whether 
Himalaya  or  Pyrenees,  Jura  or  Alps,  ought  to  be 
defended  in   this  way,  or  ever  has  been   for  long 


44  THE  CAMPAIGN  OF  OTHO 

successfully  defended  in  this  way.  Picquets  and 
outposts,  varying  in  strength,  must  be  placed  in 
tne~ actual  passes.     But  the  main  Army  of  Defence 

|  /must  be  kept  on  the  more  level  ground  behind  the 
'ridge,  concentrated  and  as  near  to  the  issues  of  the 
passes  as  the  nature  of  the  ground  allows.  From 
such  a  position  it  can  deal  a  vigorous  blow  at  its 
foes  when  these,  forcing  back  the  outposts,  struggle 
by  one  or  more  passes  with  difficulty  over  the 
mountains,    and    emerge    more    or    less    exhausted 

i  upon  the  lower  ground  beyond.  It  is  then  that 
they  must  be  attacked,  before  they  have  recovered 
from  the  stress  of  the  passage  of  the  heights,  when 
a  dangerous  country  lies  immediately  in  their  rear, 
and  when,  if  they  have  chosen  to  cross  by  more 
passes  than  one,  the  detachments  of  their  troops 
are  perhaps  separated  by  the  difficult  foothills  of 
the  mountain  ridge.  Then  the  Army  of  Defence, 
perfectly  informed  by  its  outposts  of  the  advance 
of  the  enemy,  with  its  communications  from  the 
flanks  to  the  centre  running  easily  over  the  more 
level  country  which  the  army  occupies,  can  move 
to  the  attack  with  vigour  unimpaired  and  confidence 
high,  and  by  a  tactical  offensive  give  its  strategical 
defensive  the  victory.  Such  was  the  strategy  by 
which,  for  instance,  the  Argives  ought  to  have 
defended  their  northern  rampart  of  mountains 
against  King  Agis  of  Sparta  in  418  B.C.  Such  is 
the  strategy  by  which  Italy  to-day  would  defend 
her  Alpine  barrier  against  a  foe  to  north  or  west 
of  it. 

Unhappily,  Otho  had  neither  men  enough   nor 


sec.  m  AND  THE  VITELLIANS  45 

time  enough  to  choose  this,  which  otherwise  would 
have  been  the  right,  method  of  defending  Italy. 
He  was  compelled  to  abandon  all  thought  of  hold-  \ 
ing  the  line  of  the  Alps.  He  could  not  prevent 
the  enemy's  columns,  marching  by  widely  different 
routes,  from  concentrating  in  the  plain  of  the  Po 
unhindered.  In  modern  history,  in  the  Napoleonic 
wars  and  in  the  fighting  for  the  liberation  of  Italy, 
"the  battles  lost  or  won  at  the  foot  of  the  Alpine 
passes,  and  in  the  vineyards  of  the  great  northern 
plain,  Rivoli,  Marengo,  Magenta,  Solferino," * 
decided  then  too  the  fate  of  Tuscany,  Rome,  and 
the  South.  As  Otho  could  not  guard  the  foot  of 
the  passes,  he  must  fall  back  upon  the  second 
natural  line  of  defence — upon  the  river  which  flows:j 
through  the  great  northern  plain  and  its  vineyards. 

This  line  could  be  more  easily  defended.     To  (^)ThePo. 
the  west  lay  the  great  fortress  of  Placentia,  south' 
of  the  river,  placed  upon  the  military  road  where 
it   crossed  the   Po,   and   guarding   the   passage  of 
the    river.      Placentia    if  garrisoned    strongly   and  Piacenza. 
resolutely  held  would  be  an  invaluable  "  pivot   of 
manoeuvre "    for    Otho's    defending    army,    which,  I 
with   its   left   flank  secured  by  the   fortress,,  could 
deploy  eastwards  along  the  river  in  safety. V  In  the 
same  way  the  crossing  of  the  river  to  the  east  must  | 
be  secured  and  defended,  and  at  the  same  time  the 
communications  with  the  Danube  army  at  Aquileia 
must  be  kept  open  and  safe  from  the  enemy.     A 

1   Trevelyan,  Garibaldi's  Defence  of  the  Roman  Republic,  p.  45. 
C2±3*  A  '  pivot  of  manoeuvre '  is  a  force,  fortress,  or  natural  obstacle  which 
secures  a  flank  "  (Henderson,  Science  of  War,  p.  64). 


46  THE  CAMPAIGN  OF  OTHO 

i    strong  garrison  at  Verona  or  at  Mantua  would  best 

i   achieve  this  double  object.      It  was  vital  to  Otho  to 

\  take  precautions  against  the  risk  that  the  enemy 

would  come  down  upon  Italy  by  the   Brenner  Pass 

and  seek  to  thrust  in  between  his  own  army  and 

that    at    Aquileia,    severing    the    communications 

;  between  these.     At  least  the  Mantua-Hostilia  line 

I  must  at  all  costs  be  stoutly  defended. 

The  Army  of  Italy,  therefore,  should  be  spread 
\  along  the  line  of  the  river  from  Placentia  to 
'  Hostilia,  with  special  concentration  of  strength  at 
both  ends  of  the  line.  And  as  at  the  western  end 
the  fortress  in  itself  offered  a  means  of  strong 
defiance,  the  bulk  of  the  defending  forces  must  be 
directed  to  the  eastern  part  of  the  line  of  defence. 

This  line  the  Vitellians  would  doubtless  assault 
with  vigour.  But  it  was  unlikely  that  they  would 
try  to  break  it  in  the  middle,  at  least  at  first,  or 
that,  if  they  tried,  they  would  succeed  in  the 
attempt.  The  river  here  is  wide  and  deep,  with 
shifting  sandbanks  and  dangerous  eddies,  and  its 
current,  swollen  in  spring,  is  impetuous.  It  was 
far  more  probable  that  they  would  attack  one  of 
'  the  two  ends.  A  successful  forcing  of  the  eastern 
end  would  indeed  be  ominous  of  disaster  for  Otho. 
His  army  here  must  see  to  it  that  this  did  not 
/  happen.  But  the  point  of  attack  nearest  to  the 
most  probable  place  of  concentration  for  the 
--  Vitellians  in  North  Italy  was  certainly  Placentia. 
If  then  the  enemy  combined  to  assault  this  fortress, 
if  they  even  forced  the  passage  of  the  river  here, 
then  at  once  the  advantage  which  Otho  possessed 


/ 


sEcm  AND  THE  VITELLIANS  47 

in  his  double  base  of  operations  would  come  into  \ 
play.      For  as   the   Vitellians   advanced   down   the 
great  road  from  Placentia,  the  Othonians  defending 
the  river  could  retire  before  them  unhurt,  and  fall 
..  back  upon  their  second  base  Aquileia.     This  would 
compel  the  enemy  to  choose  one  of  two  courses  of 
action.      They  might  either  neglect   this   force  or/ 
pursue  after  it.     If  they  dared  to  neglect  it,  and  to 
press   on   regardless  down   the  great   highway   for 
Rome,  by  so  doing  they  would  expose  their  own/ 
line  of  communications  defenceless  to  the  force  at 
Aquileia.     This  then,  strengthened  by  the  arrival-*/ 
of  the   Danube  army,  would  sally  forth  to  cut  the 
line.     Now  it  is  one  of  Napoleon's  sayings  that  the-^/ 
secret   of  war  lies   in   the   communications.      It   is 
true  that  under  exceptional  circumstances  an  army 
can  afford  to  cut  itself  loose  from  its  line  of  com- 
munications   with    the    base — when,   that    is,    it    is 
prepared  to  live  entirely  upon  the  country  through" 
which  it  is  marching.      But  for  the  most  part  in  all, 
warfare,  ancient  as  well  as  modern,  an  army  needs ' 
to  keep  its  communications  open  with  some  friendly 
base  in  the  rear  of  its  advance  for  the  safe  convoy 
of  supplies  and  reinforcements,  and  if  it  is  invading 
a  hostile  land  it  is  likely  to  be  extremely  sensitive 
as  to  the  perfect  safety  of  its  line  or  lines  of  com- 
munication   with    the    rear.       By    neglecting    this  j 
principle    Alexander   at    Issus    was    trapped    in    a  | 
hopeless  position,   unless   he  won   a  great   tactical 
victory.     Napoleon  at  Madrid  hurriedly  abandoned 
all  his  year's  schemes  for  the  conquest  of  Portugal 
because  a  small  British  force  moved  boldly  out  in 


48  THE  CAMPAIGN  OF  OTHO 

the  far  north  of  Spain  to  threaten  his  line  of  com- 
munications with  France.  Therefore  a  Vitellian 
invading  force  advancing  down  the  road  to  Rome 
was  not  likely  to  allow  the  enemy  to  cut  the  one 
line  by  which  reinforcements  could  come  to  it,  the 
one  line  by  which  its  own  retreat,  in  case  of 
disaster,  was  secured.  Threatened  by  an  advance 
from  Aquileia,  the  Vitellians  would  surely  turn  to 
face  the  advancing  foe.  They  would  then  find 
themselves  in  a  position  which  is  the  most  hazardous 
position  for  an  army  compelled  to  fight  a  decisive 
tactical  engagement.  This  is  the  position  techni- 
cally known  as  that  of  an  army  with  its  "  front  to  a 
flank."     A  diagram  may  make  this  clear : — 

The  Vitellians  in 
this  position  have  been 
formed  to  a  front  on 
their  left  flank.  When 
a  force  is  thus  drawn 
up,  the  enemy's  main 
attack  is  always  directed 
*c  upon  the  flank  which  is 

A  =  Vitellian  base,  over  the  river  beyond    nearest       tO      the       base. 
Placentia.  -r-.         >r   ,1         c  u 

BC  =  Direction    of    Vitellian    march     for     "  OT  II    me    lOrCe    Can    De 

c**£TZ**  »„— non,  defeated   on   this    flank 

w.th  their  base.  its  yme  0f  communica- 

D  =  Direction  of  Othonian  attack  on  CB 

from  Aquiieia.  tions     is     thereby    cut, 

E  =  Vitellian  front  to  meet  the  attack. 

and  the  whole  force  is 
separated  from  the  hope  of  succour.  Whereas  if  it 
is  worsted  on  the  flank  farthest  removed  from  the 
base,  the  line  of  communications  is  still  open  to 
the  defeated  army,   and   retreat,    if  difficult,   is  at 


sec. in  AND  THE  VITELLIANS  49 

least  possible  for  them.  As  a  general,  if  not  his 
army,  must  always  take  into  account  his  position 
in  the  event  of  defeat  as  well  as  in  that  of  victory 
(unless  he  is  staking  all  on  a  single  throw,  and 
wishes  for  no  choice  save  that  between  victory  and 
annihilation),  the  Vitellian  commander  could  not 
contemplate  with  equanimity  an  advance  which 
might  compel  him  at  any  moment  to  form  front  to 
a  flank  in  face  of  the  enemy,  if  he  was  unwilling 
to  surrender  altogether  his  line  of  communications 
to  their  mercy. 

If  then  the  Vitellians  forced  the  passage  of  the 
river  at  Placentia,  it  was  more  probable  that  they 
would  not  straightway  pursue  their  march  south- 
east along  the  road.  They  would  rather  follow 
upon  the  heels  of  the  retiring  Othonians  towards 
Aquileia.  This  would  suit  Otho  well.  He  would 
be  retreating  in  the  direction  of  the  advancing 
Army  of  the  Danube,  and  the  aim  of  his  defence 
of  the  river — concentration  with  this — would  be 
achieved.  Doubtless  it  was  better  not  to  abandon 
the  whole  of  North  Italy  to  the  invader,  for  political 
if  not  for  military  reasons.  The  invader  should 
not  be  allowed  to  cross  the  river  without  fighting, 
at  least  to  prevent  murmurs  and  discouragement 
in  Rome  and  among  the  Emperor's  troops.  But 
if,  by  fighting,  the  foe  forced  the  passage  at 
Placentia,  even  so  the  tactical  would  not  be  a 
strategical  defeat  for  Otho. 

(2)  The   first  means  of  delaying    the   Vitellian  (2)  Means 

1  .  r  of  Delay. 

advance  was,  therefore,  the  occupation  in  force  of 
a  line  of  defence   on  the  river  Po  from   Placentia  j 

E 


/ 


50  THE  CAMPAIGN  OF  OTHO         ch.i 

to  Hostilia.  Twenty  thousand  or  twenty -five 
thousand  men  could  surely  maintain  their  position 
here  for  some  time,  helped,  as  they  would  be,  by 
the  river.  It  is  true  that  this  could  be  but  a  tem- 
porary measure  of  passive  defence.  "The  defence 
of  rivers  .  .  .  has  hardly  ever  been  successful  for 
any  length  of  time.  Neither  the  Danube  nor  the 
Rhine  has  stopped  armies."  A  river,  like  a  moun- 
tain range,  is  an  "  insurmountable  impediment  which 
is  invariably  surmounted."1  The  Po  could  not  be 
permanently  held,  any  more  than  was  the  Tugela, 
in  the  face  of  repeated  and  vigorous  attempts  to 
force  the  passage,  especially  when,  as  in  both  these 
cases,  thanks  to  inferior  numbers  or  irresolution, 
no  counter-stroke  over  the  river  could  be  dealt  the 
assailants  by  the  defending  army.  But  as  a  means 
of  delay  rather  than  as  a  permanent  obstacle  the 
river  was  of  the  greatest  value  to  the  Army  of 
Italy. 

The  second  means  of  delay  was  the  fleet.  The 
command  of  the  sea  was  absolutely  Otho's.  An 
invasion  of  North  Italy  from  Germany,  it  might 
seem,  affords  the  least  possible  chances  that  the 
command  of  the  sea  should  have  any  influence 
at  all  upon  the  conduct  of  operations.  No  more 
unpromising  field  for  the  application  of  the  pet 
modern  theory,  it  might  be  urged,  could  possibly 
be  found.  Yet  none  the  less,  as  in  the  days  of  the 
second  Punic  war,  although  for  different  reasons,  so 
in  the  civil  war  of  a.d.  69,  the  invader  of  Italy  had 

1   Von    der    Goltz,    The  Nation    in   Arms,    Eng.   Trans,    p.    261.       Cf. 
Hamley,  Operations  of  War,  part  v.  chap.  iii. 


sbc.  in  AND  THE  VITELLIANS  51 

cause  to  regret  the  fact  that  the  control  of  the  sea 
rested  with  the  defender. 

The   reason    for  this   in   a.d.    69   was   that   the 
flank   of  an  army  which   proposed  to  cross  one  of '; 
the  western  passes  over  the  Alps  was  vulnerable  ' 
from  the  sea.     If  Otho  could  spare  the  troops,  a 
force    could    speedily    be    conveyed    on    shipboard 
to   Frejus,   and   there   landed.      With   the   fleet   asl  Forum 
its    base   it    could    march    up    country  to   threaten 
the  right   flank  of  a  column  crossing  the  Alps  by^ 
the  Mont  Genevre  or   Mont   Cenis  Pass.     If  the 
enemy  turned    upon   it   with   superior   numbers,   it  I 
could  retreat  to  the  coast  as  securely,  for  example, 
as  the  British  army  of  the  Peninsula  in  1808- 1809 
fell  back  on  the  fleet  at  Corunna  when  pursued  by  ; 
the  thronging  battalions  of  the  French.     And  every 
soldier  thereby  detached  from  the  invading  army, 
every  hour's  delay  to  the  final  concentration  of  the 
Vitellians  in   North   Italy,  was  so  much  pure  gain 
to  the  Emperor. 

This,  indeed,  would  be  but  a  minor  operation* 
intended  to  cause  a  diversion,  and  by  no  means! 
the  chief  drama  to  be  played  in  the  theatre  of\ 
war.  But,  as  its  object  would  be  entirely  con- 
sistent with,  and  favourable  to,  the  development 
of  Otho's  main  strategical  plan  for  the  beginning 
of  the  campaign,  it  would  be  entirely  justifiable. 
The  expeditionary  force  to  be  sent  with  the  fleet 
must  not,  indeed,  be  so  large  that  the  main  army 
on  the  Po  would  be  too  weak,  owing  to  its 
absence,  to  fulfil  the  task  of  defence  assigned 
to  it.     Nor,   again,   must    it   be   so  small   that    its 


\ 


52  THE  CAMPAIGN  OF  OTHO 

intended  menace  could  be  contemptuously  neglected 
by  the  enemy.  Some  expeditionary  force  must  be 
sent,  if  the  fleet  were  to  be  of  any  service  at  all. 
Thus  when  Napoleon's  line  of  communications  with 
France  in  his  invasion  of  Italy  in  1796  ran  along 
the  coast  through  Savona,  the  British  fleet,  although 
it  "completely dominated  the  Mediterranean  littoral," 
was  quite  unable  to  threaten  these  communications, 
since  it  had  no  force  on  board  with  which  to  strike 
a  blow  at  them.  The  use  of  an  army  for  such 
operations,  conveyed  by  and  based  upon  a  fleet, 
however  inferior  in  numbers  this  army  may  be 
to  the  enemy,  is  a  vital  element  in  the  strategy 
of  the  command  of  the  sea,  although  this  principle 
seems  hard  to  realise  from  the  days  of  Pericles 
down  to  our  own  generation. 

If  then  Otho  could  spare  a  few  thousand  men 
from  the  Army  of  Italy  to  be  carried  on  ship  and 
disembarked  at  Frejus  or  some  other  port  on  the 
coast  of  Provence— the  "Province" — this  might 
be  a  second  useful  means  of  delaying  the  advance 
and  concentration  of  the  enemy  till  such  time  as  the 
Army  of  the  Danube  arrived  at  Mantua. 

Then  at  last  would  come  the  time  for  offence, 
and  Otho's  united  army  could  be  sent  against  the 
enemy  to  hurl  them  back  through  a  land  long  since 
exhausted  by  their  stay  in  it ;  back  against  the  grim 
barrier  of  the  mountains  which  cut  them  off  from 
safety — back  with  weakened  strength  and  diminished 
numbers,  to  perish,  starved  and  fighting,  penned 
up  against  the  Alpine  wall.  "  Happy  the  soldier 
to  whom  fate  assigns  the  part  of  assailant."     Or 


sec.  in  AND  THE  VITELLIANS  53 

perhaps  Otho  need  not  wait  for  the  arrival  of  the 
entire  Danube   army  when   once   these  were  hard 
at  hand.     "The  essential  in  war  is  not  the  massing^?', 
of  troops  but  their  co-operation."  *     "  Envelopment,  1 , 
not  mere  weight  of  numbers,  is  the  true  secret  of  l^A 
decisive    success."2      Some    more    daring    plan    of 
attack  might  suggest  itself  which  promised  speedier 
victory  than  the  frontal  attack  by  a  united  army. 
Could  not  the  stubborn  fighting,  the  many  weary 
miles  of  marching  which  lay  between  the  river  and 
the  mountains,  the  last  desperate  stand  of  despair- 
ing men, — could  not  all  this   be  avoided  by  some 
masterpiece  of  manoeuvre  and  surprise  ? 

But    all    such    plans    must    for    the    present    be! 
delayed    until    the    Danube    army    should    arrive.! 
Meanwhile  one  step  was  enough.     Strategy  cannot 
look  to  the  horizon  lest  she  stumble  in  the  ditch  at 
her  feet.     "  No  plan   of  operations    can   with   any 
safety  include  more  than  the  first  collision  with  the1 
enemy's-main  force."3     So  for  the  time  the  Army 
of  Italy   should   make   resolute   defence   along  thej 
line  of  the  Po,  and  the  command  of  the  sea  should' 
be  used  to  assist  it  to  delay  the  Vitellians'  advance. 

This  strategy  surely  promised  well.  It  had,<=-— 
however,  two  defects  in  chie£/ylt  failed  to  prevent 
ultimately  the  enemy's  concentration  in  the  plain  of 
the  Po,  though  delay  might  be  caused  by  the  fleet. 
Gsx  And  the  strategy  of  defence,  however  temporary, 
might  at  any  time  impair  the  confidence  and  morale 
of  the  men  of  the  army  on  the  river,  and  especially 

1  Von  der  Goltz,  op.  cit.  p.  304.  -  Henderson,  Science  of  War,  p.  415. 

3  Von  der  Goltz,  op.  cit.  p.   187. 


54  THE  CAMPAIGN  OF  OTHO 

so  at  a  time  of  civil  war,  when  all  the  troops 
were  excited  and  impatient.  For  the  game  of  war 
as  played  in  the  field  is  anything  but  the  War-game 
of  the  drill  hall.  Would  the  Guards,  the  flower 
of  the  Roman  army,  consent  to  stand  for  some 
weeks  on  the  defensive  against  a  hated  foe?  If 
they  obeyed  such  orders,  would  their  military  fire 
and  zeal  not  be  impaired?     Such  questions  had  to 

^be  considered  by  the  Emperor.  Yet  he  knew  that 
his  men  were  devoted  to  his  cause.  The  strategy 
of  defence  on  the  river  was  the  wisest  for  him,  and 
Otho  might  well  feel  that  he  could  rely  upon  his 
men  for  any  manoeuvre — even  that  most  dispiriting 
one   of  waiting  to  be  attacked.     Further  than  the 

I  Po  he  would  not  retreat.  Not  though  the  Apen- 
nines in  spring  are  deep  with  snow,  and  their 
mountain  tracks  hazardous  and  well-nigh  impass- 
able,1 would  he  fall  back  under  cover  of  their 
shelter,  and  seek  to  lure  the  foe  on  to  venture  into 
their  recesses  or  perhaps  be  ensnared  between  them 
and  the  sea.  Retreat  to  the  river  was  far  enough. 
Beyond  the  river  the  one  maxim  laid  down  by  our 
English  general  for  an  invaded  country  held  good 

I  for  Otho  and  his  men  :  "  No  foot  of  ground  ceded 
that  was  not  marked  with  the  blood  of  the  enemy."  " 
B.  Strategical  Opportunities  of  the  Vitellians. — 
The  Vitellians,  on  the  other  hand,  enjoyed  the 
advantage  of  being  the  attacking  party,  but  very 
few  advantages  besides.  The  courage  and  con- 
fidence  characteristic    of  good    troops    who    move 

,  '  In  April  1907  snow  lay  10  to  20  feet  deep  in  places  on  the  Abetone  Pass 
between  Modena  and  Fistoia.  -  Diary  of  Sir  John  Moore,  vol.  ii.  p.  75. 


sbc in  AND  THE  VITELLIANS  55 

to    the    attack,    and    apt    to    be    lacking    in    those 
kept    on    the    defence,    might    certainly    be    theirs. 
Yet  perhaps  this  would  hardly  do  more  for   them 
than   compensate    for    their   original    inferiority   as 
troops    of  the   line   to    the    Guards.      Clausewitz's 
familiar  assertion  that  the  defensive  form  of  warfare 
is  in  its  nature  stronger  than  the  offensive,  causes 
very  great   searchings   of  heart   to   the   strategists 
among  his  countrymen  to-day.      But  if  ever  a  strat- 
egical position  were  wanted  to  justify  the  assertion, 
that  of  the  spring  of  a.d.  69  might  seem  to  be  the  one 
desired.      For  then  the  Vitellian  chances  of  prosper* 
ous  attack  seemed  somewhat  meagre  compared  with 
the  Othonian  of  happy  defence.     The  most  obvious^ 
perhaps  the  only  possible,  strategy  for  the  invaders 
was  a  rapid  descent  over  the  mountains  to  the  plain 
of  the  Po,  and  a  frontal  assault  upon  the  position 
garrisoned  by  the  Army  of  Italy.     Time  was  of  the  1 
most    vital    importance    to    the    Vitellians.       They 
must  hasten   to  move  upon   Italy  in  time  to  anti-] 
cipate   a  possible   blocking   of  the   Alpine    passes.  | 
They  must  hasten  to  fall  upon  the  Army  of  Italy 
before  the  Army  of  the  Danube  had  time  to  come 
to  its  aid.    Only  if  they  could  crush  the  former  force  ! 
before  the  arrival  of  the  latter  in  strength  would  they 
have  the  undoubted  superiority  henceforward  in  the 
strategy  of  the  war,  should  the  war  continue.     The  ' 
movement  upon   Italy  in  two  columns  by  different  I 
passes  was  necessary.    The  column  which,  travelling  \ 
by  the  nearer  route,  first   arrived  in    Italy  must,  if 
strong  enough,  attack   the  enemy  at  once  ;    if  too 
weak,  or  beaten   in   its  onset,  it  must  wait  for  the 


56  THE  CAMPAIGN  OF  OTHO 

coming  of  the  second  column  to  reinforce  it.  And 
the  race  between  the  reinforcements  of  both  sides 
would  in  truth  be  an  anxious  one.  Speed,  concentra- 
tion, and  frontal  attack  seemed  the  sole  means  to 
the  Vitellians  of  achieving  success.  And  the 
strength  of  the  defenders'  position  combined  with 
the  means  open  to  them  of  delaying  the  assailants' 
approach  might  neutralise  the  advantage  of 
numerical  superiority  enjoyed  by  the  latter. 

An  alternative  strategy  to  this  of  concentration 
Vind  frontal  attack  might  be  considered.  The 
"  strategy  of  penetration  "  justly  wields  much 
I  fascination,  and  for  modern  war  has  all  the  support 
I  of  Napoleon's  favourite  practice  behind  it.  If  the 
Vitellians  could  thrust  boldly  between  the  two 
fractions  of  Otho's  gathering  army,  could  they  not 
defeat  them  in  detail  ?  The  Brenner  Pass  in  the 
north  offered  the  easiest  access  to  Italy  of  all  the 
Alpine  passes,  and  led  straight  down  to  the  very 
centre  of  the  hostile  position.  Could  the  Vitellian 
generals  use  the  advantage  of  superior  numbers 
which  they  enjoyed  at  the  outset,  and  drive  in  a 
great  wedge  of  their  own  men,  penetrating  the 
defenders'  lines  midway  between  Placentia  and 
Aquileia  ?  This  alternative  strategy  deserved 
consideration  by  the  Vitellians  at  the  beginning 
of  the  campaign.  But  the  reasons  which  caused 
Caecina  to  reject  it  were  adequate.1 

Such  were  the  strategical  opportunities  of  both 
sides  at  the  outset  of  the  struggle.  No  campaign 
ever  yet  followed  precisely  the  course  marked  out 

1  See  below,  §  4. 


sec.  iv  AND  THE  VITELLIANS  57 

for  it  by  the  strategist.  The  weavers  at  the  loom 
of  war  might  think  to  have  but  a  common  and 
familiar  pattern  for  their  work.  But  the  designer 
who  cuts  out  the  cards  for  them  may  have  indulged 
a  free  fancy  in  the  pattern  which  he  gives  them. 


§  4.    The  March  of  the  Vitellians 

Confident  in  his  greater  numbers,  Vitellius 
issued  orders  for  the  immediate  invasion  of  Italy. 
He  divided  his  forces  into  three  parts.  Two  of 
these  were  advance  columns  of  invasion  ;  the  third 
was  the  reserve,  to  follow  later  in  support. 

The  advance  columns  were  ordered  to  penetrate 
into  the  valley  of  the  Po  by  different  Alpine  passes.1 
The  first  of  these,  the  "  Cologne  Column "  from 
Lower  Germany,  under  Fabius  Valens,  was  to 
march  through  Gaul  and  cross  the  Alps  by  the 
Mont  Genevre   Pass  on  the  west  of  the  mountains.!  Aipes 

I  Cottianae. 

This  column  was  composed  of  the  Fifth  legion 
with  its  "  eagle,"  and  of  detachments  ("  vexilla  ")  of 
the' First,  Fifteenth,  and  Sixteenth  legions,  together 
with  auxiliary  squadrons  ("  alae  "),  and  cohorts  from 
11  Germany."  Its  strength  was  reckoned  as  forty 
thousand  troops.  The  second  of  the  advance 
columns,  the  "  Vindonissa  Column,"  from  Upper 
Germany,  was  commanded  by  Aulus  Cjaeclna 
Alienus.  It  consisted  of  the  Twenty-first  legion, 
detachments  of  the  Fourth  and  Twenty  -  second, 
and  German  auxiliaries,  and  amounted  to  thirty 
thousand    men.       This    column   was    bidden    enter 

1   For  the  reasons  dictating  this  see  above,  p.  39. 


58 


THE  CAMPAIGN  OF  OTHO 


Italy  on  the  north-west  by  the  Great  St.  Bernard 
Pass.  The  reserve,  consisting  of  such  other  forces 
as  could  be  raised  in  Germany  and  elsewhere,  was 
to  march  under  the  personal  command  of  Vitellius 
himself,  advancing  with  such  greater  deliberation 
as  befitted  their  general's  ripe  years,  great  import- 
ance, and  unwieldy  frame.1  And,  in  fact,  the 
campaign  was  decided  six  weeks  before  he  himself 
arrived  upon  the  scene  of  the  decisive  battle.2 

A.  The  March  of  Valens? — From  its  place  of 
concentration  on  the  Lower  Rhine,  in  the  neigh- 
bourhood of  Cologne,  Valens'  column  marched  up 
that  river  as  far  as  the  modern  Andernach,  a  little 
short  of  the  point  where,  at  Coblenz,  the  Moselle 
enters  the  Rhine.  It  was  probably  at  Andernach 
that  Valens  left  the  main  river  and  struck  over  the 
hills  through  the  territory  of  the  Treveri  to  the 
capital  of  that  tribe,  now  Treves,  in  the  valley  of 
the  Moselle.  Thence  Valens  led  his  troops  to 
Metz,  higher  up  the  valley,  and  from  Metz  to  Toul, 
the  capital  of  the  Leuci.  At  Toul  the  column  left 
the  Moselle,  and  crossed  rolling  country  to  the 
chief  city  of  the  tribe  of  the  Lingones,  now  Langres, 
hard   by   the   source  of  the    Marne.      Here  it  was 

1  Tac.  i.  6 1.  "Vitellius  .  .  .  tota  mole  belli  secuturus,"  an  exaggerated 
phrase.      For  Valens  and  Caecina  see  above,  pp.  n-12. 

3  The  "battle  of  Bedriacum,"  April  15.  Visit  of  Vitellius  to  the  battle- 
ground, May  24. 

3  Tac.  i.  62-66  ;  ii.  27-30.  The  Corpus  Inscriptionum  Latinarum 
("C.I.L.")  enables  us  to  identify  the  places  mentioned  in  Tacitus'  narrative 
with  their  modern  equivalents.  I  use  the  latter  in  the  text,  placing  the 
Latin  titles  in  the  margin.  Certain  picturesque  incidents  which  befel  on  the 
march,  but  did  not  affect  the  military  situation,  are  omitted. 

4  C.I. I.,  xiii.   I,  fasc.  2,  p.  662. 
s   C.I.L.  xiii.  2,  p.  702. 

6   C.I.L.  xiii.  2.   1,  p.  109. 


sec.  iv  AND  THE  VITELLIANS  59 

joined  by  eight  Batavian  cohorts.  These  were 
properly  associated  as  auxiliaries  with  the  Four- 
teenth legion.  This  legion  had  belonged  to  the 
Army  of  Britain,  but  had  recently  been  transferred 
to  Dalmatia.  The  eight  cohorts,  however,  had 
refused  on  the  march  to  follow  it  to  its  new 
province,  and  were  sulkily  making  their  way  back 
to  Britain  when,  at  Langres,  the  news  of  Vitellius' 
rising  reached  them.  Here  then  they  waited 
events,  until  Valens  on  his  arrival  added  them  to 
his  force.  They  proved  but  sorry  allies.  At  once 
they  quarrelled  violently  with  his  legionaries,  and,  in 
fact,  gave  the  general  endless  trouble — not  indeed 
because  they  had  the  least  sympathy  with  the 
Othonian  cause  (as  their  actions  subsequently 
showed),  but  through  their  native  ferocity  and 
impatience  of  discipline.1 

The  Aedui,  into  whose  territory  the  army  next 
entered,  hastened  to  buy  the  goodwill  of  the 
soldiers  by  satisfying  their  hunger,  as  they  con- 
ciliated Valens  by  satisfying  his  demands  for 
money  and  arms.  This  tribe  occupied  the  high 
land  between  the  rivers  Saone  and  Loire,  but  the 
army  passed  through  their  borders  only,  following 
the  road  which  runs  from  Langres  through  Dijon 
to  Chalon-sur-Saone,  and  did  not  visit  their  chief Cabil- 

lonura. 

city,  on  the  site  of  the  modern  Autun,  which  lay 
away  on  the  right  flank.  From  Chalon  they 
marched  down  the  valley  of  the  river  Saone  to  its 
meeting  with  the  Rhone  at  Lugdunum.      Here  the 

1  Tac.  i.  59,  64  ;  ii.  27,  66,  69.     They  subsequently  did  good  service  in 
Mritain  with  Agricola  :   Tac.  Agric.  36. 


60  THE  CAMPAIGN  OF  OTHO         c.,., 

townsfolk  welcomed  them  with  open  arms.  Their 
grudge  against  Galba  for  his  favour  shown  to 
their  bitter  rivals  of  Vienne  was  deep,  and  Otho 
had  done  nothing  to  appease  it.  The  legions  of 
Germany  were  their  old  comrades  in  arms  against 
the  rebel  Gauls,  and  the  citizens  dreamed  fondly 
that  the  day  of  their  vengeance  upon  their  enemies 
lower  down  the  river  had  dawned  at  last.1  And, 
indeed,  it  had  done  so,  had  not  the  men  of  Vienne 
won  safety  by  tears,  entreaties,  and  gifts  to  gain  the 
soldiers'  pity,  and  by  enormous  bribes,  it  was  said, 
to  the  general.  At  Lugdunum  Valens  strengthened 
his  force  by  the  addition  of  the  greater  part  of  the 
garrison  of  that  city,  the  First  legion  Italica  and 
the  ala  Tauriana,  and  left  only  the  Eighteenth 
cohort  to  guard  the  place.  Then  the  army  marched 
away  down-stream,  passed  peaceably  through  the 
streets  of  Vienne,  between  the  river  and  the  over- 
hanging heights,  and  followed  the  left  bank  of  the 
Rhone  through  the  Allobroges'  territory  southwards 
by  way  of  Valence  as  far  as  its  tributary  the  Drome. 
Thence  the  column  pursued  its  slow  and  terrifying 
way  up  this  stream,  extorting  money,  by  menaces 
and  fire,  from  the  little  places  through  which  it 
passed,  and,  when  there  was  no  money,  appeasing 
its  disappointment  by  gratifying  its  lust.  One 
hope  only  of  speedy  relief  from  the  army's  presence 
was  left  to  the  landowners  and  magistrates  on  the 
army's  route.  Valens'  cupidity  would  sell  even 
this  relief  for  a  price.  Such  profitable  trafficking 
had    long    been    known    to    Roman    generals,   and 

1   See  above,  p.  27. 


sec.  iv  AND  THE  VITELLIANS  61 

Valens,  too,  reaped  his  harvest  of  gain  as  he  led  his 
troops  on  their  infamous  march  to  Italy.  In  such 
wise  the  army  marched  by  the  modern  little  towns 
of  Die  and  Luc-en-Diois  on  the  Drome  to  Briancon,  Dea 

Vocoi\- 

and,  crossing  the  Alps  by  the  Mont  Genevre  Pass,  tiorum. 
came  down  the  Doria  Riparia  valley  by  Susa  to  the  Augjusti. 
Po  at  Augusta  Taurinorum  (Turin).  Brigant,a 

But  the  mountains  were  not  the  only  difficulty 
which  Valens  had  to  surmount.  At  some  point 
during  the  march  from  Vienne  to  the  Po,  and 
probably  before  the  army  had  crossed  the  Alps, 
he  had  become  conscious  that  the  activity  of  the  j 
enemy's  fleet  on  the  coast  at  Frejus  threatened  the 
security  of  his  right  flank.1  Anxious  to  strengthen,' 
the  Vitellian  forces  on  the  coast,  he  therefore 
detached  a  few  of  the  unruly  Batavian  cohorts 
from  his  main  army  and  sent  them  away  to  the 
rescue.  Also  they  were  turbulent  boasters,  and 
he  was  glad  of  the  chance  to  be  rid  of  them.  But 
strategical  needs  above  all  dictated  the  general's 
orders.  His  other  troops,  however,  discovered 
their  departure,  and  resented  it.  Ignorant  of  the 
imperative  strategical  necessity  of  guarding  their 
flank  and  their  sole  line  of  communications,  they 
thought  only  of  the  strength  of  their  own  force,  and 
clamoured  that  "  so  powerful  a  limb  should  not  be 
torn  from  the  body."  They  would,  they  declared, 
all  proceed  to  the  Province,  or  they  would  all 
march  to  Italy  together.2  When  Valens  attempted 
to    quell    the    growing    disturbance,    open    mutiny 

1  See  below,  §  5. 
a  For  the  strategical  aspect  of  the  question  see  above,  pp.  50-52. 


62  THE  CAMPAIGN  OF  OTHO         ch.i 

broke  out.  The  general  fled  in  terror,  and  while 
he,  clad  in  a  slave's  garb,  lurked  concealed  in  a 
subaltern's  tent,  the  happy  troops  ransacked  his 
quarters,  searching  vainly  for  the  gold  of  Vienne. 
When  the  excitement  died  down,  and  anger  gave 
place  to  repentance  and  shame,  Valens  pitifully 
presented  himself  again  to  his  men.  General  and 
soldiers  wept  together  in  the  joy  of  reconciliation, 
and  a  few  rioters  were  punished  by  reprimand. 
And  this  is  a  Roman  army !  Truly  may  the 
Roman  historian  remark  that  in  times  of  civil 
war  the  soldiers'  license  exceeds  that  of  the 
general.1 

So  Valens  led  his  repentant  troops  over  the 
Pavia.  Alps  to  the  Po,  and  down  this  river  to  Ticinum. 
\  Here  at  the  beginning  of  April  his  army  was 
engaged  in  fortifying  a  camp  when  the  news 
reached  them  that  their  comrades  of  the  Second 
Army  were  in  a  sore  plight  lower  down  the  river, 
at  Cremona.  Already  Caecina  and  his  Vindonissa 
column  had  met  the  enemy  and  had  been  defeated. 
The  news  called  for  instant  action.  Valens  broke 
up  his  camp  and  marched  at  full  speed  to  join  his 
colleague  at  Cremona.  He  arrived  here  in  the 
second  week  of  the  month. 

B.  The  March  of  Caecina? — The  headquarters 
of  Caecina  and  the  Second  Army  were  at  Vindo- 
nissa. This  camp  was  situated  on  the  tongue  of 
land  which  lies  between  the  rivers  Reuss  and  Aare. 
A  few  miles  to  the  north,  the  Aare  flows  into  the 
Rhine    between    Schaffhausen    to    the     east    and 

1  Tac.  ii.  29.  2  Tac.  i.  67-70. 


sec.  iv  AND  THE  VITELLIANS  63 

Sackingen  to  the  west.  In  the  neighbourhood  of 
the  camp  and  to  south  and  south-west  of  it  lay  the 
tribe  of  Helvetii,  once  the  early  and  stalwart  foes  of 
Julius  Caesar,  but  now  a  civilised  folk,  who  supplied 
auxiliaries  to  the  Roman  army,  and  had  gained  in 
culture  what  they  had  lost  in  pluck.  Opportunities 
for  plunder  afforded  by  civil  war  were  even  more 
welcome  to  Caecina  than  they  were  to  the  more 
indolent  Valens.  The  tribe  and  its  belongings 
offered  an  easy  prey  to  his  disciplined  troops,  and  a 
pretext  for  attack  upon  them  was  the  more  easily 
found  as  they  were  ignorant  of  Galba's  death  and 
refused  to  accept  Vitellius.  It  was  Caecina's  clear 
duty  to  his  Emperor  to  press  on  into  Italy  with  all 
possible  speed.1  Instead  of  this  he  let  loose  his 
army  upon  the  unlucky  natives,  and  sent  bidding 
his  friends,  the  procurator  of  Raetia  and  his 
auxiliary  troops,  to  harry  them  from  east  and  south 
while  he  himself  descended  upon  them  from  the 
north.  The  task  was  as  easy  as  it  was  profitable, 
and  the  miseries  of  the  tribe  were  great.  Their 
country  was  laid  waste  far  and  wide  ;  their  resist- 
ance was  spiritless  and  availed  them  nothing.  The 
hot  medicinal  springs  of  Baden  on  the  river  Limmat, 
some  sixteen  miles  to  the  north-west  of  the  town 
and  lake  of  Zurich,  from  which  lake  the  river  flows, 
were  already  known  at  this  time,  and  the  place, 
under  the  title  of  Aquae  Helvetiorum,  had  become 
a  popular  watering-place,  frequented  by  the  tribe.2 


1  See  the  strategical  reasons  for  this  above,  p.  55. 

-  "  Locus  .   .   .   amoeno  salubrium  aquarum  usu   frequens.'      Cf.    C.I.L. 
iii.  6017  ;  xiii.  5233  ;  xiii.  2,  p.  42. 


64  THE  CAMPAIGN  OF  OTHO         ch.i 

Caecina  sacked  and  destroyed  it.  The  Raetian 
auxiliaries  swept  down  from  Lake  Constance  upon 
the  victims.  The  refugees  from  the  general 
slaughter  flung  away  their  arms  and  fled  to  the 
Mons      hills    of   the    Botzbem    (which    form    the    extreme 

Vocetius.  T  i- 

north-east  end  of  the  Jura  range,  and  lie  a  few 
miles  west  of  Windisch).  They  were  driven  out 
by  a  cohort  of  Thracian  auxiliaries,  and  thousands 
more  were  butchered  in  the  pursuit.  Caecina  next 
marched  upon  the  capital  city  of  the  tribe,  Aven- 
ticum.  This,  the  modern  Avenches,1  lies  hard  by 
the  lake  of  Morat,  five  miles  to  the  east  of  the  lake 
of  Neuchatel.  Aventicum  surrendered.  The 
general  allowed  the  townsfolk  to  appeal  for  mercy 
to  Vitellius.  They  won  pardon  with  difficulty, 
thanks  to  a  sudden  change  of  feeling  towards  them 
on  part  of  Vitellius'  army,  which,  of  course,  took 
upon  itself  to  decide  the  matter.  If  the  Roman 
legionary  of  the  time  could  once  be  induced  to 
weep,  then  there  was  hope  of  mercy  at  his  hands. 
Streams  of  tears  saved  the  remnant  of  the  Helvetii 
from  the  extremes  of  Imperial  vengeance.  Later, 
Vespasian  the  Emperor  bestowed  the  Jus  Latinum 
on  the  place  in  recompense  for  its  sufferings  at  the 
hands  of  his  enemies,  and  under  the  proud  title  of 
Colonia  Pia  Flavia  Constans  Emerita  Helvetiorum 
the  town  might  perhaps  forget  its  earlier  woes.2 

While   Caecina  stayed   dallying    at   Aventicum, 
news    reached    him    from    Italy    which    called    for 

1  German,  Wiflisberg. 

2  C.I.L.  xiii.  5089,   5093,   5063,  and   Zangemeister,  ap.  C.I.L.  xiii.   2, 
PP-  5.  6»  18. 


sec.  iv  AND  THE  VITELLIANS  65 

prompt  action  on  his  part.  At  the  outbreak  of  the 
war,  as  has  been  said,  Otho's  only  troops  actually 
in  the  valley  of  the  Po  were  one  squadron  of 
auxiliary  horse,  the  ala  Siliana.  But  this  same 
ala  had  chanced  to  be  quartered  in  Africa  eight  I 
years  before,  at  the  time  when  Vitellius  adminis- 
tered the  affairs  of  that  province.  Its  officers 
therefore  succeeded  in  persuading  the  men  to 
throw  in  their  lot  with  their  old  governor,  and  the 
more  easily  as  the  numbers  and  valour  of  the  Army 
of  Germany  inspired  the  regiment  with  much  awe. 
It  therefore  at  this  point  revolted  from  Otho,  and 
brought  over  with  it  to  the  Vitellian  side  the  four 
strongest  towns1  of  the  Transpadane  district  in 
North-west  Italy,  which  lay  north  of  the  Po.  If 
these  four  towns,  Mediolanum  (Milan),  Novaria 
(Novara),  Vercellae  (Vercelli),  and  Eporedia(Ivrea), 
once  fell  into  the  hands  of  the  enemy,  the  Othonians 
had  no  choice  but  to  limit  their  attempts  at  defence 
to  the  line  of  the  river  itself.  This  was  the  news/ 
which  reached  Caecina  at  Aventicum.  The  whole 
of  North-west  Italy,  with  its  garrison  towns  com- 
manding the  issues  of  the  Great  St.  Bernarcl, 
Pass,  was  offered  to  him.  By  this  pass  he  haaj] 
been  ordered  to  enter  Italy.  But  the  one  little 
regiment  of  horse  could  not  hold  the  tract  for  long 
in  face  of  the  enemy  moving  up  from  the  south. 
Caecina  therefore,  without  more  delay,  hurried  strong 
reinforcements  over  the  Alps  to  its  help.  These 
consisted  of  Gallic,  Lusitanian,  and  British  auxiliary 
cohorts,  detachments  ("vexilla")  of  German  troops, 

1   "  Firmissima  municipia." 


66  THE  CAMPAIGN  OF  OTHO 


CH.  1 


and  another  squadron  of  cavalry,  the  ala  Petrlana.1 
I  For  the  moment,  these  troops  on  arrival  might  well 
hold  in  check  any  attempt  by  the  enemy  to  cross 
the  Po  in  force.  And  then  Caecina  had  to  set 
himself  anxiously  to  consider  what  his  main  plan  of 
campaign  should  now  be. 

To  carry  out  Vitellius'  orders,  it  was  right  for 
him  to  follow  with  the  rest  of  his  army  over  the 
Great  St.  Bernard,  in  the  wake  of  his  advance 
guard,  as  speedily  as  possible.  But  the  idea  of  an 
alternative  and  more  daring  strategy  was  first  at 
least  to  be  considered. 

This  was  the  "strategy  of  penetration."  If 
Caecina  could  cross  the  Alps  farther  to  the  east  he 
could  perhaps  thrust  his  force  between  the  two  parts 
of  the  Othonian  defending  army  upon  the  Po.  For 
this  purpose  the  Brenner  Pass,  by  which  the  great 
road  from  Innsbruck  crosses  the  Alps  to  Verona, 
was  most  suitable.  It  was  by  far  the  easiest,  the 
lowest,  and  the  best  known  of  all  the  Alpine  passes. 
If  Caecina  came  down  upon  Verona  by  this  route  he 
might  hope  to  cut  completely  the  communications 
of  the  Danube  army  at  Aquileia  with  the  Army  of 
Italy  upon  the  river,  by  penetrating  between  them. 
If  Valens,  then,  approached  from  the  west,  and  if 
the  mobilisation  of  the  Danube  legions  at  Aquileia 
proceeded  slowly,  surely  the  Army  of  Italy  would 
be  caught  in  the  middle  and  annihilated  with  ease. 

The  plan  was  a  tempting  one,  but  very  daring. 
There  was  always  the  chance  that  the  penetrating 
army   would   find   the  tables   turned  and   itself  be 

1   Cf.  Tac.  ii.  17. 


sec.  iv  AND  THE  VITELLIANS  67 

surrounded.  If  the  movements  of  the  Danube 
army  should  be  quicker  than  those  of  Valens'1 
column,  Caecina  at  Verona  or  Mantua  might  find 
himself  very  uncomfortably  placed,  especially  if  he 
found  any  large  part  of  the  Army  of  Italy  blocking 
the  way  on  the  Adige  at  Verona.  Valens,  too,  had 
already  started  on  his  march  through  Gaul.  Even 
if  Caecina  sent  him  an  urgent  message  to  hurry,  he 
would  not  improbably  disregard  it  entirely. 

But  geographical  obstacles  proved  finally  fatal 
to  the  plan.  The  road  from  Innsbruck  to  Verona 
over  the  Brenner  was  indeed  an  easy  one.  The 
top  of  the  pass  is  under  4500  feet  in  height.  But 
it  was  by  no  means  such  an  easy  matter  to  get 
trom  the  north  of  Switzerland,  where  Caecina  then 
was,  to  the  valley  of  the  Inn  over  the  "  Raetian 
heights."  For  this  purpose  he  would  have  to 
march  by  Lake  Constance  to  the  valley  of  the  111 
river  at  Feldkirch  and  then  cross  the  mountains 
to  Innsbruck  by  the  Arlberg  Pass.  This  pass  is 
nearly  6000  feet  in  height.  In  April  1906  it  was 
deep  in  snow  for  several  miles  on  either  side  of  the 
top,  and  could  be  crossed  only  on  snow-shoes.  In 
February  a.d.  69  it  was  not  likely  to  have  been  less 
difficult  of  passage  by  an  army.]     That  Caecina  did 

1  That  "Raeticis  jugis  "  means  the  Arlberg  Pass  is  surely  beyond  doubt. 
The  suggested  alternative  is  that  of  the  Stelvio  (or  Umbrail)  Pass,  which 
leads  from  Nauders  in  the  Engadine  to  Bormio  and  Tirano,  and  so  to  Como 
or  to  the  Val  Camonica  and  Iseo.  To  reach  this  pass  on  the  Tyrolese  side 
Caecina's  route  would  have  lain  by  Ragatz,  and  the  Fliiela  and  Ofen  Passes 
first.  The  heights  of  the  passes  are  as  follows  :  Arlberg,  about  5900  feet ; 
Stelvio,  9000 ;  Fliiela,  7835 ;  Ofen,  7070.  It  seems  at  least  uncertain 
whether  tracks  existed  over  these  last  three  passes  in  Roman  days.  How- 
ever this  may  be,  Caecina  could  hardly  have  contemplated  coming  this  way  in 
February.  The  difficulties  of  the  Arlberg  would  be  as  nothing  to  those  of 
the  Stelvio  ;  the  latter  pass  would  have  been  no  service  to  him  against  the 


68  THE  CAMPAIGN  OF  OTHO         ch.i 

consider  the  plan  is  shown  by  the  action  of  the 
procurator  of  Noricum,  Petronius  Urbicus,  who  was 
ranged  upon  the  enemy's  side.  Some  fifty  miles 
east  of  Innsbruck  down  the  Inn  river  lay  the 
frontier  of  his  province.  The  procurator,  who,  like 
his  fellow-procurator  of  Raetia,  had  only  auxiliary 
troops  at  his  disposal,  promptly  destroyed  all  the 
bridges  over  the  river,1  trusting  to  that  fiercely- 
rushing  stream  to  save  him  from  attack  if  Caecina 
came  that  way.  As  a  matter  of  fact,  if  Caecina  had 
crossed  the  Arlberg  it  was  not  Noricum,  with  its 
petty  force  and  unimportant  governor,  which  would 
have  been  in  his  thoughts,  but  rather  the  Brenner 
Pass  and  Italy  ;  and  the  procurator  had  in  any  case 
alarmed  himself  needlessly.  Had  Caecina  come 
that  way,  Petronius  Urbicus  should  indeed  have 
made  vigorous  demonstrations  on  the  flank  of  the 
enemy's  column  at  Innsbruck,  with  intent  to  delay 
its  passage  over  the  Brenner.  Instead  of  this  he 
was  preparing  to  stand  stoutly  on  the  defensive. 
It  is  a  well-known  fact  of  military  history  that  "all 
local  commanders  are  firmly  convinced  always  that 
it  is  upon  them  that  the  brunt  of  the  fighting  is 
destined  to  fall."2     Petronius  Urbicus  saw  that  the 

Othonian  position.  Even  had  lie  diverged  east  from  Tirano  lie  would  have 
come  down  on  Iseo,  Brescia,  and  Cremona.  These  places  were  much 
more  easily  reached  from  Neuchatel  by  the  Great  St.  Bernard.  And  the 
Stelvio  could  not  possibly  have  excited  the  attention  of  the  procurator  of 
Noricum.  Hence  my  choice  of  the  Ailberg-Brenner  combination.  I  owe 
thanks  to  Mr.  K.  L.  Poole  for  calling  my  attention  to  the  proposed  alternative. 

Of  course  the  Great  St.  Bernard — height  8 no  feet — was  hard  enough 
to  cross,  as  Napoleon  found  it  in  May  1S00.  But  the  Stelvio  would 
be  much  worse.      And  the  Roman  column  would  not  march  with  artillery. 

1  Tacitus  says  "  interruptis  Jhitninum  pontibus."  The  plural  is  un- 
meaning.     Only  the  Inn  is  concerned. 

-  Sir  Ian  Hamilton,  A  Staff-Officers  Strap-Book,  vol.  i.  p.  94. 


sec.  iv  AND  THE  VITELLIANS  69 

pass  over  the  "  Raetian  heights "  led  the  enemy 
to  the  Inn,  and  concluded  that  Caecina,  if  he  came 
that  way,  meant  to  attack  him.  For  Petronius,  in  an 
anxious  position  and  with  a  small  handful  of  troops, 
there  was  much  excuse.  But  there  is  little  excuse 
for  the  strategical  blindness  of  the  Roman  historian 
Tacitus,  who  complacently  reproduces  the  tale  that 
Caecina  hesitated  whether  to  attack  Noricum  or 
Italy,  and  finally  preferred  Italy  as  being  more 
important.  Such  a  question  could  need  no  de- 
liberation at  all.  Italy  was  from  first  to  last' 
Caecina's  objective.  The  very  important  problem 
for  his  decision  was  that  of  the  pass  by  which 
he  should  deliver  his  attack  on  Italy.  Petronius 
Urbicus  might  easily  fail  to  recognise  this.  Tacitus 
has  no  right  to  be  so  blind.1 

Caecina  therefore  gave  up  the  plan  of  strategical 
penetration  of  the  enemy's  position.  Apart  from 
the  military  risks  which  it  involved,  the  delay  which 
the  snows  of  winter  would  have  caused  compelled  \ 
him  to  abandon  it.  Probably  it  was  with  reluct- 
ance that  he  rejected  it  ;  for  a  few  weeks  later  he 
strove,  on  the  Po,  to  pursue  this  same  device  of 
piercing  the  enemy's  lines,  though  on  the  later 
occasion  his  attempt  was  one  of  tactical  rather  than 
strategical  penetration.2 

He  therefore  now  resumed  his  march  for  Italyv 
by  the  direct  route,  carrying  out  the  original  orders 
which  Vitellius  had  given  him.      From  Aventicum 
he  followed  in  the  steps  of  his  advance  guard  to 

1  See  below,  Note  E,  "Tacitus  as  Military  Historian." 
-  See  below,  p.  94. 


jo  THE  CAMPAIGN  OF  OTHO         ch.i 

the  head  of  Lake  Geneva,  and  up  the  Rhone  valley 
to  Martigny.  Thence  by  the  Great  St.  Bernard 
Pass  he  crossed  the  "  Pennine  Alps  "  to  Aosta,  and 
proceeded  down  the  valley  of  the  Dora  Baltea  to 
Eporedia.  By  the  second  week  in  March,  a  month 
before  Valens'  column  entered  Italy,  Caecina  had 
/crossed  the  Alps,  and  was  in  touch  with  the 
/  enemy's  outposts  in  the  valley  of  the  Po. 

§  5.    The  Othonian  Measures  of  Defence 

Meanwhile  the  Emperor  Otho  was  busy  arrang- 
ing for  the  defence  of  Italy  against  the  invader. 
)His  preliminary  measures  of  defence  were  two  in 
number.  In  the  first  place,  he  sent  part  of  his  force 
of  Guards  on  board  the  fleet  to  Narbonese  Gaul,  to 
threaten  the  flank  of  Valens'  invading  column,  and  at 
least  delay,  if  they  could  not  interrupt,  its  march. 
■  Secondly,  a  strong  advance  guard,  under  the  two 
generals  Annius  Gallus  and  Vestricius  Spurinna, 
was  ordered  to  leave  Rome  at  once  for  the  north. 
This  force  consisted  of  five  Praetorian  cohorts,  the 
First  Adjutrix  legion,  two  thousand  gladiators,  and 
some  detachments  of  cavalry.  It  was  their  duty  to 
secure  the  line  of  the  Po.  As  soon  as  the  treachery 
of  the  ala  Siliana  was  known,  it  was  clearly  im- 
possible to  hope  to  hold  the  country  west  and  north 
■  \  of  Placentia,  or  to  attempt  to  prevent  the  muster  of 
I  \  the  Vitellians  north  of  the  river.  Spurinna  there- 
fore, with  part  of  the  troops,  threw  himself  into 
Placentia  to  block  the  passage  of  the  river  at  that 
fortress,  and  therewith  the  great  road  to  the  south. 


secv  AND  THE  VITELLIANS  71 

Gallus,  for  his  part,  moved  with  the  rest  of  the 
advance  guard  to  secure  the  communications  with 
the  Danube  army  then  collecting  at  Aquileia. 
This  he  would  best  do  by  crossing  the  river  at 
Hostilia  and  moving  forward  to  a  position  in  the 
neighbourhood  of  Mantua,  which  covered  that  cross- 
ing and  the  road  to  the  east.1 

To  Otho  himself  at  Rome  there  was  still  left  a 
considerable  number  of  excellent  troops  in  fighting 
trim  and  eager  for  the  fray.     He  left  the  city,  at  , 
the  head  of  these,  for  the  theatre  of  war  in  North  j 
Italy  on   March    14.      He   was  accompanied   by  a 
large  general   staff,   including   the   most   renowned^ 
general  of  the  day,   Suetonius   Paulinus.      In   his 
train  came  many  unhappy  senators,  reluctantly  and 
of  gentle  compulsion.     These  were  left  behind  at 
Mutina  when  the  army  moved  forward  to  the  front. 
Their  panics  and  distresses,  their  discomforts  and 
perils,  their  shifts  and  evasions,  form  a  somewhat 
humorous  relief  to  what  henceforward  is  but  a  grim 
and  sombre  story.2 

These  military  dispositions  were  sound.      The  | 
best    defensive    position     was    thereby    occupied.  • 
Means  for  delaying  the  muster  of  the  enemy  were 
employed.     And,  above  all,  the  communications  of 
the   army  on    the    Po    with    that   of    the    Danube; 
provinces   were    secured.       The    superstitious    and 
silly    folk   in    the    capital     criticised    gloomily    the 

1  Tacitus  does  not  tell  us  anything  of  Gallus'  movements  when  he 
separates  from  Spurinna.  But  as  when  Gallus  next  appears  he  is  marching 
from  the  east  on  Placentia,  the  statement  in  the  text  is  the  probable  infer- 
ence, especially  as  some  of  the  Danube  army  is  then  with  him.  CI.  Tac.  it. 
18,  23. 

2  Tac.  i.  87  ;  ii.  II,  52. 


72  THE  CAMPAIGN  OF  OTHO         en.  i 

Emperor's  speedy  departure  from  Rome.  It  was 
the  month  of  the  dancings  of  the  sacred  priesthood, 
the  Salii,  and  the  Ancilia,  the  twelve  famous  shields, 
could  not  be  restored  to  their  resting-place  in  Mars' 
Sacrarium  until  the  month  was  ended.  Surely, 
they  urged,  Otho  ought  to  wait  in  Rome  for  the 
sixteen  days,  until  such  time  as  the  Salii  had 
finished  their  dancings  and  the  sacred  shields  were 
restored  to  their  shrine.  The  very  gods  seemed 
to  desire  this.  Fearsome  omens  were  recorded, 
and  Mars'  plain  was  flooded  deep  by  one  of 
those  terrible  inundations  which  the  Tiber  from 
time  to  time  inflicts  on  the  city  up  to  this  day.1  No 
muster  or  review  of  the  Imperial  Army  could  be 
held  on  the  Campus  Martius,  and  the  Via  Flaminia, 
the  great  high-road  to  the  north,  which,  after  leaving 
the  city  gate,  ran  parallel  to  and  not  far  from  the 
river,  was  blocked  for  some  distance.  But  such 
scruples  and  hindrances  could  not  delay  Otho.  The 
Emperor  was  a  soldier,  and  was  well  aware  of  the 
supreme  importance  of  time  in  operations  of  war. 
.  Caecina,  he  knew,  had  already  crossed  the  Alps, 
u Delay  had  proved  fatal  to  Nero;  it  should  not 
/  imperil  him.  By  the  earliest  date  which  his  final 
preparations  allowed,  the  14th  of  March,  he  led 
his  troops  out  from  Rome.  The  jumping  priests 
must  jump  for  a  fortnight  without  him.  They  will 
jump  all  the  more  merrily  next  year  if  he  comes 
back   in   triumph.      "  Deorum   injurias   dis   curae." 

1  The  Campus  Martius,  lying  on  the  river  bank,  is  peculiarly  liable  to 
these  floods.  The  tablets  recording  the  height  of  the  river  at  such  times, 
e.g.  in  the  Piazza  of  the  Pantheon,  are  familiar  to  visitors  to  Rome.  In 
some  cases  they  are  as  high  or  higher  than  the  reader's  head. 


sec.v  AND  THE  VITELLIANS  73 

Otho  will  trust  rather  to  the  devotion  of  his  ! 
soldiers  than  to  divine  patronage.  Armed  with 
cuirass  of  iron  the  Emperor  marched  on  foot  in 
front  of  the  standards,  rough  in  dress  and  look,  and  j 
careless  of  comfort.  This  was  not  the  Otho  whom 
men  had  known  in  former  days.  It  was  a  soldier 
Emperor  marching  to  defend  his  empery.1 

Yet,  excellent  though  these  measures  of  defence 
were,  the  course  of  events,  as  they  developed,  dis- 
covered three  flaws  in  them.  The  force  sent  on 
shipboard  acted  too  feebly  ;  the  Army  of  the  Danube 
gathered  together  too  slowly;  and  Cremona  fell  intov 
the  hands  of  the  foe. 

(1)   The  Action  of  the  Fleet} — Otho  entrusted 
the  command  of  the  naval  expedition  to  three  men. 
Their    incapacity   was    equalled    only  by  their    in- 
significance.    Otho  lacked  entirely  that  admirable) 
faculty  of  being  able  to  choose  men,  which  his  pre-j 

^decessor  Nero  had  possessed  in  marked  measureJ 
^\6lt  was  this  lack  which  ruined  him.     Of  the  three 

•commanders  of  the  fleet,  one  was  put  in  irons  by  his 
troops  ;  the  second  was  unable  to  control  them  ;  the 
third  to  control  himself.  There  was  no  discipline, 
and  the  fleet  sailed  up  the  north-west  coast  of  Italy 
like  a  pirate  fleet,  ravaging  and  murdering.  One 
of  the  victims  was  the  mother  of  Agricola  ;  and  the 
Roman  historian,  who  found  in  this  general  the  hero 
of  his  youth,  writes  bitterly  of  the  ferocity  shown 
by  the  Othonian  troops  who  were  sent  on  this 
expedition  : — 

1  C(.  Tac.  i.  S6-89;  ii.  II. 

2  Tac.  ii.  12-16  ;  Agricola,  c.  7. 


74  THE  CAMPAIGN  OF  OTHO         ch.i 

It  seemed  they  entred  not  into  Italie,  their  owne  native 
countrey  and  soile  :  as  if  it  had  bene  forraine  coasts  and  cities 
of  enemies,  they  burned,  wasted,  and  spoiled,  with  so  much 
the  more  outrage  and  harme  because  no  such  invasion  was 
feared,  and  therefore  nothing  provided  against  it :  the  fields  lay- 
full  of  commodities  ;  the  houses  wide  open  ;  the  masters  meeting 
them  with  their  wives  and  children,  through  the  security  of 
peace,  were  overtaken  with  the  misery  of  warre.1 

The  hill  men  who  came  down  to  the  rescue  were 
easily  routed  and  dispersed  by  the  regular  troops, 
who  proceeded  in  their  wrath  to  sack  the  town  of 
Intimilium,  now  Ventimiglia,  the  Riviera  town  on 
the  frontier  between  France  and  Italy.2 

The  pezants  were  beggerly,  their  armour  not  worth  taking 
up ;  and  beside,  being  swift  of  foote  and  skilfull  in  the  countrey, 
they  could  not  be  taken  :  but  the  sacke  of  the  poore  innocent 
towne  paied  the  reckening,  and  contented  the  covetous  soldier. 
The  odiousnesse  of  which  fact  was  greatly  increased  by  a  notable 
example  which  happened  there  of  a  Ligurian  woman :  who 
having  hid  her  sonne,  the  soldiers  supposing  she  had  hid  her 
money  withall,  and  thereupon  by  torture  examining  hir,  where 
she  had  hid  him,  shewing  hir  belly  answered  that  there  he  was 
hid :  neyther  could  she  by  any  manner  of  torment  afterwarde, 
or  death  at  the  length,  be  induced  to  change  that  worthy 
answere. 

While  the  fleet  was  thus  pleasantly  occupied,  a 
messenger  rode  off  in  hot  haste  to  Valens  imploring 
help.  That  general  saw  at  once  the  danger  which, 
if  the    plunderers    were    not    checked,     threatened 

1  This  and  the  following  quotation  are  from  Sir  II.  Savile's  translation  of 
the  Histories,  A.  D.  1 591.  This  cannot  always  be  trusted  for  accuracy,  but  it 
deserves  to  be  remembered,  and  that  not  only  by  Mertonians,  if  for  no  other 
reason  yet  at  least  for  its  splendid  address  "To  the  Reader." 

2  There  is  no  doubt  that  Intimilium,  not  Albintimilium,  is  the  right 
form. 


secv  AND  THE  VITELLIANS  75 

not  only  the  coast  of  the  Maritime  Alps  and  the 
Province,  but  also  his  own  march  to   Italy.       He 
therefore   sent  at  once   a   strong   force,   mainly   of 
cavalry,  including  the  ala  Treverorum.     This  last 
squadron  was  under  command  of  a  man   soon   to 
become  only  too  well  known  to  the  Romans,  but  then 
merely  a  subordinate  officer  of  auxiliaries  in  the  Army 
of  Germany.      But  this  man,  Julius  Classicus,  was  a 
great  noble  in  his  clan,  and  one  of  the  leaders  in  the 
near  future  of  the  rebellion  on  the  Rhine.1     Of  this 
force,  part  was  ordered  by  Valens  to  strengthen  the 
garrison  of  Forum  Julii  (Frejus)  against  the  risk  of  I 
attack.     The  rest  joined  the  local  troops  collected  1 
for  purposes  of  defence,  and  marched  against  the  ' 
enemy.     These  they  seemed  to  have  found  in  the  v 
neighbourhood.     The  brief  campaign  which  followed 
was,   though  indecisive   in   results,  not  without  its; 
interest.     It  served  to  illustrate  the  superiority  of 
the  Guards  as  soldiers  over  the  Vitellian  auxiliary  r 
troops,   and   supplied  another  example  of  the  use 
which  may  be  made  of  a    fleet    during    a   tactical 
engagement  upon  the  coast,  whereby  men  can  be  . 
landed  from  it  ashore  on  the  rear  of  the  enemy's  1 
battle  line,  thus  recalling  the  device  employed  by  J 
Pompey   in   old  days  at   Dyrrhachium.      The   tide 
of  success  in  the  fighting  ebbed  and  flowed,  but  on 
the  whole  the  Othonians  had   the  better  of  it,  and 
Valens  had  to  send  off  more  reinforcements  to  the 
scene  of  war.J     But  at  the  end  both  sides  withdrew 
their  forces  at  such  a  distance  from  one  another  that   1 
neither  the  cavalry  of  the  one  side  nor  the  fleet  ot 

1   See  below,  Chap.  III.  -  See  above,  p.  61. 


76  THE  CAMPAIGN  OF  OTHO         ch.i 

i  the  other   could    cause    any  sudden   alarm.      The 
Antipoiis.  1  Vitellians  retired  to  the  low  promontory  of  Antibes  ; 

,  the    Othonians    to   Albingaunum,  now    the    many- 

I  towered  little  city  of  Albenga,  which  lies  in  the 
small  swampy  plain  of  the  Centa,  sixty-five  miles 
east  of  Antibes.  At  Albenga  there  is  one  of  the 
few  pieces  of  open  shore  of  any  extent  along  the 
whole  rocky  coast  of  the  Riviera,  from  Savona  to 
Cannes,  and  the  ships  could  here  be  beached.  But 
it  was  unnecessary  to  retire  so  far  eastwards  for  the 
purpose,  as  the  small  harbour  of  Porto  Maurizio 
would  have  given  excellent  shelter  to  the  fleet,  and 
have   been  more  in   touch  with  the  enemy.     The 

,  fact  was  that  both  sides  had  by  tacit  consent  ceased 
from  all   warlike  activity,   and   the  short   maritime 

■  campaign  was  ended.  Sardinia  and  Corsica  were, 
it  is  true,  secured  for  Otho,  but  peace  henceforward 

\  reigned  on  the  coast  of  Provence. 

If  these  tentative  and  half-hearted  operations  of 
the  fleet  and  the  force  which  it  carried  are  to  be 
viewed  merely  as  an  isolated  episode  in  the  history 
of  the  war  (and  it  is  in  this  light  that  the  Roman 

,  historian  Tacitus  regards  them),  Otho  is  convicted 
of  a   strategical   blunder  in   commanding   them  to 

'.  take  place.  For  thereby  he  weakened  his  numbers, 
already  inferior  to  those  of  the  enemy,  by  detaching 

1a  force  for  the  purpose  of  a  minor  operation  of  war, 
when  his  whole  energies  should  have  been  concen- 
trated on  the  main  issue,  the  defence  of  North  Italy. 
Otho's  order  would,  in  this  case,  be  but  an  example 
the  more  of  that  familiar  failing  in  generalship,  for 
'  which  the  dispassionate  German  critic  blames  our 


sec.v  AND  THE  VITELLIANS  y7 

own  commanders  in  the  recent  Boer  war.1  To 
plunder  a  town  or  two  on  the  Riviera,  to  secure 
Corsica  and  Sardinia,  to  worry  the  local  forces  of 
the  enemy, — these  were  petty  successes  which 
counted  for  nothing  in  the  general  course  of  the 
war.2  Whereas  the  loss  of  strength  which  these 
successes  inflicted  upon  Otho's  main  army  counted  / 
for  a  good  deal. 

But  this  view  of  the  matter  is  short-sighted,  and 
based  rather  on  the  actual  results  of  the  campaign  / 
on  the  coast  than  on  the  intentions  of  the  Emperor/ 
when  he  sent  the  fleet.     Otho  did  not  intend  to  use 
the   command   of  the   sea   merely    to    secure    such- 
secondary  objects.     The  fleet's  activity  was  meant  to 
harass  Valens,  to  weaken  his  army,  to  delay  its  march.-' 
The  operations  of  the  fleet  were  therefore  intimately 
connected  with  the  concentration  of  the  Army  of  the 
Danube   at   Aquileia,  and  were  part  of  the  whole 
well-designed  strategical  plan.     To  a  certain  extent : 
Otho's  hopes  were  realised.     Twice  Valens  had  to 
detach  troops  from  his  column  and  send  them  to  the1 
coast.      And  the  disturbance  caused  in  his  army  on 
one  of  these  occasions,  which  has  been  narrated,3 1 
was    a   greater    success    for    Otho   than    he   could 
reasonably  have  anticipated.      It  would  indeed  have 
been  a  curious  freak  on  the  part  of  the  Genius  of  War 

1  "The  mania  of  the  British  leaders  for  detaching  troops  for  minor 
operations,  whereby  they  weakened  themselves  prior  to  a  crisis,  often  had 
disastrous  results,  and  might  easily  have  led  to  a  catastrophe  at  Driefontein 
also"  {German  Official  History  of  the  Boer  War,  Eng.  Trans,  ii.  p.  49). 

At  Driefontein  we  were  compelled  to  throw  all  our  scanty  reserves  into 
the  line  of  battle  to  carry  the  position  in  one  last  effort. 

-  Tacitus  himself  rightly  comments  on  the  small  importance  of  Corsica  in 
this  connection,  ii.  16. 

3  See  above,  p.  62. 


;8  THE  CAMPAIGN  OF  OTHO 


CU.  1 


had  the  command  of  the  sea  in  this  year  by  the  foe 
cost  Valens  his  life,  thanks  to  the  mutinous  spirit 
of  his  troops.  In  this  event  Caecina  would  in  all 
likelihood  have  been  crushed,  and  Valens'  leaderless 
column  have  recoiled  on  Gaul.  Such  events  were 
not  to  happen,  nor  could  Otho  have  expected  them. 
But  that  the  action  of  his  fleet  did  not  bring  him  at 
least  a  greater  measure  of  success  than  it  did  was 
due,  not  to  the  Emperor's  strategy,  but  to  his  officers' 
misuse  of  their  chances.  Had  these,  after  their 
victory  over  the  Vitellians  on  the  coast,  dared  to 
push  up  country  in  the  direction  of  Briancon 
instead  of  weakly  retiring  to  Albenga,  Valens' 
whole  march  over  the  Alps  might  have  been 
arrested  for  at  least  some  days,  if  not  weeks.  This 
delay  might  well  have  resulted,  as  events  showed, 
in  the  entire  destruction  of  Caecina's  column  at 
Cremona.  For  only  Valens'  arrival  saved  his 
colleague.  The  Othonian  force  marching  on 
Briant^on  could  probably  have  retreated  in  safety 
to  its  base  the  fleet,  had  Valens,  as  might  then  be 
hoped,  turned  savagely  upon  it  with  his  whole 
army.  Even  had  it  been  cut  off  and  perished,  it 
would  by  its  defeat  have  won  victory  for  its  Emperor 
in  the  main  campaign.  In  this  way  failure  some- 
times spells  triumph  in  the  lesson-book  of  war.  In 
this  way  Sir  John  Moore,  in  like  manner  trusting  to 
the  fleet  at  Corunna  for  his  base,  splendidly  hurled 
his  little  column  at  Napoleon's  line  of  communica- 
tions in  North  Spain  and  saved  the  Peninsula.  But 
Otho's  officers  lacked  either  the  pluck  or  the  insight 
of  such  a  general.      It  may  be  that  they  distrusted 


sec.v  AND  THE  VITELLIANS  79 

the  uncertain  character  of  their  initial  success  ;  it  may 
be  that  they  dared  not  run  the  great  risk  involved. 
It  is  probable   that   they  were  quarrelling   among 
themselves.    Whatever  the  reason,  the  fleet  and  the' 
force  on  board  of  it  missed  their  chances,  and  Otho's  1 
position  of  defence  was  greatly  weakened  thereby. 

(2)   The  Mobilisation  of  the  Danube  Ar?ny. — A 
second  and  more  damaging  flaw  in  Otho's  armour  * 
of  defence  was  the  fact  that  the  Army  of  the  Danube 
was  concentrating  at  Aquileia  far  too  slowly.     The  ■ 
four  legions   of  Pannonia  and   Dalmatia  were  the\ 
nearest  to  the  scene  of  the  coming  struggle.     To 
some  extent  they  realised  the  importance  of  speed, 
and,  since  they  themselves  were  not   yet  ready  to 
march  as  whole  units,  each  sent  forward  a  special 
detachment  ("vexillum")  of  its  men  in  front  of  it. 
These  detachments  were  each  two  thousand  strong. 
But  even  of  these  only  one  arrived  in  time  to  takes 
part  in  the  first  battle  in  the  field,  that  of  Locus  I 
Castorum  on  April  6.      Behind  this  solitary  detach- 
ment  were  the   three   similar  ones ;    behind   these 
the   bulk   of  their  legions ;   and   farthest   from  the 
theatre  of  war  were   the   Moesian   legions,  slowly 
assembling.      This    slackness  on   the    part   of  the 
Danube  army,  says  the  Roman  historian,  was  due  I 
to    over  -  confidence :    "E    fiducia   tarditas   inerat." 
Yet  it   was   a  time   when    every  day   that    passed 
before  that  army  came  in  strength  might  spell  ruin 
to  their  cause  and  to  their  Emperor.     There  is  no 
excuse   for   their  deliberation   in  movement  or  for  1 
their  confidence.1 

1  See  Note  A,  "  The  Movements  of  the  Danube  Legions." 


8o  THE  CAMPAIGN  OF  OTHO         ch.i 

(3)  The  Loss  of  Italy  north  of  the  Po,  including 
Cremona. — Finally,  the  speed  of  Caecina's  advance 
guard  had  already  cost  Otho  dear.  When  Spurinna 
arrived  at  Placentia,  he  found  that  a  cohors  Pan- 
noniorum  of  the  Army  of  Defence  had  already  been 
captured  at  Cremona,  and  this  strongly-walled  city 
fell  into  the  enemy's  hands  about  the  same  time. 
I  "  All  the  fields  and  cities  between  the  Po  and  the 
Alps  were  held  by  the  Vitellian  forces."  l  The  loss 
of  Cremona,  though  unavoidable,  was  a  serious 
disaster.  This  city  lies  on  the  north  bank  of  the 
Po,  some  twenty-five  miles  from  Piacenza  to  the 
west,2  and  forty  from  Mantua  on  the  east.  It  had 
originally  been  built  three  centuries  before  as  a 
Roman  outpost  north  of  the  river  to  keep  the 
Gallic  tribes  in  check,  and  it  was  still  strongly 
fortified.  There  does  not  seem  to  have  been  any 
bridge  in  a.d.  69  at  the  city  over  the  river.  To-day 
there  is  a  bridge  over  the  Po  of  enormous  structure 
and  great  length,  separated  from  the  city  gate  by  a 
mile  of  weary  road.  But  Cremona  in  a.d.  69  pos- 
sessed no  "  bridge-head  "  on  the  southern  bank  of 
the  Po,  which  would  have  put  a  force  holding  the 
town  in  a  position  to  turn  the  flank  of  the  gar- 
rison of  Placentia,  and  render  the  holding  of  that 
fortress  useless.  Yet  its  seizure  by  Caecina  gave 
the  Vitellian  general  a  strong  base  of  operations  for 
movements  of  offence  against  the  Othonian  line  of 
defence,  a  place  of  refuge  in  the  event  of  defeat 
or  the  advance  of  the  enemy  in  force,  and  a  safe 

1  Tac.  ii.  17. 
-  By  road  vid  Codogno  on  the  left  bank  ;  in  a  bee-line,  16  miles  only. 


sbc. vi  AND  THE  VITELLIANS  81 

resting-place  in  which  he  could  await  the  arrival  of 
his  colleague,  Valens.  Its  loss,  therefore,  to  Otho 
was  serious.  It  was  only  the  Emperor's  strategical 
brilliance  which  later  all  but  turned  this  loss  to 
positive  gain,  and  made  the  fortress  not  Caecina's 
harbour  of  refuge  but  a  prison-house  for  him  and 
his  army.1 

In  these  three  respects,  therefore,  Otho's  position 
at  the  beginning  of  the  "  tactical  chapter "  in  the 
history  of  the  campaign  was  not  so  favourable  as  it 
would  have  been  had  his  generals  been  abler  men. 
He  himself  had  made  no  mistake  in  his  measures^ 
of  defence.  But  the  force  on  shipboard  had  not 
done  its  duty ;  the  legions  of  the  Danube  army 
were  sluggish  ;  and  all  the  north  bank  of  the  Po  I 
from  Alps  to  Cremona  had  fallen  at  once  into  the 
enemy's  hands.  But  if,  so  far,  Fortune  had  been 
unkind  to  the  Emperor,  yet  his  main  line  of  defence 
south  of  the  river  was  stoutly  held,  and  Fortune! 
never  yet  showered  all  her  favours  on  one  side  only 
in  any  war.  And  now  at  last  the  armies  were  in 
touch.'  Spurinna  at  Placentia,  Gallus  at  Mantua, 
held  the  two  ends  of  the  defensive  position.  They 
had  to  expect  that  the  enemy  would  seek  to  break 
through,  and  that  immediately. 

§  6.    The  First  Encounters 

When  Spurinna  reached  Placentia  he  found  the, 
enemy  hard  at  hand.     An  outpost  force  of  his  own  I 

1  See    below,   §   7.      Cf.    Note  B,    "The   Capture   of  Cremona   by   the 
Vitellians." 

G 


82  THE  CAMPAIGN  OF  OTHO         ch.i 

troops,  numbering  eleven  hundred  men,  was  cut  off 
by  them  between  the  fortress  and  Ticinum,  the 
present  town  of  Pavia.  His  patrols  also  speedily 
came  into  touch  with  their  skirmishers  and  were 
roughly  handled  by  them.  Their  Batavian  and 
German  auxiliaries,  excited  by  success  and  by  the 
sight  of  a  great  river  again  after  their  many  weary 
miles  of  mountain  roads,  adepts  as  they  were  at 
the  means  of  crossing  such  an  obstacle,  swarmed 
across  the  Po  higher  up  stream  and  bore  down  on 
Spurinna's  lines.  The  general,  however,  was  con- 
vinced, as  was  indeed  the  case,  that  Caecina  himself 
—  and  his  legionary  army  had  not  yet  arrived.  It 
was  his  obvious  policy  to  make  a  reconnaissance  in 
force  westwards  along  the  river  bank  to  discover, 
;if  he    could,   the    strength    and    intentions   of  the 


enemy.1 


With  this  intent  he  marched  out  from  Placentia 
over  the  Po  a  day's  march,  and  when  night  fell 
fortified   his   position   by  the   river.2     He   had   not 


1  Tacitus  asserts  (ii.  18)  that  he  was  compelled  to  march  out  against  his 
own  wishes  by  his  insubordinate  troops,  who  realised  their  folly  next  day. 
This  seems  to  me  a  silly  story,  told  later  in  the  camp,  where  generals'  actions 
are  always  pulled  to  pieces. 

2  "  In  conspectu  Padus  "  (ii.  19).  These  words  greatly  trouble  the  German 
editors.  Heraeus,  e.g.,  proposes  "e  conspectu  Padus,"  on  the  ground  that, 
as  Placentia  lay  on  the  river  and  as  Spurinna  crossed  it,  he  must  have  led  his 
men  north,  and  have  been  "out  of  sight  of  the  river,"  therefore,  when  he 
halted  them.  If  Oxford  were  threatened  by  an  enemy  who  lay  in  the  direc- 
tion of  Reading,  the  commander  of  the  Oxford  garrison  would  surely  conduct 
his  reconnaissance  and  feel  his  way  along  the  Thames,  choosing  for  his  night's 
camp  a  position  at  Abingdon.  Would  he  strike  for  the  Ridge  Way  at,  e.g., 
Wantage?  Classen  proposes  "in  conspectu  hostis."  Would  Spurinna  be 
likely  to  choose  such  a  position  for  his  camp  ?  In  the  first  place,  it  would 
have  been  foolish.  In  the  second  place,  during  his  reconnaissance  he  did 
not  come  into  touch  with  the  enemy  at  all.  The  words  "in  conspectu 
Padus  "  present  no  difficulty  at  all,  from  a  military  point  of  view.  The  river 
would  be  an  additional  protection  to  the  camp. 


sec. vi  AND  THE  VITELLIANS  83 

met  with  any  opposition,  and  to  advance  farther 
along  the  stream  next  day  was  not  prudent.  He 
therefore  left  patrols  along  the  river  bank  and  with- 
drew to  the  fortress.  His  reconnaissance  had  been 
fruitless  of  results,  and  his  troops  were  well  content 
to  be  back  again  in  shelter. 

Shortly  afterwards  Caecina  himself  with  his  main 
army  arrived  outside  the  fortress.  His  march 
through  the  plain  of  North  Italy  had  been  a  rapid 
one,  and  he  had  kept  his  men  well  in  hand,  not  allow- 
ing them  to  plunder  the  towns  through  which  they 
passed.  Caecina,  in  fact,  had  "  left  his  cruelty  and 
profligacy  on  the  other  side  of  the  Alps."  In 
presence  of  the  enemy,  other  and  more  soldierly 
qualities  had  to  take  the  place  of  these.  The  citizens 
of  those  towns  were  reduced  to  grumbling  at 
the  general's  "  barbarian  costume,"  disliking  the 
trousers  which  he  wore,  and  at  his  wife  Salonina's 
gallant  display  on  horseback  in  a  purple  robe. 
They  were  happy  that  they  had  no  other  cause  for 
grumbling  than  a  man's  novel  taste  in  dress  and 
a  woman's  usual  love  of  show  and  finery.  Caecina 
therefore  had  no  reason  to  dread  any  rising  in  his 
rear  when  he  crossed  the  Po  above  Placentia  and 
marched  down-stream  upon  the  town.  It  was  indeed' 
imperative  for  him  to  seek  to  take  that  fortress. 
If  he  neglected  it  and  passed  it  by,  it  threatened 
his  one  line  of  communications  with  Valens' 
column.  Of  this  general's  approach  there  was  as 
yet  no  sign,  and  Caecina  must  keep  the  road  open 
at  least  on  the  north  of  the  river  as  far  as  Cremona, 
a  town  now  held  by  his  troops.      It  was  doubtless 


84  THE  CAMPAIGN  OF  OTHO         ch.i 

very  expedient  to  win  an  early  success.1  It  was  an 
attractive  method  of  attack  upon  the  foe's  line  of 
defence  to  assault  the  extreme  flank  of  it  and  seek 
to  "roll  it  up"  from  west  to  east.  But,  above  all, 
it  was  necessary  for  him  to  guard  his  own  com- 
munications north  of  the  river  from  the  constant 
menace  of  interruption  by  a  sally  over  the  bridge 
from  Placentia,  if  that  fortress  remained  in  posses- 
sion of  the  enemy's  garrison.  Caecina  was  bound 
to  get  possession  of  it  at  once  if  possible. 

Negotiations  were  opened  between  the  two 
sides,  but  resulted  in  nothing  save  mutual  revilings. 
It  was  indeed  "easier  to  blame  than  to  praise"  the 
characters  of  Otho  and  Vitellius.  Caecina  therefore 
wasted  little  time  on  words,  but  for  two  days 
delivered  a  vigorous  assault  on  the  fortress.  This 
was  stubbornly  and  successfully  resisted  by  the 
garrison,  and  the  Vitellians  sullenly  admitted  their 
first  reverse.  The  first  wave  of  attack  spent  its 
fury  in  vain  and  retreated.2  Caecina  drew  off  his 
defeated  army  and  marched  for  Cremona,  into  which 
town  he  threw  himself.  His  strategical  position 
was  one  already  of  some  risk.  To  the  west  lay  his 
imperilled  line  of  communications  and,  if  necessary, 
retreat.  In  front  of  him,  in  the  centre  of  the 
Othonian  line  of  defence,  was  a  small  but  active 
,  and  annoying  body  of  gladiators,  under  Martius 
Macer,  who  made  stinging  raids  over  the  river 
— a    little    hornets'    nest    which    it    was    hard    to 


1  This  is  the  one  and  only  motive  ascribed  to  Caecina  by  Tacitus  !     It  is 
by  far  the  weakest  of  the  three. 

-  Siege  of  Placentia  at  length  in  Tacitus,  ii.  1S-23. 


sec. vi  AND  THE  VITELLIANS  85 

reach.  And  on  the  east  a  third  enemy  speedily 
appeared. 

For  Gallus  at  Mantua,  on  hearing  of  the  assault 
on  Placentia,  had  moved  forward  towards  the  west, 
to  relieve  the  pressure  on  the  fortress,  the  import- 
ance of  which  he  well  knew.  While  on  the  road, 
north  of  the  river,  he  received  news  from  Spurinna 
that  Caecina  had  raised  the  siege  and  drawn  off 
his  troops.  Gallus  therefore  halted  his  army  and 
promptly  fortified  his  position.  He  was  then  at  a 
village  hitherto  unknown  to  history,  but  hencefor- 
ward to  be  doubly  famous  in  the  annals  of  war — the 
village  of  Bedriacum.1 

The  broad,  dusty  Italian  high-road  which  runs 
to-day  through  the  great  plain  of  the  Po  westwards 
from  Mantua,  after  crossing  the  Oglio,  one  of  the 
larger  tributaries  of  the  main  river,  some  fifteen 
miles  from  the  city,  passes  by  two  small  towns  named 
Bozzolo  and  Piadena  on  its  way  to  Cremona.  Mid- 
way between  these  towns,  which  are  nine  kilometres 
apart,,  and  a  quarter  of  a  mile  north  of  the  road, 
there  lies  a  tiny  hamlet,  consisting  of  a  church  and 
a  cluster  of  small  houses  nestling  together  under 
its  spire.  This  hamlet  is  named  Calvatone.  It  is 
as  peacefully  remote  from  the  dust  and  traffic  of 
the  highway  as  any  Cots  wold  village  which  just 
escapes  the  great  Bath  road.     But  here  at  Calvatone 

1   I  think  this  should  be  pronounced  Bedriacum,  despite  the  analog}-  of 
such  names  as  Moguntiacum.      For  Juvenal's  line  runs  : 

"  Bebriaci  in  campo  spolium  affectare  Palati"  (ii.  106). 
Of  course  the  poet  could  not  manage  Bebriacum  at  all,  and  it  is  possible 
that  he  may  have  taken  a  licence  of  mispronunciation,  as  modern  poets  have 
done  with  the  name  Trafalgar.      But,  as  we  have  no  other  evidence  at  all, 
perhaps  we  had  better  half-heartedly  follow  the  poet. 


86  THE  CAMPAIGN  OF  OTHO         ch.i 

in  Roman  days  two  great  roads  met — that  from 
Hostilia  at  the  crossing  of  the  Po  to  the  south,  and 
that  from  Mantua,  both  making  for  Cremona.  The 
village  where  they  joined  was  named  Bedriacum. 
As  a  military  post,  when  war  swept  over  the  plain, 
it  was  both  important  and  easily  defensible.  A 
force  entrenched  here  covered  both  the  line  of  com- 
munications with  Aquileia  and  the  east,  and  also 
the  crossing  of  the  river  by  the  road  which  led  to 
the  great  highway  to  Ariminum  and  Rome.  A 
short  distance  to  the  north  of  it  there  flowed  the 
river  Oglio,  a  broad  muddy  stream  with  steep  high 
bank  on  the  southern  side,  not  easy  to  cross,  whose 
rapid  and  broad  current  turns  two  long  rows  of 
water-mills  to-day,  not  unlike  those  on  the  Danube. 
There  might  have  been  one  danger  to  Gallus  when 
he  encamped  his  legions  here — the  fear  lest  the  foe 
should  come  down  into  Italy  by  the  Brenner  Pass 
and  so  cut  him  off  from  the  Danube  army.  But  by 
this  time  it  was  known  that  Caecina  had  not  chosen 
this  way,  and  there  were  no  enemy  in  the  pass.1 

At  Bedriacum,  therefore,  Gallus  halted  his 
troops.  And  here  he  was  soon  afterwards  joined 
by  the  main  Army  of  Italy  under  its  generals 
Suetonius  Paulinus  and  Marius  Celsus.  A  most 
welcome  detachment  also  of  two  thousand  men 
from  the  Thirteenth  legion  of  Pannonia  marched 
into  camp,  seeming  an  earnest  of  the  rest  of  the 
Army  of  the  Danube  to  come,  and  with  them  came 
six  auxiliary  cohorts  and  one  cavalry  squadron 
belonging   to    this   army.     Caecina  and  his  troops 

1  See  Note  C,  "The  Site  of  Bedriacum." 


sec.  vi  AND  THE  VITELLIANS  87 

were  still  at  Cremona,  a  few  miles  down  the  broad 
white  road,  alone.  There  was  no  news  of  Valens' 
column.  And  the  garrison  at  Placentia  kept  grim 
watch  upon  his  line  of  retreat. 

But  the  Vitellian  general  was  not  down-hearted. 
Always  in  war  the  force  which  attacks  is  likely  to 
fight  better  than  that  which  is  attacked.  Caecina's 
army  was  a  strong  one — stronger  in  numbers  almost 
certainly  than  the  full  muster  of  Otho's  men  at 
Bedriacum.  He  resolved  at  once  to  take  the 
offensive,  lest  delay  should  bring  the  Danube  army 
upon  the  scene.  Purely  frontal  attacks,  however, 
were  hazardous  and  costly.  He  planned  to  lure 
the  Othonians  into  a  snare  and  so  destroy  their 
army. 

Some  twelve  Roman  miles  east  of  Cremona,1  the 
Postumian  way  (which  as  a  great  Roman  military 
road  was  raised  up  high  above  the  level  of  the  rich, 
damp,  cultivated  land  on  either  side  of  it)  ran  for 
a  short  distance  through  woods  on  both  sides.  On 
emerging  from  the  trees  it  passed  through  vine- 
yards. These  Italian  vineyards  of  the  north  are 
not  the  forest  of  short  upright  stakes,  such  as  those 
which  line  and  disfigure  the  banks  of  the  Rhine  and 
the  Moselle,  but  in  North  Italy  the  vine  stems  are 
linked  from  fruit-tree  to  fruit-tree  in  long  droop- 
ing and  graceful  festoons,  while  the  rich  earth  in 
April  supplies  enough  nourishment  also  to  cover 
the  ground  under  the  vines  and  under  the  fruit- 
trees  with  a  green  carpet  of  corn.  It  is  indeed 
hard    to    make    one's    way    through    the    fields    by 

1  I.e.  about  io|  English  miles. 


88  THE  CAMPAIGN  OF  OTHO 

Cremona  save  where  road  or  track  is  cut  through 
the  vineyards. 

Here,  just  where  the  road  left  the  cover  of  the 
trees,  Caecina  placed  an  ambush,  at  a  place  called 
Locus  Castorum,  ten  miles  away  from  the  camp  at 
Bedriacum.  Some  auxiliary  infantry  were  hidden  in 
the  woods  on  either  side  of  the  way.  The  cavalry 
were  ordered  to  ride  forward  towards  the  enemy 
and  provoke  them  to  attack.  Then  they  were  to 
fall  back  along  the  road  in  feigned  retreat,  drawing 
the  Othonians  in  pursuit  after  them  through  the 
wood,  when  the  infantry  would  sally  out  from  the 
trees  and  have  them  at  their  mercy  on  either 
flank. 

It  was  not  a  very  brilliant  plan,  and  Caecina 
might  have  foreseen  that,  in  a  time  of  civil  war,  it 
would  be  promptly  betrayed  to  the  other  side. 
Moreover,  he  managed  the  ordering  of  his  troops 
clumsily  enough.  No  Roman  general,  except 
Julius  Caesar,  seems  ever  to  have  been  a  master 
of  ambushes  and  surprises.  In  this  case  the  biter 
was  bit.  Otho's  generals,  Suetonius  and  Celsus, 
were  duly  told  the  whole  of  Caecina's  plan,  and  took 
their  measures  accordingly.  Tijey  marched  out 
from  camp,  three  Praetorian  cohorts  in  column  on 
the  road  :  on  the  right  flank  the  First  legion,  two 
auxiliary  cohorts  of  foot,  and  five  hundred  cavalry  ; 
on  the  left  flank  the  two  thousand  legionaries  of  the 
Thirteenth  legion,  four  auxiliary  cohorts  of  foot,  and 
five  hundred  cavalry.  These  cavalry  formed  the 
extreme  wing  on  either  side,  and  finally  one  thousand 
other  cavalry  formed  the  rear-guard,  keeping  open 


sec  vi  AND  THE  VITELLIANS  89 

the  communications  with  the  camp  at  Bedriacum, 
ready,  in  case  of  retreat,  to  open  out  and  let  the' 
infantry  through,   and  cover  their  retirement.      In 
this  order  the  Othonians  moved  out   to  meet  the 
foe.     Presently  the  Vitellian  cavalry  came  in  sight,— 
charging  along  the  road.     At  this  Suetonius  halted 
his  infantry,  throwing  forward  slightly  his  auxiliary 
foot  on  either  side  of  the  road  ;  the  cavalry  under 
Celsus  slowly  trotted  forward  as  if  to  receive  the 
charge  of  the  enemy's  horse.     These,  obedient  to— 
orders,  promptly  began  to  retire.     But  then  Celsus  — 
for  his  part  halted  his  men.     At   this  unexpected 
turn  of  events  the  Vitellian  infantry  in  the  wood  """ 
seem  to  have  lost  patience  ;  for  they  rose  from  their 
ambush,  and    pouring    tumultuously  out  upon  the 
road  came  charging  down  upon  Celsus,  with  their 
own  cavalry,  it  may  be,  thrust  forward  in  their  van 
like  the  foam  of  a  wave.     Nothing  could  have  suited-^ 
Celsus  better.     Quietly  withdrawing  his  men,  when 
he  came  into   touch   with   his   infantry,   he   passed 
through  their  ranks,  which  opened  out  to  allow  this, 
and  the  pursuing  Vitellians  suddenly  found  them-" — 
selves  trapped  in  the  middle  of  the  enemy's  foot. 
The  Guards  faced  them  in  the  front ;  the  auxiliaries 
threatened  them  on  the  flanks  ;  and  Celsus  with  his 
thousand  horse,  who  had   led  them  into  the  snare, 
now  emerged  again  from  behind  his  infantry  and 
threatened  to  fall  upon  them  on  the  rear.     It  was,- — 
in  fact,  very  nearly  a  second  Cannae.     For  it  was  by 
this  same  device  that  Hannibal  long  years  ago  had 
caught  and  massacred  the  Romans. 

But  Suetonius   Paulinus  was  no   Hannibal,  and  1 


90  THE  CAMPAIGN  OF  OTHO 

all  became  confusion.  While  he  was  hesitating  and 
troubling  himself  about  ditches  and  extensions  to 
the  flank,  as  if  it  were  drill  upon  a  field  day,  not  a 
grim  melde  on  the  field  of  battle,  the  Vitellians 
slipped  away  from  the  closing  circle  into  the  com- 
fortable shelter  of  the  friendly  vineyards.  When 
however  they,  with  greater  daring  than  prudence, 
reappeared  from  cover,  the  Othonians  charged  at 
'  last,  and  Caecina's  men  were  driven  in  rout  off  the 
field.  Nor  was  their  plight  in  any  way  redressed 
by  their  general ;  for  he  had  hurriedly  sent  to 
Cremona  for  reinforcements,  and  these  came  hasten- 
ing down  the  road  in  small  detachments,  only  to  be 
involved  in  the  general  rout  and  to  make  confusion 
worse  confounded.  The  battle  on  the  road  became 
a  nightmare  of  disorder,  until  Suetonius  checked 
the  pursuit  and  recalled  his  men.  The  discomfited 
Vitellians,  grateful  for  the  respite,  ignominiously 
made  good  their  escape  to  Cremona.  They  were 
sated  with  ambushes  for  the  future.  They  were 
the  smaller  of  the  two  Vitellian  armies.  Where, 
they  disconsolately  asked,  were  Valens  and  his 
larger  force  ?  Were  they  themselves,  a  "  mere 
handful "  in  comparison,1  to  bear  the  whole  brunt 
of  the  fighting  ?  2 

In  reviewing  the  story  of  this  extraordinary 
battle,  the  reader  must  see  that  both  Suetonius  and 
Caecina  may  be  blamed  too  easily.     There  were,  it 

1    "  Tanto  pauciores,"  ii.  30. 

J  Tacitus's  account  of  the  battle  of  Locus  Castorum  (ii.  23-26)  is  in  itself 
a  historical  nightmare.  The  account  in  the  text  is  an  attempt  to  make  sense 
out  of  it.  But,  even  so,  infantry  and  cavalry  must  have  been  horribly  mixed 
up  together.  A  Roman  road  is  not  a  Salisbury  Plain.  And  cf.  Hardy, 
Plutarch 's  Galba  and  Otho,  pp.  239-241. 


AND  THE  VITELLIANS  91 


v 


is  true,  bad  mistakes  made  on  both  sides.     Caecina's 
ambush  was  a  clumsy  affair ;  he  could  not  keep  his  I 
troops  in  hand,  and  when  the   battle   took   an   un- 
expected course  and  went  against  him,  he  lost  his 
head    completely.       To    hurry    reinforcements    in 
driblets  into  the  firing  line  when  it  is  yielding  is  a 
not  uncommon  device  of  inferior  generals,  and  is  I 
always  worse  than  useless.     But  in  his  general  idea 
of  attacking  the  Othonians  at  Bedriacum  promptly 
before  the  Danube  army  had  arrived  to  their  succour) 
Caecina  was  absolutely  right.     The  Roman  historian 
blames    him    for   attacking    "  more    hurriedly    than 
wisely,"  and  ascribes  his  haste  to  a  mere  jealous  fear 
lest,   if  he   waited,   Valens  should   acquire  all   the/ 
renown  for  the  war.1     Tacitus  has  a  genius  for  mis- 
understanding the  essentials  of  a  military  situation. 
It  might  so  easily  have  happened  that  the  Danube 
army  outstripped  Valens  in  its  coming.     Caecina's 
was  a  sound  strategical  plan  spoilt  by  faulty  tactical! 
execution. 

In   like    manner,    Suetonius   (who   displayed   nox 
small  tactical  skill)  was  over-cautious  on  the  field  of 
battle,  and  let  slip  a  good  opportunity  for  crushing 
the  enemy.     The  vigour  which  he  had  once  shown  1 
in  the  black  days  of  the  rebellion  in  Britain  seemed  \ 
to  have  deserted  him.     In  battle,  something  must.v 
be  risked  or  nothing  will  be  achieved.      But  that 
he  was  thoroughly  justified  in  calling  his  men  off 
from  the  pursuit  can  hardly  be  doubted,  although 
men  blamed  him  for  this  at  the  time,  and  his  reputa-j 
tion    has    suffered    for    this    ever   since.       Had   he\ 

1  Tac.  ii.  24. 


92  THE  CAMPAIGN  OF  OTHO 

allowed  the  pursuit  to  continue,  his  critics  urged, 
(the  Vitellian  army  must  have  been  annihilated. 
The  result  in  reality  would  have  been  very  different. 
Caecina's  force  must  have  outnumbered  his  own  in 
a  proportion  of  three  to  two.  The  fighting  had 
taken  place  some  ten  miles  from  his  base  camp  at 
Bedriacum.  The  pursuit,  if  allowed,  would  have 
I  followed  yet  another  ten  miles  up  to  the  walls  of 
I  Cremona.  There  it  would  have  been  stopped 
abruptly  by  Caecina's  entrenchments  ;  and  Suet- 
onius' straggling,  exhausted,  tired  troops  would 
at  once  have  been  exposed  to  a  counter-stroke  of 
the  rudest  and  most  effective  kind.  In  refusing  to 
permit  his  men  to  incur  this  risk,  Otho's  general 
displayed  a  sound  common  sense  which  is  lacking 
in  the  critics  of  his  generalship.1 

§  7.    The  Strategies  of  the  Final  Struggle 

-*  A.  The  Vitellian"  Strategy  of  Penetration" — By 
the  time  that  Caecina  had  fought  and  lost  the  battle 
of  Locus  Castorum,  Valens  had  at  last  crossed  the 
Alps  and  arrived  at  Ticinum.  When  the  news  of 
the  defeat  reached  him  he  acted  promptly,  and 
marched  at  once  to  Cremona,  where  he  joined 
forces  with  his  colleague.  At  last  the  Army  of 
Germany  was  united.  And  not  only  were  Valens' 
numbers  by  this  time  nearly  twice  as  large  as  those 
of  Caecina,  but  the  misfortunes  of  the  smaller  army 
stiffened  discipline    through    the   entire    force.       It 

1   Even  Tacitus  seems  to  approve  and  understand  Suetonius's  caution  (ii. 
26,  fin.). 


skcvii         AND  THE  VITELLIANS  93 

was  realised  that  the  enemy  were  prepared  to  offer 
a  stout  resistance.  They  had  already  given  unwel- 
come proof  of  their  valour.  The  crisis  of  the 
whole  campaign  was  at  hand,  and  the  Vitellians 
braced  themselves  for  a  sterner  effort. 

In  camp  together  at  Cremona  the  two  generals 
took  counsel  what  to  do.  One  possible  plan,  now 
that  they  had  so  large  an  army  at  their  command, 
was  to  renew  the  direct  attack  upon  the  enemy  at 
Bedriacum,  not  now  by  any  attempt  at  a  lure  or 
ambush,  but  by  an  honest  frontal  attack  down  the 
Postumian  way.  But  they  hesitated  to  adopt  this 
plan.  The  last  fighting  had  gone  very  badly.  That 
Otho  was  now  himself  present  and  was  receiving 
reinforcements  was  well  known.  Any  day  might 
see  the  arrival  of  the  whole  Army  of  the  Danube 
in  his  camp.  Even  if  they  carried  the  position  by 
the  dangerous  method  of  a  frontal  attack  before  its 
arrival,  the  enemy  would  only  fall  back  upon  the 
support  and  shelter  of  this  army.  Caecina's  energy 
and  impulsiveness  had  ended  in  defeat.  The  older 
and  more  cautiousJValens  induced  his  colleague  to 
agree  to  a  more  cautious  strategy.  Their  Emperor 
Vitellius  would  presently  arrive  with  large  reinforce- 
ments. They  must  wait  for  him,  unless  themselves 
attacked.  On  his  arrival,  with  the  whole  of  their 
resources  available,  they  would  proceed  to  the  final 
struggle.  But  meanwhile  much  could  be  done. 
The  assault  on  Placentia  iiad  been  a  bad  failure ; 
but  the  fortress  still  menaced  both  their  own  position 
and  Vitellius'  when  he  came.  From  it  the  enemy's 
line  stretched  eastwards  along  the  southern  bank  of 


/ 


94  THE  CAMPAIGN  OF  OTHO         ch.  i 

the  Po,  and  immediately  opposite  Cremona  Macer 
and  his  gladiators  were  annoying  them.  They 
-  decided  to  penetrate  the  enemy's  line  here  in  the 
centre,  where  very  probably  it  was  weakest,  as  it 
relied  on  the  cover  afforded  by  the  river.  Certainly 
-  the  troops  of  the  enemy  here  employed  were  not 
his  best  soldiers.  Every  Roman  army  was  skilled  in 
the  art  of  bridge-building  ;  therefore  the  engineers 
were  set  to  work  under  Caecina's  directions  to 
bridge  the  Po  opposite  Cremona. 

The  attempt  was  vigorously  opposed  by  Macer 
and  his  gladiators  on  the  southern  bank.  Every 
device  was  employed  to  interrupt  the  building.  But 
despite  direct  attacks  on  the  bridge  by  boats,  and 
the  use  of  fire-ships,  the  Vitellians  were  able,  by 
dint  of  steadfast  and,  for  the  most  part,  successful 
fighting,  slowly  to  push  on  the  work.  It  appeared 
as  if  the  plan  of  the  tactical  penetration  of  the 
hostile  line  would  be  achieved  and  Placentia  hope- 
lessly isolated.  Caecina  was  urging  on  the  effort 
when  a  hurried  message  reached  him  from  Valens' 
headquarters  in  Cremona.  The  enemy  had  appeared 
in  force  marching  for  the  town.  He,  Valens,  was 
moving  out  to  face  them.  Caecina  must  instantly 
bring  all  his  available  men  up  to  the  front  to 
assist. 

In  fact,  while  the  Vitellian  commanders  had  been 
busy  on  their  own  tactical  plan,  the  Othonians  had 
been  far  from  idle.  If  the  Vitellians  remained  quiet 
at  Cremona,  the  initiative  had  passed  to  Otho's 
men,  so  long  as  their  general  on  the  river  made 
good  the  defence  of  the  centre.     The  Emperor  had 


sec.  vii         AND  THE  VITELLIANS  95 

seized  the  chance  offered  him  by  Valens'  cautious 
strategy.1 

B.  Othos  Council  of  War} — Shortly  after  the 
victory  of  Locus  Castorum,  the  Othonian  generals 
welcomed  the  arrival  into  camp  of  the  Emperor  and' 
his  brother  Titianus.  Otho  had  summoned  Titianus 
from  Rome,  intending  him  to  take  command,  when 
he  learnt  that  Suetonius  had  displeased  the  troops* 
by  checking  their  pursuit  after  the  battle.  But, 
besides  the  full  muster  of  the  Army  of  Italy,  the, 
Army  of  the  Danube  was  now  fast  approaching.^ 
Already  after  the  victory  the  rest  of  the  Thirteenth 
legion,  under  its  legate  Vedius  Aquila,  had 
arrived  in  camp,  to  join  its  vexillum,  which  it  had 
sent  before  in  time  to  take  part  in  the  fighting. 
There  also  came  a  detachment  of  the  Four- 
teenth legion  from  Dalmatia,  and  a  second  cavalry 
squadron  belonging  to  the  Danube  forces.3  To 
increase,  their  numbers  still  further,  Spurinna  wasj 
summoned  from  Placentia  with  the  bulk  of  the) 
garrison  of  that  fortress.  He  left  only  enough  in 
the  town  to  hold  it  against  sudden  surprise.  In 
case  of  any  more  serious  emergency  it  could  easily 
be  reinforced  again. 

By  the  second  week  of  April  these  forces  were] 
mustered    at    Bedriacum.       The    enemy   lay  quiet 
at  Cremona,  trying,  it  seemed,  to  bridge  the  river,, 
and  Macer  must  frustrate  this.     But  they  certainly1 
showed  no  signs  of  advancing  to  attack  Bedriacumi;' 

1  The  Bridge  -  Building :    see    Tac.   ii.    34-36,    41;    Plutarch,    Otho,    10. 
On  Tacitus'  short-sighted  view  of  this  see  Note  E. 

2  Tac.  ii.  31-33. 

3  Tac.  ii.  24,  43,  66  ;  iii.  2. 


96  THE  CAMPAIGN  OF  OTHO 

Otho  summoned  a  council  of  war  to  discuss  the 
situation. 

At  this  council  of  war  the  three  generals 
Suetonius,  Celsus,  and  Gallus  agreed  in  the  advice 
which  they  offered.  All  three  had  been  remarkably 
successful  hitherto  by  their  defensive  strategy. 
Thanks  to  it  the  victory  of  Locus  Castorum  had 
been  won.  They  had,  it  is  true,  been  reinforced, 
but  the  Vitellians  had  been  much  more  strongly 
reinforced.  If  the  enemy  did  not  move  to  the 
attack,  why  should  they  themselves  desert  the  safe 
course  and  take  the  offensive  ?  The  bridge-build- 
ing was  not  a  matter  of  great  concern.  It  was  an 
easy  matter  to  look  to  this. 

Suetonius  therefore,  supported  by  the  other  two, 
strongly  urged  Otho  to  remain  on  the  defensive 
luntil  the  summer  came.  As  a  "second-best" 
strategy,  if  the  Emperor  should  be  reluctant  to  wait 
so  long,  let  him  at  least  do  nothing  for  a  few  days 
longer  until  the  bulk  of  the  Fourteenth  legion 
and  the  "  Moesian  army  "  arrived  in  camp.  Already 
detachments  of  the  three  Moesian  legions  had 
arrived  at  Aquileia.1  But  the  three  generals,  for 
their  part,  advised  the  first  plan,  of  a  strict  defen- 
sive until  the  summer.  Their  arguments  were,  in 
I  the  main,  three  in  number. 

In  the  first  place,  they  urged,  the  enemy  could 
expect  very  few  more  reinforcements,  if  any.  The 
Gallic  provinces  were  in  a  ferment :  Narbonensis, 
for  instance,  was  seriously  alarmed  by  the  operations 
of  the  fleet.     The  troops  in  Britain  had  their  own 

1  Suetonius,   Vesp.  6. 


sec.vm         AND  THE  VITELLIANS  97 

native  foes  in  the  island  to  keep  them  busy,  and, 
moreover,  the  sea  rolled  between.  Spain  had 
scarcely  any  troops  to  send.  The  bank  of  the 
Rhine  had  to  be  guarded  by  some  troops  against 
the  tribes  over  that  river.  The  generals  therefore 
were  of  opinion  that  even  when  Vitellius  himself! 
arrived,  he  would  bring  but  a  scanty  force  with  him/ 

In  the  next  place,  delay  and  defence  on  their 
own  part  would  involve  the  foe  in  very  serious 
commissariat  difficulties.  No  supplies  could  reach 
him  by  sea.  The  Alps  were  a  great  hindrance  to 
the  carriage  of  supplies  by  land,  and  the  strip  of 
country  on  which  he  depended — at  least,  immedi- 
ately for  food  and  forage — that  between  the  Po  and 
the  mountains,  was  laid  waste  and  already  exhausted.1 
All  that  was  wanted  to  complete  the  demoralisation 
of  a  hungry  army  of  northerners  was  the  hot  sun  of ' 
an  Italian  summer. 

Very  different    in   all    respects,   they   concluded 
finally,  was  their  own  position.     The   longer  they 
themselves   could  delay,  the   stronger  they   would   | 
become.     They  had  vast  quantities  of  money.      In 
civil    war    money    was    stronger    than    the    sword,  j 
Desertion    was    easily    bought.       Upon    their    side  J 
they  had  all  the  provinces  of  the  Danube  and  the 
east,  with  their  strong  and  vigorous  armies,  which 
as  yet  had  taken  no  part  in  the  fighting  and  were 
theirs  to  employ.      Italy  was  theirs;  theirs,  above 
all,    was    the    favour   of   the    Senatus    Populusque  I 
Romanus.    Theirs  was  the  rightful  cause.    Let  them 
then  wait   till   summer  came.     Their  position  was 

1  Surely  "  vastam  "  in  ii.  32  can  only  be  in  error  for  "  vastatam  "  ? 

H 


\ 


98  THE  CAMPAIGN  OF  OTHO 

impregnable,  as  Placentia's  resistance  had  shown. 
And  men  of  the  Army  of  Italy  had  no  reason  to 
dread  the  fierceness  of  an  Italian  July  sun:  they 
'  were  used  to  it. 

For  these  reasons  Suetonius,  supported  by  his 
colleagues,  gave  counsel  of  delay. 

There  is  no  reason  whatever  to  suspect  him  of 
any  but  the  best  motives.  A  wild  story,  indeed,  was 
presently  afloat  that  the  general  hoped  that  such 
delay  would  lead  both  armies  to  weary  of  their 
emperors  and  depose  them  ;  that  the  Senate  would 
thereupon  proceed  to  choose  another  prince  ;  that 
his  own  great  reputation  would  then  carry  the  elec- 
tion.1 This  indeed  was  a  wild  tale  enough.  The 
soldiers  of  both  armies  were  devoted  to  their 
respective  emperors.  Suetonius  was  already  un- 
popular with  his  own  men.  Even  the  Roman 
historian,  despite  his  pitiless  insight  into  men's 
baser  thoughts,  rejects  this  story.  When  Suetonius 
counselled  delay,  he  believed  this  to  be  Otho's 
'  wisest  strategy. 

And  Suetonius'  words  would  carry  weight.  He 
was  "  the  most  experienced  general  of  the  day." " 
His  fame  had  been  early  won  in  Mauretania,  when 
he  had  been  the  first  Roman  to  cross  the  Atlas 
range  of  mountains.3  If  imperilled  in  Britain 
recently,  that  fame  had  been  vindicated  by  his 
notable  victory  over  the  savages  in  the  island,  and 
he  had  crushed  their  furious  rebellion.4     He  came 


1  Tac.  ii.  37.  -  Tac.  ii.  31. 

3  Pliny,  Hist.  Nat.  v.  1,  14  ;  Dio  Cassius,  lx.  9. 

4  See  my  History  of  Nero,  chap.  vi.  §  3. 


sec.  vii         AND  THE  VITELLIANS  99 

to  the  council  of  war  fresh  from  the  field  of  victory  ; 
and  he  adduced  grave  arguments  to  support  his 
views. 

Yet  a  good  tactician  is  not  always  a  good  strate- 
gist, and  all  Suetonius'  successes  had  been  tactical. 
Now,  in  the  council  of  war,  he  was  urging  delay  , 
and  caution  as  a  strategical  plan.  That  such  a 
cautious  strategy  would  be  recommended  was 
indeed  likely. 

How  greatness  of  intellect,  which  in  times  of  peace  enjoys 
the  highest  consideration,  decreases  in  value  in  times  of  war 
when  opposed  to  will,  is  shown  by  the  result  of  nearly  every 
council  of  war.  It  cannot  be  denied  that  in  an  assembly  of 
experienced  and  capable  men,  the  highest  aggregate  of  intelli- 
gence must  be  collected.  Yet  Frederick  the  Great  was  right  in 
peremptorily  forbidding  his  generals  to  hold  a  council  of  war. 
That  keen  judge  of  human  nature  knew  full  well  that  nothing  is 
ever  gained  by  it  save  a  majority  for  the  "timid  party."  The 
intelligence  collected  in  a  council  of  war  is  wont  to  be  productive 
of  no  other,  advantage  but  that  of  assiduously  searching  out  all 
the  weak  points  of  an  army,  and  of  demonstrating  the  danger  of 
action.1 

Suetonius  had  certainly  produced  "  a  series  of 
plausible  arguments  for  leaving  well  alone."  -  But 
indeed  the  Emperor  had  reason  to  criticise  them.3 
It  certainly  was  not  true  that  Vitellius  would  bring" 
only  scanty  reinforcements.  Many  Gallic  levies 
and  as  many  as  eight  thousand  chosen  men  of  the 
Army  of  Britain,  in  spite  of  sea  and  savages,  joined 
his   march,  and   he   was   over  the   Alps   by   May.4 

1  Von  der  Goltz,  op.  tit.  p.  64. 

-  Ian  Hamilton,  Staff-Officer s  Scrap-Book,  p.   120. 

3  Tacitus  adduces  no  reasons  at  all  for  the  rejection  of  Suetonius'  advice  ; 
for  his  "imperitia  properantes  "  of  ii.  33  is  shallow.      Cf.  Note  E. 

4  Tac.  ii.  57  ;  cf.  i.  61. 


ioo         THE  CAMPAIGN  OF  OTHO         ch.  » 

I  Commissariat  difficulties,  also,  might  be  foolishly 
jexaggerated.  Caecina  had  kept  a  stern  control  of 
his  men  on  his  march  through  North  Italy,  and  the 
land  had  not  been  indiscriminately  pillaged.  All 
that  could  safely  be  admitted  was  that  the  power  of 
the  Italian  sun  in  July  is  fatal  to  any  prolonged 
activity  at  all.  But  despite  it,  the  immobile  defend- 
ing force  was  more  likely  to  lose  heart  because  of 
an  inactivity  imposed  by  command  than  was  the 
attacking  force  because  of  a  rest  imposed  by  the 
>heat.  It  was  on  all  accounts  inexpedient  to  prolong 
the  war  passively  till  the  summer.     Suetonius  was 

^  impressed  by  the  strength  of  their  line  of  defence  ; 

/  but  in  his  admiration  of  the  river  and  the  forts  he 
was  forgetful  of  men's  hearts.     Otho's  troops  de- 

y  manded  to  be  led  against  the  invader.  They  were 
not  machine-made  puppets,  that  they  could  have 
their  keen  zeal  blunted,  their  passion  disappointed, 
with  impunity. 

Therefore  Otho  rejected  the  advice  offered  him 
in  his  council  of  war.  He  determined  to  take  the 
offensive  against  the  enemy,  and  that  at  once.  His 
plan  embraced  all  his  forces  and  was  brilliant -in 
conception.  Because  it  failed  in  execution";  the 
ancient  historian  failed  to  understand  it. 

C.  Othds  "Strategy  of  Envelopment."  —  The 
Vitellian  generals  at  Cremona  were  seeking  to 
carry  out  the  plan  of  the  tactical  penetration  of 
the  enemy's  line.  Otho's  answer  to  this  was  a 
scheme  for  the  strategical  envelopment  of  the 
(entire  Vitellian  army.  The  elements  of  the 
scheme  were  these  : — 


sec.  vn         AND  THE  VITELLIANS  101 

A  large  part  of  the  Army  of  the  Danube  was 
already  at  Aquileia  ;  some  of  its  troops  were  perhaps 
already  on  the  ,road  between  Aquileia  and  Bedri- 
acum.1  This  army  was  to  concentrate  at  Bedriacum 
with  all  possible  speed. 

Meanwhile  the  troops  already  at  Bedriacum  were 
to  be  moved  to  the  west  of  the  enemy's  position  at 
Cremona,  and  flung  across  the  foe's  one  and  only 
line  of  communication  and  succour.  Seven  miles 
to  the  west  of  Cremona  the  river  Adua,  the  modern 
Adda,  flows  from  the  north  into  the  Po.  This  is 
a  broad  and  navigable  stream,  the  outflow  of  the 
waters  of  Lake  Como.  A  strong  force  posted  at'; 
the  confluence  of  the  Po  and  Adda,  behind  the 
latter  river,  would  isolate  an  enemy  at  Cremona. 
The  Cremona  force,  cooped  up  in  the  town,  reduced 
J  for  its  supplies  to  the  few  miles  of  country  in  its 
immediate  neighbourhood,  unable  to  force  its  way 
over  the  Po  to  the  south,  would  be  envejuped  and1 
invested.  Hunger  would  speedily  compel  it  to  try 
and  cut  its  way  out  through  the  force  posted  on  the 
Adda.  It  might,  it  is  true,  find  a  way  of  retreat 
open  on  the  north  towards  Brescia ;  but  this  way 
led  it  nowhere,  save  up  against  the  barrier  of  the 
unfriendly  Alps,  and  still  the  foe  on  the  Adda  lay 
between  it  and  its  Emperor  Vitellius.  If  it  could  , 
not  cut  its  way  through  this  force  it  must  very  soon 
capitulate.  Even  if  it  did  force  the  passage  of  the 
Adda  with  heavy  loss,  it  would  be  an  escaping 
army,  fleeing  back  to  a  far-off  base.      Its  prestige 

1  E.g.  the  vexilla  of  the  Seventh  legion  from  Pannonia  and  Eleventh  from 
Dalmatia  (ii.  1 1). 


102         THE  CAMPAIGN  OF  OTHO         ch... 

-  would  be  gone  and  Vitellius'  cause  ruined.  If  it 
chose  rather  to  force  the  passage  of  the  Po,  then 
Placentia  lay  like  a  lion  in  its  path,  and  the  army 
on  the  Adda  could  reach  and  notably  strengthen 
the  garrison  long  before  the  disheartened  column 
of  the   Vitellians    in    retreat   could   arrive    outside 

II  that  fortress.  The  enemy  at  Cremona  should  be 
"enveloped"  by  the  transferring  of  the  army  at 
Bedriacum  to  the  line  of  the  Adda,  while  the  Army 
of  the  Danube  marched  to  take  its  place  at 
Bedriacum  and  complete  the  investment. 

When  Otho  had  once  conceived  the  main  idea, 
three  questions  arose  in  connection  with  the  method 
of  its  execution  : — 

( i )  The  Route  of  the  Flank  March. — The  Emperor 
directed  his  army  to  march  by  the  northern  bank  of 
the  river.  At  some  safe  distance  from  Cremona, 
when  the  generals  thought  the  time  had  come,  the 
force  was  to  leave  the  Postumian  way,  and  strike 
off  to  their  right  flank,  to  circle  round  on  the  north 
side  of  the  town  of  Cremona,  and  so  to  come  down 
upon  its  intended  position  on  the  Adda.  This 
position  was  therefore  to  be  reached  by  a  flank 
march  in  the  immediate  proximity  of  the  enemy. 
The  dangers  of  this  route  seemed  obvious.  Why 
'  then  did  not  Otho  choose  the  route  on  the  southern 
side  of  the  Po  ?  This  would  have  been  absolutely 
safe  ;  and  the  army  would  have  crossed  by  Placentia 
and  turned  eastwards  thence  to  the  confluence. 

But  this,  the  safer  route,  could  not  be  chosen.  The 
march  from  Bedriacum,  in  that  event,  would  first  have 
been  an  apparent  retreat  to  cross  the  river  at  Hostilia. 


sec.  vii         AND  THE  VITELLIANS  103 

On  the  way  to  Hostilia  the  army  might  then 
have  met  the  forces  coming  up  from  Aquileia  to; 
take  its  place  at  Bedriacum.  The  confusion  which 
would  have  resulted  would  have  been  inextricable. 
There  was  but  a  single  road,  and  the  country  on 
both  sides  of  it  was  either  a  marsh  or  heavily  culti- 
vated. And  the  effect  of  the  apparent  retreat  on^ 
the  enemy  at  Cremona  had  to  be  considered.  If 
the  force  in  front  of  them  had  thus  disappeared, 
their  attention  must  at  once  have  been  excited. 
Had  they  guessed  Otho's  intended  manoeuvre  they 
would  have  had  abundance  of  time  to  occupy  the  line 
of  the  Adda  themselves,  and  safeguard  their  line  of 
communications  to  the  west.  The  whole  plan  of 
••> strategical  envelopment  would  fail  dismally.  Had 
they,  however,  as  was  far  more  probable,  pursued 
after  the  retreating  column,  the  feigned  retreat 
might  easily  have  become  a  real  one  ;  or  if  the  Army 
of  Bedriacum  safely  crossed  the  Po  at  Hostilia,  the 
Danube  army,  moving  up  piecemeal,  might  have 
met  the  foe  hotly  pursuing  the  others,  and  in  that 
case  would  have  been  rolled  up  in  disaster,  and 
flung  back  discomfited  upon  Aquileia.  Once  more 
the  plan  of  envelopment  would  be  ruined  ;  and  the 
Army  of  Italy  would  find  itself  between  a  victorious 
foe  to  the  east  of  it  and  Vitellius'  army  approaching 
on  the  west. 

But  if,  on  the  contrary,  the  force  at   Bedriacum 
marched  by  the  northern   route,  not  only  was  this 
less   than   half  the   distance,  but   its  effect  on  the 
enemy  would   surely  be  just   that  which  was  most  » 
desired.     The  foe  was  apparently  quiescent  in  his 


io4         THE  CAMPAIGN  OF  OTHO 

entrenched  lines  at  Cremona.  On  hearing  of  a 
forward  movement  on  part  of  the  force  at  Bedriacum, 
surely  they  would  be  tempted  to  cling  all  the  more 
closely  to  their  fortified  lines,  and  thus  give  the 
Othonians  exactly  the  opportunity  which  they  de- 
sired for  passing  unmolested  round  the  town  on 
the  Brescia  side.  Before  the  Vitellians  recovered 
from  their  surprise  the  flank  march  would  be  ended, 
and  the  invaders  be  trapped  at  Cremona.  There 
was  also  another  consideration  :  the  enemy  were 
trying  to  bridge  the  Po.  If  they  succeeded,  an  army 
marching  by  the  southern  route  might  be  assailed 
violently  on  the  flank  where  they  deemed  themselves 
safe.  This  would  be  fatal ;  for  a  flank  march  is 
horribly  dangerous,  chiefly  when  the  men  engaged 
upon  it  do  not  realise  the  nature  of  the  movement, 
and  any  sudden  appearance  of  the  foe  upon  their 
own  flank  is  therefore  utterly  unexpected. 

The  soldiers  in  the  marching  columns  always  assume  that 
their  commanders  suppose  the  enemy  straight  before  them.  If 
the  latter  suddenly  appears  on  the  flank,  the  men  may  easily 
imagine  that  they  are  surprised,  and  this  destroys  their  confidence. 
Flank  marches,  which  even  the  private  soldier  knows  to  be  such, 
are  easy  to  execute.  This  is  proved  by  the  numerous  marches 
within  the  investing  lines  before  Metz  and  Paris  in  1870,  for  the 
purpose  of  concentrating  troops  at  certain  points.  They  all  in 
their  nature  were  flank  marches  in  relation  to  the  enemy's  forces 
stationed  between  and  behind  the  works  of  the  fortress.  But 
here  the  whole  situation  was  clear,  for  every  soldier  knew  that 
during  the  march  they  could  only  be  attacked  from  the  side  of 
the  fortress,  and  the  feeling  of  being  placed  in  an  extraordinary 
position  disappeared.  The  troops  marched  quite  unconcernedly 
along  or  close  behind  the  line  of  investment.1 

1  Von  der  Goltz,  The  Nation  in  Arms,  p.  237. 


bbc. vii         AND  THE  VITELLIANS  105 

No  soldier  marching  from  Bedriacum  to  the 
Adda  by  the  northern  route  could  have  failed  to 
realise  the  situation.  On  the  southern  route  he— ~ 
might  have  been  unexpectedly  attacked  and  dis- 
mayed. Whereas,  if  the  northern  route  were  chosen, 
the  closer  the  attention  which  the  enemy  paid  to 
their  bridge-building,  the  better  the  chance  of  passing 
quietly  by  them  on  the  opposite  side  of  the  town. 
Cremona  was  to  be  the  Metz  of  the  campaign.  The 
Vitellians  were  the  French ;  the  Othonians  the 
Germans. 

The  northern  route,  then,  was  the  one  rightly 
chosen  by  Otho  for  the  flank  march  of  the  envelop- 
ing column. 

(2)  The  Command  for  Simultaneous  Movement. — 
But  why,  it  might  be  asked,  was  not  Suetonius' 
advice  accepted  that  Otho  should  at  least  wait  a' 
few  days  for  the  arrival  of  the  Danube  army  at 
Bedriacum  ?  Why  did  he  rely  rather  on  a  simul- 
taneous movement  of  both  armies — the  one  to  the 
Adda  from  Bedriacum,  the  other  to  Bedriacum 
from  Aquileia  ? 

The  answer  to  this,  again,  is  not  far  to  seek. 
Had  the  Othonians  made  no  movement  until  the. 
Danube  legions  had  arrived  at  Bedriacum,  the 
enemy  at  once  must  have  heard  of  the  arrival  of 
these,  and  have  been  anxiously  upon  their  guard. 
For  the  moment  they  seemed  lulled  in  inactivity  by ' 
a  false  sense  of  security.  It  was  this  false  sense  of 
security  of  which  Otho  could  so  brilliantly  avail 
himself.  Once  let  the  Danube  army  arrive  at 
Bedriacum,  any  forward  movement  after  its  arrival ' 


106         THE  CAMPAIGN  OF  OTHO         ch.i 

t 
J  would    find   the   foe  ./thoroughly   awakened    out   of 

sleep.     There  would  be  no  chance  of  envelopment 

by  a  flank  march  without  stubborn  fighting  front  to 

front.     The  strategical  opportunity  would  be  lost. 

The  simultaneous  movement  of  two  or  more  columns 

to  the  same  place  needs,  indeed,  the  most  careful 

timing  to  be  successful.     Yet  it  is  by  this  that  in 

modern  warfare  the  most  striking  triumphs  are  won, 

such   as   that   of  Koniggratz   in   1866.      More  and 

more  such  a  device  of  the  simultaneous  movement 

of  converging   columns    will    be    employed.       But 

Otho's   columns   had   not   so   hard  a  task,  as  their 

simultaneous  movement  was  directed  on  different 

places.     There  was  no  reason  why  it  should  not  be 

properly  carried  out  by  both  armies.     Certainly  the 

whole  idea  of  strategic  envelopment  depended  on 

this  simultaneous  movement. 

(3)  The  Position  of  the  Emperor. — Otho  himself 
neither  waited  for  the  arrival  of  his  Danube  army 
at  Bedriacum,  nor  did  he  put  himself  at  the  head 
of  the  column  of  march  for  the  Adda.  He  himself 
crossed  the  river  to  a  place  named  Brixellum,  the 
modern  little  town  of  Brescello,  which  lies  on  the 
southern  bank  of  the  Po,  about  midway  between 
Hostilia  to  the  east  and  Cremona  to  the  west. 
With  him  he  took  a  considerable  force  of  Guards, 
light-armed  troops,  and  cavalry.  The  Bedriacum 
column  was  to  be  led  by  the  Emperor's  brother 
Titianus  and  the  prefect  of  the  Guards,  Licinius 
Proculus,  with  Suetonius  and  Celsus  to  help  them. 

No  part  of  the  whole  strategical  plan  has  been 
more  misrepresented  and  misunderstood  than  Otho's 


SEC.  VII 


AND  THE  VITELLIANS  10: 


withdrawal    to    Brescello.      The    Roman    historian 
thoughtlessly  ascribes  it  to  the  Emperor's  care  for-" 
his  own  personal  safety.      His  troops  adored  him  ;^. 
he  had   endeared   himself  to  them  by  a  display  of 
sterling  military  qualities  on  the  march  from  Rome-p- 
he  was  presently  to  die  with  calm  courage.     And 
yet  Tacitus  believes  that  at  this,  the  very  crisis  of 
the  whole  campaign,  he  ran  like  a  coward.      More- 
over, at  a  time  when  his  Army  of  Bedriacum  was 
greatly  inferior  in  numbers  to  the  enemy,  and  was'^ 
about  to  make  a  flank  march  of  no  small  risk,  he,  the 
Emperor,  still  further  weakened  it  by  detaching  the 
troops  which  he  took  with  him  to  Brescello  to  serve 
as  his   own   personal   bodyguard.     "  That  day  was 
the  beginning  of  ruin  to  the  Othonian  cause,"  wrote 
the  historian  ;  "  the  spirit  of  the  troops  left  behind 
was  broken."  l     Like  the  Duke  of  Argyle  in  face  of 
Montrose,  Otho,  according  to  this  view,  found  his 
courage  fail  him,  and,  when  urged  by  his  staff  that  | 
his  life  was  more  valuable  than  his  presence,  was  | 
easily  persuaded  to  withdraw.     And  the  officers  and 
men  of  the   Guards   felt   his  desertion,  as  did   the 
knight  of  Ardenvohr  that  of  the  chief  of  his  clan  : — 

"  It  is  better  it  should  be  so,"  said  he  to  himself,  devouring 
his  own  emotion  ;  "  but — of  his  line  of  a  hundred  sires,  I  know 
not  one  who  would  have  retired  while  the  banner  of  Diarmid 
waved  in  the  wind  in  the  face  of  its  most  inveterate  foes  ! "  2 

Surely  Otho's  Roman  courage  and  his  Imperial 
position  might  well  take  the  place  even  of  a  hundred 
sires,  now  that  the  very  last  stake  was  to  be  played. 

i  Tac.  ii.  33. 
2  Cf.  Sir  W.  Scott's  Legend  of  Montrose^  chaps,  xviii.-xx. 


/ 


io8         THE  CAMPAIGN  OF  OTHO 

The  galley  on  Loch  Eil  saved  the  craven  chieftain's 
life  when  the  battle  was  lost.  No  galley  on  the  Po 
could  save  the  Emperor  from  the  last  consequences 
of  defeat. 

In  actual  fact,  it  was  not  Otho's  duty  to  lead  the 

y  flanking  column.  He  was  bound  to  take  up  such  a 
central  position  as  to  be  able  from  it  to  control  the 
development  of  all  parts  of  the  combined  scheme  of 
envelopment.  One  force  was  to  march  to  the 
Adda,  another  to  Bedriacum,  and  all  the  while  the 
defence  of  the  line  of  the  river  on  the  south  had  to 
be  maintained.  The  commander-in-chief  was  bound 
to  occupy  as  headquarters  a  place  where  he  could 
be  in  touch  with  all  his  separate  forces  which  were 
co-operating  to  secure  one  end.  The  commander- 
in-chief  in  such  a  case  is  not  allowed  to  take  the 
personal  command  of  one  of  those  separate  forces, 
not  even  of  that  exposed  to  the  greatest  risk  of 
contact  with  the  foe.  He  must  be  found  in  a 
situation  whence  he  has  a  grip  of  the  whole  de- 
velopment of  the  main  idea,  from  which  he  can,  if 

.  '  necessary,  send  troops  to  any  vital  and  threatened 
point  in  the  whole  area  of  events.  Such  a  central 
position  was  Brescello.  It  was  in  touch  with 
Bedriacum,    with    Hostilia,    and    with    the    whole 

^  southern  bank  of  the  Po  as  far  as  Placentia. 
From  it,  as  a  matter  of  fact,  Otho  could  send  a 
member  of  his  staff  to  take  Macer's  place  when  the 
latter's  resistance  to  the  bridge-building  at  Cremona 
was  proving  inadequate,  so  great  was  the  advantage 
of  the  presence  of  the  Emperor  at  the  central  posi- 
tion.    Because   Brescello  was   out   of  harm's   way, 


sec.  vii         AND  THE  VITELLIANS  109 

because  the  Emperor  took  no  share  in  the  fighting 
which  presently  fell  to  the  lot  of  the  Bedriacum 
column,  men  forgot  that  he  was  commander-in- 
chief  and  not  a  mere  general  of  division,  and 
foolishly  accused  him  of  cowardice.  Such  easy 
imputations  are  part  of  the  heavy  burden  of  supreme 
command.  "  The  magnitude  of  the  personal  re- 
sponsibility inseparable  from  command  against  the 
enemy  " l  is  surely  load  enough  for  the  general-in- 
chief  without  his  having  also  to  bear  such  charges, 
lightly  brought  and  readily  believed. 

The  more,  therefore,  that  clear-sightedness  and  intelligent 
direction  in  the  development  of  a  battle  is  demanded  of  a 
general,  the  greater  the  reason  that  he  should  keep  out  of 
serious  danger.  The  best  post  for  a  commander-in-chief  is  one 
from  which  he  has  a  clear  view  of  the  lines  of  advance  of  his 
columns  as  well  as  of  the  enemy's  line  of  battle.  Such  places 
are  usually  found  only  at  a  considerable  distance  completely 
beyond  the  range  of  fire ;  but  it  would  be  an  entirely  false  sense 
of  honour' to  reject  them  on  that  account.  By  displaying  his 
contempt  of  death,  a  commander-in-chief  can  scarcely  effect 
more  than  any  subordinate  officer;  but,  by  clearness  and  cool 
deliberation  in  his  plans,  he  will,  on  the  other  hand,  become 
the  benefactor  of  hundreds  of  thousands.2 

The  principle  applied  here  in  the  sphere  of 
tactics  is  true  in  that  of  strategy.  The  great 
scheme  of  envelopment  meant  the  moving  of  many 
pieces  together  on  the  strategical  chessboard.  The 
Emperor  must  be  the  player  and  set  them  all  in 
motion  from  Brescello,  not  be  himself  the  hardy 
knight  to  cry  checkmate  after  its  skilful  moves.     The 

1  Henderson,  Science  of  War,  p.  50. 
2  Von  der  Goltz,  The  Nation  in  Arms,  p.   133. 


) 


no         THE  CAMPAIGN  OF  OTHO         ch.i 

flank  march  of  the  one  column  had  to  be  entrusted 
/  to  his  divisional  commanders.     The  general  officer 
commanding  must  rely  on  the  sagacity  and  bravery 
of  his  subordinates.      If  they  prove  incompetent  he 
jjj  pays  the  penalty,  as  did  Lee  at  Gettysberg. 

A  rough  diagram,  then,  may  illustrate  the  plan 
of  strategical  envelopment  as  it  might  have  pre- 
sented itself  to  Otho's  mind,  when  the  investment 
was  completed  : — 


Ta^51"" 


<*?      vg      ;t...  \-  .,j»r-- 

*»..       V^. '      ••••  \o  fin''1'' 

:T9_Pquia_ 1;.V  , L,         *• RA=^  ■{o  t-'~ 


m 


\ 


p  Fo  Og  Po  *0       Po      h 

A  =  proposed  position  of  the  force  marching  from  Bedriacum  (B)  by  dotted  line 

and  based  on   Placentia  (P),   thus  cutting  the  communications 

of  the  Vitellians  at   Cremona  (C)  with   the  west,    via   either    Milan    or 
Pavia  (Ticinum). 
B  =  proposed  position  at  Bedriacum  of  the  force  from  Aquileia. 
C  =  the  Vitellians  "  enveloped  "  (as  a  result  of  the  flank  march)  at  Cremona. 
"0  =  Otho's  headquarters  at  Brescello. 
^>G  =  Macer's  gladiators    defending    the    river   against    the    bridge -building   at 

Cremona. 
PandH  =  the  only   passages   of  the  Po,   at    Placentia   (P),   and   Hostilia  (H)  ; 
both  in  Otho's  hands. 
The  road  to  Brescia  is  a  cul-de-sac  for  the  Vitellians,  as  the  passage  of  the 
Adda  farther  north   (via  Bergamo)  would  be  easily  controlled  by  the  foe 
at  A. 

To  achieve  the  envelopment  of  the  enemy,  Otho 
ordered  the  force  at  Bedriacum  to  advance  towards 
the  enemy  upon  its  flank  march  to  the  confluence 
of  the  Adda  and  the  Po. 

D.  The  Possibility  of  Success. — The  crucial  move- 
ment in  the  strategical  plan  of  envelopment,  upon 
which  its  whole  fortune  depended,  was  the  flank 
march   of  the    Bedriacum   army.      The  plan   itself 


sec.  vn         AND  THE  VITELLIANS  in 

was  a  brilliant  one.  For  modern  war  it  has  been 
declared  that  "  envelopment,  not  mere  weight  of 
numbers,  is  the  true  secret  of  decisive  success."1 
Such  was  the  strategy  by  which  the  Emperor  Otho 
planned  to  defeat  the  enemy,  rather  than  stay  idly 
on  the  defensive,  as  his  older  generals  recom- 
mended, or  make,  with  inferior  numbers,  a  frontal 
attack  on  the  foe.  But  were  not  the  dangers  of  the1 
flank  march  too  great  to  allow  success  ?  The  plan 
was  a  daring  one.  Was  it  not  also  a  rash  one, 
which  merited  failure  ?  Criticism  might  fasten  on 
two  points — on  the  enormous  risk  of  the  flank 
march  itself,  and  on  the  precarious  situation  of  the 
force  at  the  confluence  even  if  the  flank  march  was 
accomplished. 

Neither,  however,  of  these  dangers  was  such  as 
reasonably  to  deter  Otho  from  carrying  out  his 
scheme. 

(i)  The  Flank  March.  —  "These,"  writes  the 
German  expert  concerning  flank  marches,  "have 
the  reputation  of  being  difficult  and  dangerous 
undertakings.  Military  history,  however,  teaches 
us  that  in  the  matter  of  flank  marching  one  may 
venture  more  than  theory  jwould  seem  to  allow. 
Frederick  the  Great  at  Prague  made  a  flank  march 
round  the  right  wing  of  the  Austrians,  and  at  Kolin 
even  along  their  whole  front  .  .  .  even  in  manoeuvres 
flank  marches  are  successfully  executed  even  within 
sight  of  the  enemy."  2 

All  depended  on  the  immobility  of  the  enemy  and, 

1  Henderson,  Science  of  War,  p.  415. 

2  Von  der  Goltz,  op.  cit.  pp.  234,  235. 


ii2         THE  CAMPAIGN  OF  OTHO 

especially,  on  the  discretion  of  Otho's  generals. 
They  must  act  rapidly  and  with  decision,  and  must 
choose  the  fitting  place  for  striking  away  north- 
wards from  the  main  road  before  they  came  within 
sight  of  Cremona  or  in  touch  with  the  Vitellians  in 
the  town.  The  object  to  be  gained  by  the  march 
was  worth  many  risks. 

The  greatest  advantage  of  all  turning  movements  is  that,  if 
they  succeed,  they  finally  result  in  the  whole  of  the  enemy's 
army,  or  a  part  of  it,  being  caught  between  two  fires.  Scharn- 
horst  expressed  himself  to  the  effect  that  "  troops  attacked  upon 
more  than  one  side  may  be  regarded  as  defeated."  This  pro- 
nouncement is  not  true  unconditionally,  yet  it  is  founded  upon 
the  fact  that  he  who  finds  himself  between  several  enemies 
threatening  him  from  different  directions,  is  constrained  to 
eccentric  action  which  tends  to  split  up  his  forces  and  thus  to 
weaken  him,  while  the  former  work  concentrically  and  gain  in 
strength.1 

If  the  march  were  successful  the  foe  would  be  in 
the  trap,  and  if  they  escaped  at  all,  it  would  be  only 
to  retreat  with  loss  and  disgrace.  The  risks  must 
be  run.  "  He  who  would  always  in  war  be  on  the 
safe  side  will  hardly  ever  attain  his  object."  J 

(2)  The  Position  on  the  Adda. — Otho's  force, 
if  it  reached  the  confluence,  might  seem  to  be 
very  uncomfortably  placed  between  the  enemy  at 
Cremona  and  Vitellius'  approaching  reinforcements; 
but  this  in  reality  was  hardly  the  case.  Vitellius 
had  not  crossed  the  Alps,  and  the  crisis  of  the 
situation  must  come  in  a  very  few  days  after  its 
arrival  at  the  confluence.     Valens  and  Caecina  had 

1  Von  der  Goltz,  op.  cit.  p.  283. 
2  A  saying  of  Von  Moltke's  a  propos  of  the  campaign  of  1866. 


sRc.vn         AND  THE  VITELLIANS  113 

not  supplies  enough  to  stay  sulkily  in  Cremona,  nor 
indeed  would  their  troops  be  likely  tamely  to  submit 
to  this.  This  force,  then,  at  Cremona  was  the 
danger,  and  the  Adda  was  a  splendid  stream  for- 
Otho's  men  to  defend  even  against  a  much  larger 
army,  especially  when  their  comrades  from  the  old 
camp  would  be  pressing  upon  the  rear  of  the 
attacking  enemy.  The  strong  fort  which  is  to-day 
placed  on  both  banks  of  that  river  at  Pizzighettone, 
where  the  road  and  rail  for  Cremona  cross  the 
stream,  shows  the  value  still  placed  upon  the  Adda 
as  a  military  obstacle.  Here,  then,  the  force  would 
serenely  expect  the  attack  of  the  desperate  foe, 
even  though  it  had  temporarily  surrendered  its 
own  line  of  communications  with  the  east. 

It  never  occurred  to  any  one  in  the  German  army  at  the  time 
that  on  August  18,  1870,  we  were  fighting  a  great  battle  with 
reversed  front,  and  that,  in  our  outflanking  attack  upon  the 
French  right,  we  had  completely  cut  ourselves  off  from  our 
established  lines  of  communication.  All  attention  was  centred 
forwards  in  victory  and  not  backwards  in  retreat.1 

But  the  Othonians  at  the  confluence  would  have . 
been  more  happily  placed  in  the  event  of  a  reverse 
than  were  the  Germans  outside  Metz ;  for  the 
former  could  fall  back  on  Placentia,  and  so  regain 
safety  and  their  communications  once  more.  And 
they  were  Romans.  Otho's  strategical  plan  of 
envelopment  was  bold  in  conception,  and  needed 
energy  and  intelligence  in  execution.  But  its 
daring  and  possibility  merited  success  ;  and  the 
alternatives  to  it  of  quiescence  or  of  frontal  attack 

1  Von  der  Goltz,  op.  cit,  p.  355. 

I 


ii4         THE  CAMPAIGN  OF  OTHO         ch.j 

promised    very    little,     if    any,     triumph    for    the 
Emperor. 


§  8.    The  "  Battle  of  Bedriacum  " 

After  Otho  had  departed  from  the  camp  at 
Bedriacum,  the  generals  left  in  command  prepared 
to  carry  out  his  orders.  Part  of  the  force  under 
Gallus  was  kept  in  camp  to  guard  it,  and  to  await 
the  arrival  of  the  Danube  army.  The  rest  of  it 
marched    out    on    April     14,    along    the    road    to 

'  Cremona.  In  the  day's  march  they  covered  four- 
teen miles  and  halted  for  the  night.1  The  generals 
judged  it  safe  to  approach  within  eight  miles  of  the 
city  before  diverging  to  the  north.  The  troops 
were  marching  to  take  up  a  new  position  at  the 
confluence  of  the  Adda  and  Po,  and  entrench  them- 
selves there.  They  therefore  were  naturally  in  full 
marching  kit  and  accompanied  by  a  baggage  train. 
It  was  most  desirable  to  keep  to  the  broad,  paved 
way  as  long  as  possible.  Hence  the  generals  ven- 
tured along  it  as  far  as  fourteen  miles,  and  en- 
camped for  the  night.  Their  ultimate  objective, 
the  confluence,  lay  some  fifteen  miles  away  in  a 
straight  line.  The  next  day's  march  would,  how- 
ever,  have   to  be   a  longer  one  by  reason  of  the 

J  detour  round  Cremona. 

But  neither  generals  nor  troops  were  in  good 
spirits.  Even  in  April  the  sun  can  be  extremely 
hot  and  the  road  exceedingly  dusty  between  Calva- 

1   Reading   "ad  quartum  decimum "  for  "ad  quartum ''   in  ii.   39.      See 
Note  D,  "The  Distances  in  Tacitus  ii.  39,  40." 


sec.  viii         AND  THE  VITELLIANS  115 

tone  and  Cremona.     The  fourteen  miles  had  been 
fatiguing,  and  the  troops  had  been  distressed   for 
lack  of  water.     This  indeed  was  not  the  generals' 
fault,    unless   (which   seems   improbable)    they  had 
been     able     to    improvise    water  -  carts    and    had 
neglected  to  do  so.     For  although  in  the  flat  plain 
to   the   east  of  Cremona  there  are  to-day  ditches, 
innumerable,  yet  in  April  these  were  either  dry  or* 
contained  only  a  little  stagnant  filthy  water.1     Of 
rivers  there  were  none ;  for  every  step  along  the 
road   took  the    thirsty  troops    farther   and    farther 
from   the   Oglio,   and   their   camp  for  the  night  in 
the  neighbourhood  of  the  modern  hamlet  of  Pieve 
Delmona  lay   midway   between  the   Oglio   on   the 
north   and  the  Po  on  the  south,  and  some  six  or 
seven  miles  from  both.      In  the  immediate  presence! 
of  the  unsuspecting  enemy  the  men  could  not  bej 
allowed   to  straggle  in  search  of  water,  either  on) 
the  march  or  from  the  evening's  camp.      It  might 
indeed  have  been  better   if  the  generals   had   left 
the  main   road    earlier    and    encamped    beside   the 
Oglio   for  the   night.      But   the  attractions   of  the, 
highway  proved  too  strong.2 

The  soldiers  were  therefore  in  a  bad  temper  and    ■ 
angry  with  their  generals.      In  their  discontent  and 

1  This  at  least  was  the  case  between  Calvatone  and  Piadena  in  April 
1906. 

2  Tacitus  ii.  39  :  "  Adeo  imperite  ut  quamquam  verno  tempore  anni  et  tot 
(sic)  circum  amnibus  penuria  aquae  fatigarentur. "  This,  of  course,  is  mere 
nonsense  if  the  march  were  only  four  miles.  If  fourteen,  still  it  is  the  private 
soldier's  view  (as  always)  which  puts  his  sufferings  down  to  lack  of  skill  on 
part  of  his  generals.  Tacitus  never  realises  a  military  situation.  He  does 
not  even  tell  us  that  the  march  lasted  two  days,  although  we  should  have 
inferred  it  from  probability,  had  not  Plutarch  directly  told  us.  For  the 
impossibility  of  reconciling  Plutarch"s  whole  story  with  Tacitus,  cf.  Note  D. 


n6         THE  CAMPAIGN  OF  OTHO         ch.i 

impatience  they  loudly  lamented  the  Emperor's 
absence.  The  generals  meanwhile  were  fiercely 
quarrelling  among  themselves.  Suetonius  and 
Celsus  disliked  and  distrusted  the  whole  scheme 
from  the  beginning.  They  now  gloomily  pointed 
(out  its  risks  to  Titianus  and  Proculus,  who  were, 
/for  their  part,  eager  and  ready  to  carry  out  Otho's 
■  orders.  The  foe,  urged  the  malcontents,  were  all 
but  in  sight.  In  case  of  attack  these  had  but  four 
miles  to  tramp  (a  characteristic  underestimate). 
But  their  own  troops  were  in  marching  order,  not 
fighting  trim,  and  wearied  by  the  march.  These 
j  recriminations  and  gloomy  reflections  came  too 
late,  and  were  indeed  out  of  place.  The  troops 
could  not  but  mark  the  acrimonious  dissensions 
-between  their  leaders,  and  these  must  have  the 
worst  effect  upon  them,  especially  in  their  present 
temper.  There  was  no  doubt  that  they  had  come 
too  far  along  the  road,  too  near  the  enemy,  for 
safety.  And  now  the  generals  were  busy  discuss- 
ing again  what  had  already  been  decided.  It  was 
Ih  grand  error  on  Otho's  part  to  entrust  the  column 
/to  a  committee  of  generals  in  place  of  one  supreme 
commander.  Roman  generals  did  not  always  agree 
together.  Two  were  bad  enough,  but  a  council  of 
four  was  indeed  likely  to  ruin  any  plan.  The 
Emperor  sought  to  remedy  the  evil  by  his  own 
control.  He  despatched  a  Numidian  mounted 
orderly  from  Brescello  with  the  stern  and  impera- 
tive order  to  the  generals  to  advance.  It  may 
be  that  they  misread  the  order,  *and  thought  that 
it  countermanded  the  original  plan  in  favour  of  a 


sec.  viii         AND  THE  VITELLIANS  117 

direct  attack  upon  the  Vitellians  at  Cremona.  -£)r 
perhaps  Otho  himself,  hearing  that  the  force  had 
come  so  near  to  the  enemy,  judged  that  there  was 
no  room  for  the  flank  march  left,  and  himself  com- 
manded a  frontal  attack  instead.  Or,  again,  the 
generals  may  have  relied  on  the  enemy's  inactivity 
and  still  moved  forward,  intending  to  strike  north 
presently,  allured  by  the  fatal  attractions  of  the 
highway.  Whose  the  blunder  was  can  never  nowr 
be  determined.  All  that  is  certain  is  that  Otho's  Uv 
whole  strategical  scheme  miscarried  ;  for  when  the 
column  resumed  its  march,  obedient  to  orders,  on 
April  15,  they  blundered  straight  upon  the  foe. 
The  head  of  the  column  suddenly  found  the 
enemy's  horse  charging  full  upon  them. 

Valens  had  not  been  caught  unready  that*v. 
morning.  Under  screen  of  his  cavalry  charge  he 
marched  his  army  out  of  camp  and  drew  it  up 
ready  for  battle.  Caecina  and  his  men  were  quickly 
summoned  from  the  half- made  bridge.  The  full 
Vitellian  army  stood  ready  to  fight,  drawn  up 
quietly  despite  the  near  approach  of  the  foe. 
Thick  brushwood  on  either  side  of  the  road  hid  the 
Othonian  approach,  and  in  consequence  the  Vitellian 
regiments  moved  to  their  allotted  places  without 
alarm  or  disorder.  Their  cavalry  indeed  came 
presently  reeling  back,  for  the  head  of  the  Othonian 
column  stood  its  ground  valiantly  and  repulsed 
them.  It  needed  the  levelled  pikes  and  the  taunts 
of  the  First  infantry  legion  of  the  Vitellians  toj 
compel  the  shaken  horse  to  pull  bridle  and  rally./ 
Then  the   whole  army  moved  forward  on  a  wide 


n8         THE  CAMPAIGN  OF  OTHO         ch.  i 

I  front  stretching  some  distance  on  either  side  of  the 

'  road.     The    repulse    of  the    enemy's    cavalry   had 

j  given  the  generals  on  the   Othonian  side  time   to 

extend    their    front,   and    dress    it    to    meet    their 

opponents   to   some   degree.      But   their  confusion 

was   still  great.      Some   indeed   believed   that   the 

advancing  foe  had  abandoned  Vitellius'  cause  and 

[were  joining  them  in  all   love   and  amity.     Some 

pressed    boldly  on    to    the    front   seeking   honour ; 

some   hurried  as  eagerly  to  the  rear  in  search   of 

safety.      There   was   more   uproar  than   there  was 

discipline.      At   the    height    of   the    confusion    the 

.Vitellian  line  charged. 

But  the  fighting  was  stubborn.  Between  the 
river  and  the  road  two  legions  strove  fiercely. 
The  Othonian  First  Adjutrix,  eager  to  gain  its  first 
laurels  (for  it  had  been  but  recently  levied),  rushed 
fiercely  upon  the  Vitellian  Twenty- first,  a  legion 
of  old  renown,  overthrew  its  first  ranks,  and  carried 
off  its  eagle  in  triumph.  In  bitter  anger  the 
veterans  rallied  and  thrust  hard  upon  the  foe. 
The  legate  of  the  First  fell,  his  men  were  routed, 
and  the  loss  of  the  eagle  was  made  good  by  the 
v  capture  of  many  colours  from  the  enemy.  On  the 
other  flank,  the  men  of  the  Fifth  legion  of  the 
German  army  drove  the  Pannonian  Thirteenth 
legion  in  flight  off  the  field.  The  detachment  of 
the  Fourteenth  legion,  the  famous  legion  of  Britain, 
stood  true  to  their  absent  comrades  and  the  tradi- 
tions of  the  regiment.  But  they,  a  mere  handful, 
could  not  save  the  day  for  Otho,  and  they  fought 
vainly   but   desperately,    surrounded   by   a   ring    of 


sec.  via         AND  THE  VITELLIANS  119 

foes.  It  was,  like  Inkermann,  a  soldiers'  battle. 
Otho's  generals  had  already  done  their  utmost  to~ 
ruin  his  cause  by  their  quarrels.  They  had  igno- 
rantly  exposed  their  army,  unprepared  and  in  dis- 
order, to  the  frontal  attack  of  a  more  numerous 
and  well-ordered  veteran  force.  One  thing  only 
was  lacking.  They  fled  from  the  field.  But  their 
men  went  on  fighting.  From  the  southern  bank 
of  the  Po,  the  gladiators  crossed  the  river  in  boats 
to  help  their  comrades.  Then  the  Vitellians  made 
their  last  supreme  effort.  Valens  and  Caecina  flung 
their  reserves  into  the  battle-line.  The  enemy's 
centre  was  pierced.  The  Batavian  auxiliaries  of 
the  German  army  cut  the  gladiators  to  pieces  even 
before  they  reached  dry  land,  and,  hastening  in  the 
flush  of  victory,  came  charging  upon  the  left  flank 
of  the  stubborn  foe.  This  flank  charge  decided 
the  issue."  The  Othonians  broke  and  fled  wildly. 
The  battle  was  ended. 

The  pursuit  rolled  on  for  many  miles.  No 
quarter  was  given,  for,  says  the  Roman  historian 
grimly,  "  captives  in  civil  wars  cannot  be  turned 
to  profit."  The  ways  were  heaped  high  with  the 
bodies  of  the  slain.  The  survivors  of  the  rout 
found  refuge  only  with  Gallus  and  the  camp  at 
Bedriacum,  twenty  miles  away.  The  Vitellians 
checked  their  pursuit  four  miles  from  the  camp, 
and  bivouacked  for  the  night  just  west  of  the 
modern  town  of  Piadena.  Next  day,  April  16, 
they  advanced  to  Bedriacum,  and  the  garrison 
surrendered.1 

1  Tac.  ii.  39-45- 


120         THE  CAMPAIGN  OF  OTHO         ch.  i 

Thanks  to  the  mistakes  of  the  enemy,  Vitellius' 
generals  had  won  for  their  master  the  final  victory, 
and  with  it  the  Empire.  He  himself  visited  the 
scene  of  the  struggle  on  the  24th  of  May,  six 
weeks  after  it  had  been  fought.  No  attempt 
had  been  made  to  bury  the  dead.  Caecina  and 
Valens  showed  their  Emperor  over  the  battle-field 
and  explained  to  him  the  details  of  that  bitter  fight. 
His  one  saying  is  recorded  :  — 

When  hee  came  into  the  fields  where  the  battaile  was 
fought,  and  some  of  his  traine  loathed  and  abhorred  the 
putrified  corruption  of  the  dead  bodies,  he  stuck  not  to  harten 
and  encourage  them  with  this  cursed  speech :  that  an  Enemie 
slaine  had  a  very  good  smell,  but  a  Citizen  farre  better.  How- 
beit  to  qualifie  and  allay  the  strong  savour  and  sent  that  they 
cast,  hee  poured  downe  his  throat  before  them  all  exceeding 
great  store  of  strong  wine,  and  dealt  the  same  plentifully 
about.1 

Vitellius  is  the  one  utterly  contemptible  figure 
of  the  century. 

Thus  the  first  "  Battle  of  Bedriacum " 2  was 
fought,  and  Vitellius  won  his  throw  for  Empire. 
A  review  of  the  military  measures  of  both  sides 
shows  that  both  made  mistakes,  or  at  least  failed 
in  their  intentions.  Neither  the  plan  of  tactical 
penetration  on  the  one  side,  nor  that  of  stra- 
tegical envelopment  on  the  other,  was  fully  carried 
out.     As  Otho's  had  been  the  more  brilliant  and 

1  Suetonius,  Vitellius,  10.  Philemon  Holland's  translation  (a.d.  1606). 
Cf.  Tac.  ii.  70. 

2  The  title  "  Battle  of  Bedriacum  "  is  a  misnomer,  as  the  actual  fighting 
took  place  just  outside  Cremona,  twenty  miles  away,  and  that  of  the  "  Battle 
of  Cremona  "  would  be  more  fitting.  But  the  former  name  is  consecrate  by 
custom. 


VITELLIUS. 

From  a  bust  in  Vienna. 


SEC.  VIII 


AND  THE  VITELLIANS  121 


daring  conception,  so  its  failure,  which  precipitated 
the  final  fight,  was  the  more  ruinous,  and  brought 
defeat  and  death  upon  him. 

The  great  cause  of  its  failure  was  the  incompetence^ 
of  Otho's  generals.  As  the  story  of  the  battle  shows, 
they  clung  to  the  main  road  too  long,  calculating 
too  confidently  upon  the  enemy's  immobility.  But 
the  enemy  gladly  sallied  out  to  attack,  and  Otho's 
troops  were  caught  encumbered  with  baggage  and 
tools,  unready  for  a  fight  and  not  expecting  it. 
To  incompetence  the  generals  added  cowardice,  of 
which  later  they  blandly  made  a  merit,  and  won 
Vitellius'  pardon  by  this  means.  \]  There  was  small 
wonder  that  with  such  generals  Otho  lost  the  day. 
His  troops  fought  well  for  him  against  all  possible 
odds.  It  had  been  better  for  Suetonius  had  he 
died  eight  years  earlier  amid  Boadicea's  war- 
chariots.  ■  He  saved  Britain.  But  now  he  had 
lost  himself. 

Otho's  generals  failed  him,  as  Lee's  subordinates 
failed  to  carry  out  their  orders  in  the  Gettysberg 
campaign.  Decisive  defeat  was  the  result  in  both 
cases. 

But  the  whole  campaign  has  been  misunder- 
stood by  the  Roman  historian.  Otho's  strategy 
was  hidden  from  him.  The  reason  of  this  blind- 
ness on  Tacitus'  part  is  easily  to  be  found.  The 
Emperor  could  not  explain  his  strategy  to  his 
troops  lest  the  enemy  should  hear  of  it  before- 
hand. "  The  one  fixed  law  of  all  military 
experience  is  that  whatever  is  believed  in  one's 
own    camp    is    believed    also    to    be    true    in    that 


122         THE   CAMPAIGN  OF  OTHO 

of  the  enemy."1  Very  little  happened  in  the  camp 
at  Bedriacum  without  the  foe  being  at  once  in- 
formed of  it.  Civil  war  produces  a  rich  crop  of 
traitors  on  both  sides — above  all  when  the  com- 
batants are  fighting  on  behalf  of  persons  and  not 

s  on  behalf  of  principles.  On  the  very  morning  of 
April  15  itself,  two  tribunes  of  the  Praetorian 
Guard  sought  for,  and  were  "granted,  an  interview 
with  Caecina,  which  was  interrupted  only  when 
the  battle  called  the  general  away  in  haste.2  Otho 
therefore,  having  planned  to  envelop  the  enemy  by 
a  flank  march,  was  bound  to  seek  to  deceive  them 
as    to   his    intentions.       If  they    misread    his   own 

,  departure  to   Brescello  as  a  sign  of  cowardice,  so 

I  much  the  better.  They  would  be  all  the  less  on 
their  guard.  But  by  misleading  the  enemy,  the 
Emperor  misled  also  the  common  soldier  in  his 
own  camp,  and  in  his  train  he  misled  the  most 
unmilitary  of  historians.  The  soldier  saw  that 
Otho's  orders  to  advance  had  led  to  the  battle. 
He    jumped    to    the    conclusion    that     Otho    had 

1  intended  that  battle  from  the  first.  Tacitus 
solemnly  repeats  his  view.  The  soldier  regretted 
his  Emperor's  absence  and  wondered  at  it.  The 
historian  explains  it  with  great  satisfaction  as  due 
to  cowardice.  The  soldier  found  the  battle  badly 
mismanaged,  and  heavy  defeat  the  result.  The 
historian  put  it  all  down  to  a  foolish  order  to 
advance  for  a  frontal  attack.  True,  the  latter  had 
discovered   in   his   records  or  inquiries   some   faint 

1  Maj.-Gen.  Maurice,  Diary  of  Sir  John  Moore,  ii.  p.  354. 
-  II.  41  j  cf.  ii.  34. 


sEc.vn,         AND  THE  VITELLIANS  123 

traces  of  an  idea  of  reaching  the  confluence.  But 
as  the  direct  way  to  it  from  Bedriacum  lay  straight 
through  Cremona,  he  concluded  that  the  troops 
were  bound  to  march  that  way  and  hence  would 
have  to  fight.  The  troops  notoriously  did  march 
that  way.  The  proof  was  complete.  Otho  was 
reckless,  impatient,  foolish,  a  coward !  He  had 
made  no  mistake  up  to  the  time  when  he  issued 
his  last  orders.  This  was  all  the  more  reason  for 
an  accumulation  of  errors  in  them.  His  troops, 
adored  him  after  the  defeat  as  before.  It  was 
curious  ;  but  what  will  not  defeated  troops  do  ?  He 
died  with  unshaken  serenity.  Any  coward  can  do 
that !  He  would  not,  as  he  might  well  have  done,, 
prolong  the  war,  falling  back  on  his  Danube  army,' 
where  the  line  of  safe  retreat  was  open  to  him  \ 
he  would  not  challenge  Fortune's  verdict  upon 
second  field.  He  would  redeem  Italy  from  war'i 
horrors  by  the  willing  sacrifice  of  his  own  life] 
This  was  conduct  truly  worthy  of  a  coward.  The 
common  soldier  was  too  ignorant  to  see  fully  his 
general's  incapacity  and  shrinking.  The  historian 
took  from  him  the  tales  of  what  befel  and  of  what 
was  said,  and  wisely  added  the  explanations. 

That  the  military  knowledge  of  the  common 
soldier,  with  all  its  hopeless  limitations,  should 
become  the  wisdom  of  the  journalist  is  a  feature 
of  historical  writing  but  too  familiar  to  us  of  late 
years.  Otho  sought,  as  it  were,  to  make  a  Metz 
of  Cremona.  Had  his  strategical  idea  succeeded, 
Tacitus  might  have  realised  its  meaning  if  not  its 
brilliance.      It   failed,   and  in   consequence  left   but 


i24         THE  CAMPAIGN  OF  OTHO 

one  puzzling  trace  of  itself  in  the  historian's 
narrative,  when,  that  is,  he  speaks  of  the  inten- 
tion to  reach  the  confluence  of  the  Adda  and 
the  Po.  But  the  historian  does  not  see  the 
meaning  of  this,  and  gives  us  in  consequence 
a  story  of  the  whole  campaign  which  is  indeed 
"  unintelligible  from  a  military  point  of  view." ' 
Even  had  the  Germans  failed  in  their  attempt  to 
invest  Metz — and  they  too  came  near  failing — their 
effort  would  not  have  been  caricatured.  To  that 
extent,  at  least,  military  science  has  advanced  since 
the  days  of  Otho,  and  left  its  mark  even  upon  the 
intelligence  of  the  historian. " 


§  9.    The  Death  of  Otho 

When  the  tidings  of  defeat  reached  the  Emperor 

\  at    Brescello,    his    troops    there    implored    him    to 

continue  the  struggle.     The  legions  from  Moesia 

were    hastening    to    the    front    and    hard    at  hand. 

I  Had  Otho  willed  to  live,  he  might  yet  have  been 

the  victor. 

But  he  refused  to  be  cause  of  bloodshed  any 

longer.     Though  his  men  were  eager  to  fight — if 

need  were,  to  die — for  him,  he  would  not  suffer  it. 

The  wife,  the  children,  the  brother  of  his  triumphant 

1  rival  were  in  his  power.     He  would  take  no  venge- 

)  ance  upon  them.      He  blamed  none,   neither   men 

1  Mommsen. 

2  The  material  upon  which  is  based  the  view  of  the  campaign  contained 
in  this  section,  and  a  discussion  of  the  difficulties  of  the  Tacitean  story,  are  to 
be  found  in  Note  D.  I  have  judged  it  best  to  give  my  conclusions  in  the 
form  of  a  direct  narrative. 


sec. ix  AND  THE  VITELLIANS  125 

nor  gods,  for  the  calamity  which  had  befallen.^ 
"  Such  blame  rather  befitted  him  who  still  longed 
to  live."  Otho  had  no  such  longing.  He  had 
played  gallantly  for  Empire;  he  had  staked  his; 
fortune  on  a  throw  and  lost.  He  himself  would 
pay  the  forfeit,  but  no  other  one  besides.  The 
miseries  of  civil  war  had  lasted  long  enough,  and 
he  would  not  prolong  them.  If  only  his  own  life 
stood  now  in  the  way  of  amity  and  peace,  the  way 
should  speedily  be  made  open. 

Intrepid  in  his  looks,  courteous  in  his  entreaties, 
he  now  besought,  now  commanded,  his  officers  to 
hasten  to  make  their  peace  with  the  victor,  and 
himself  rebuked  the  wrath  of  the  troops  with  those 
who  obeyed  and  hurried  from  the  camp.  Nor 
would  he  rest  until  he  knew  that  all  had  fled.  His 
young  nephew  at  his  side  was  panic-stricken.  He 
cheered  the  boy  and  bade  him  hope  for  the  new 
Emperor's  clemency.  "  Be  brave,"  he  said,  "  and 
grasp  life  sturdily ;  remember  that  Otho  was  your 
uncle,  yet  remember  it  not  overmuch."  He  sought 
out  and  destroyed  all  letters  in  which  were  written 
any  words  of  love  for  him,  of  hatred  for  Vitellius. 
In  such-wise  and  in  leave-taking  of  his  friends 
Otho's  last  day  drew  to  evening.  Then,  when 
darkness  fell,  he  quenched  his  thirst  with  a  little 
cold  water  and  lay  down  in  his  tent  quietly  to 
sleep. 

The  light  of  dawn  woke  him,  and  he  called 
to  his  freedman  in  the  tent.  Had  his  friends,  he 
asked,  who  had  left  him  the  day  before,  fared  well 
upon  their  going?     His  servant  answered  that  no 


126         THE  CAMPAIGN  OF  OTHO         ch.i 

ill  had  befallen  any.  "  Go  then,"  said  the  Emperor, 
"  and  show  yourself  now  to  the  troops,  lest  thou 
die  miserably  at  their  hands  as  having  brought 
death  upon  me."  The  man  went  out.  Then  Otho 
drew   from    under   his    pillow   a   dagger  which    he 

\  had  chosen  and  hidden   there  the  evening  before, 

;  and,  turning,  fell  upon  it.  Hearing  a  single  groan, 
his  slaves,  still  faithful,  and  Firmus,  his  loyal  pre- 
fect of  the  Guards,  rushed  into  the  tent  and  found 
their  Emperor  dead.  Death  had  come  quickly  to 
him,  nor  did  he  die  with  any  Stoic  pose.  "  It  is 
the  coward  who  talks  much  about  his  death,"  he 
had  yesterday  told  his  soldiers.  He  might  have 
added,  "  and  the  philosopher."  Otho  was  a  soldier, 
\\   and  spent  not  many  words  on  death. 

His  troops  carried  his  body  to  the  pyre,  weeping, 

1  kissing,  now  his  hands,  now  his  wounded  breast. 
The  flame  was  kindled  beneath  the  funeral  pile, 
and,  as  it  blazed  upwards,  some  of  his  men  slew 
themselves  beside  it  for  very  rivalry  of  honour  and 
of  sorrow  for  their  Prince.  Others  too,  when  they 
heard  of  it,  in  the  camps  at  Bedriacum,  Placentia, 
and  elsewhere,  did  the  like.     The  officers  might  flee. 

I  The  men  followed  their  Emperor  through  the  gate  of 
death.  "  They  had  received  nothing  of  great  price 
at  the  dead  man's  hands,  nor  did  they  think  to 
suffer  any  dread  doom  at  his  conqueror's.  But  in 
no  tyrant  ever,  it  seemeth,  in  no  monarch  hath 
there  ever  been  begotten  so  terrible,  yea  so  mad, 
a  lust  for  rule  as  was  their  lust  to  yield  obedience 
and  submit  themselves  to  Otho's  governance.  Verily 
that  fierce  longing  left  them  not,  no  not  though  he 


sec.  ix  AND  THE  VITELLIANS  127 

was  dead,  but  it  abode  and  passed  in  the  ending  of 
it  into  loathing  unquenchable  for  Vitellius."1 

It  is  hard  calmly  to  appraise,  dispassionately  to 
measure  out,  such  love.      It  is  hard  to  pass  judg- 
ment  of  indifference    or    disparagement    upon    the 
Emperor    who    inspired    it.       Otho    was    careless,  11 
licentious,   ambitious,   frankly   selfish,   treacherous  ;'J. 
but  he  died  like  a  true  Roman  when  all  was  done. 
Fate  gave  him  but  thirty-seven  years  of  life  and  a 
bare  three  months  of  Empire.      He  bade  farewell 
to  both  unmoved,  as  one  who  goes  a  journey  and 
will  presently  return.     Like  Petronius,  he  is  scornful  | 
of  life  with  a  quiet  contempt  born  of  native  courage./ 
For  him,  too,  the  thought  of  death,  and  of  the  loss 
by  death  of  those  good  things  of  life  which  he  has 
enjoyed  to  the  full,  cannot  cast   a   shadow  on   his 
peace  when  the  last  call  sounds.       Like  Antony,  as 
he  answers  to  that  call,  he  makes  one  claim  to  an 
immortality  of  renown,   if  there  be  any  such  ;   for 
he,  too,  has  won  the  love- -of  his  men  in  surpassing 
measure.     And   to  have  won  such  love   is   not   to 
have  failed  utterly  in  life  after  all. 

1  Plutarch. 


CHAPTER    II 


THE    FLAVIAN    INVASION    OF    ITALY 

How  oft,  indeed, 
We've  sent  our  souls  out  from  the  rigid  north 

To  climb  the  Alpine  passes  and  look  forth, 

Where  booming  low  the  Lombard  rivers  lead 
To  gardens,  vineyards,  all  a  dream  is  worth. 

E.  B.  Browning  :   Casa  Guidi  Windows 


§  I.    Vit  el  litis  and  his  Army  in  Rome 

The  first  "  Battle  of  Bedriacum "  was  fought  on 
April  15,  a.d.  69,  and  Otho  slew  himself  next 
morning.  The  news  of  the  victory  reached  his 
rival  Vitellius  at  Lugdunum,  where  he  was  met  by 
his  victorious  generals  Caecina  and  Valens  as  well 
as  by  the  fugitive  leaders  of  the  defeated  army, 
Suetonius  and  Proculus.  The  former  were  suitably 
honoured  by  the  new  Emperor ;  the  latter,  when 
they  pleaded  that  their  own  treachery  to  Otho  had 
lost  him  the  battle,  were  acquitted  of  the  charge  of 
honour  and  received  pardon.  From  Lugdunum 
Vitellius  went  on  his  way  slowly  to  Rome,  escorted 
by  his  generals,  who  showed  him  the  battle-field1 
and    entertained    him    with   gladiatorial    shows    at 

1  On  May  24. 
128 


sec.  i    FLAVIAN   INVASION  OF  ITALY    129 

Cremona  and  Bononia.  To  his  large  and  trium- 
phant army  which  accompanied  him  was  given  on 
the  march  every  license  of  plunder  and  debauchery, 
and  it  did  not  hesitate  to  follow  the  example  set  by 
its  Emperor.1  News  speedily  reached  Vitellius  that 
the  legions  of  the  East,  under  Mucianus  in  Syria 
and  Vespasian  in  Judaea,  had  accepted  the  fact  of 
his  victory  and  recognised  him  as  Emperor.  His 
last  anxiety,  therefore,  was  allayed,  and  he  gladly 
abandoned  himself  and  his  army  to  the  full  enjoy- 
ment of  the  sweets  of  power.  His  mercy  and 
his  cruelty  were  alike  capricious.  Otho's  brother, 
Salvius  Titianus,  was  pardoned.  Galerius  Trachalus, 
the  orator,  who  was  suspected  of  writing  Otho's 
spirited  harangues  for  him,  happily  enjoyed  the 
protection  of  his  relative  Galeria,  Vitellius'  second 
wife.2  But  some  of  the  centurions  of  the  enemy's 
army,  whose  crime  was  that  of  military  loyalty  to 
their  dead  Prince,  were  executed  in  cold  blood. 
They  had  not  the  wit  of  their  generals  to  plead 
treachery  as  their  reasonable  apology.  Thus  dis- 
pensing favours  to  some  and  punishments  to  others, 
and  always  chiefly  intent  on  the  pleasures  of  the 
appetite,  the  glutton  Emperor  made  his  slothful 
progress  to  Rome.  He  entered  the  city  in  great 
state  at  the  head  of  sixty  thousand  troops  and  a 
larger  rabble  of  camp-followers.  The  troops  spread 
themselves  over  the  city,  lodging  where  they  liked 
and    doing   what   mischief  they  pleased.     All    dis- 


1  Tac.  ii.  56,  71. 

-  His  first  wife  was  Petronia,  now  divorced  and   married  to  Dolabella — 
who  was  presently  slain  for  his  temerity,  ii.  64. 

K 


i3o    FLAVIAN   INVASION  OF  ITALY   ch.ii 

cipline  was  at  an  end.  The  officers  had  no  control 
over  the  men,  the  men  none  over  their  appetites. 
The  torrid  heat  of  the  Roman  summer,  the  un- 
healthiness  of  the  city,  the  self-indulgence  of  the 
troops,  completed  a  demoralisation  begun  by  victory 
and  plunder.  Many  of  the  men  encamped  on  the 
right  bank  of  the  Tiber,  upon  the  low-lying  plain 
of  the  "  Vatican."  This  flat  land,  now  occupied  by 
the  crowded  "  Leonine  City,"  St.  Peter's, "and  the 
Papal  palace,  has  always  been  notoriously  un- 
healthy ;  and  then,  when  the  troops  new  come  from 
the  cold  north  hastened  without  self-restraint  to 
quench  their  raging  thirst  with  the  foul,  polluted 
river  water,  disease  took  an  ominous  toll  of  life.1 
Even  regimental  esprit  de  corps  was  suffering  ;  for 

*  Vitellius,  having  disbanded  all  Otho's  Praetorian 
Guards,  set  to  work  to  enrol  twenty  new  regiments 
of  Guards  (sixteen  Praetorian  cohorts  and  four 
Urban,  each  a  thousand  strong).  The  men  were 
chosen  at  haphazard,  with  scant  regard  to  their 
merit  or  their  services,  and  as  a  result  the  legions 
were    depleted,    but    no    really   efficient    corps    of 

y  Guards  was  created  to  compensate  for  this.2  Such 
thoughtless  army  reorganisation  did  but  corrupt 
and  spoil  a  fine  force  in  its  attempt  to  remedy 
an  existing  deficiency.  Recruiting  also  for  the 
legions  was  stopped,  with  intent  to  save  money, 
and  many  of  the  troops  were  invited  to  accept  their 
discharge  from  the  ranks.     The   Gallic  auxiliaries 

1  I  see  no  reason  to  suppose  that  this  disease  was  malaria,  as  suggested  by 
Mr.  W.  H.  Jones  in  his  essay  on  "Malaria."  This  disease  is  not  rapidly 
fatal,  even  to  northerners 

2  Tac.  ii.  93,  94. 


sec.  i    FLAVIAN   INVASION  OF  ITALY     131 

were  sent  off  home,  and  the  unruly  Batavian  cohorts 
despatched  to  Germany,  there  soon  to  kindle  savage 
rebellion.  Death  and  folly  played  havoc  with  the 
splendid  Army  of  Germany,  and  at  the  end  of  six 
months'  loose  living  in  Rome  it  seemed  to  be 
going  to  rack  and  ruin.1 

Meanwhile  the  Emperor  played  at  "constitu- 
tional government,"  and  devoted  his  more  serious 
thoughts  to  problems  of  the  palate.  When  he  was 
at  Lugdunum,  it  was  said,  men  heard  the  roads 
which  led  to  the  city  ringing  with  the  hurrying  feet 
of  those  who  came  carrying  the  dainties  of  all  lands 
to  whet  his  appetite,  his  "  foul  insatiable  maw." 2 
But  the  capital  offered  him  nobler  opportunities  of 
delicacies,  and  during  his  few  months'  stay  in  Rome 
he  is  said  to  have  spent  nine  hundred  millions  of 
sesterces.3  He  had  at  least  the  merit  of  a  con- 
sistency of  taste,  whether  the  object  of  his  ex- 
travagance was  large  or  small.  Nero  had  built 
a  palace  for  his  soul's  delight,  famous  and  hated  as 
the  "  Golden  House."  For  it  he  had  clothed  the 
squalid  slopes  and  dusty  purlieus  of  the  Esquiline 
with  woodland  glades  and  garden  greenery,  re- 
freshed them  with  cool  waters  and  with  quiet  shade, 
and  made  the  arid  desert  of  Rome's  hovels  blossom 
as  the  rose.4  Otho,  the  "  second  Nero,"  had  added 
to  its  beauties  and  extent.  Vitellius  complained 
at  it  :  he  felt  himself  cramped  by  such  a  meagre 
habitation.       But    if    he    himself   could    not    roam 

1  Tac.  ii.  67,  69. 

2  "  Epularum  foeda  et  inexplebilis  libido,"  ii.  62. 

3  Over  ,£7,000,000. 

4  See  my  Life  of  Nero,  pp.  243-247. 


132    FLAVIAN   INVASION  OF  ITALY   ch.h 

afield  as  widely  as  a  fitting  pleasaunce  might 
have  suffered  him,  no  such  limits  could  fetter 
the  activity  of  his  mind's  intelligence.  A  new 
recipe  for  hotch-potch  was  the  child  of  that 
intelligence,  planned  on  so  vast  imperial  a  scale 
that  no  mere  potter  could  fashion  a  dish  large 
enough  to  contain  it.  The  silversmith  alone 
succeeded  where  the  potter  failed,  and  his  silver 
dish  remained  an  object  of  wonder  to  succeeding 
generations  until  the  thrifty  Hadrian  melted  it 
down  for  coin.  In  drunkenness  and  revelling,  in 
gluttony  and  foulness,  the  Emperor  Vitellius  spent 
his  few  months  of  rule.1  And  all  the  while  his 
splendid  army  was  decaying  and  its  two  victorious 
generals  grew  more  jealous  each  of  the  other  every 
day.  "  Truly  it  was  to  the  State's  good  that 
Vitellius  was  vanquished."2 

§  2.    The  Gathering  of  the  Storm 

Meanwhile  heavy  storm-clouds  were  gathering 
on  the  far  horizon  to  east  and  to  north-east.  Vitel- 
lius' treatment  of  the  victorious  army  was  senseless 
enough,  even  though  he  believed  all  danger  of 
further  war  at  an  end ;  but  his  method  of  dealing 
with  the  vanquished  army  was  not  of  such  wisdom 
as  to  warrant  such  a  belief.  Some  small  efforts 
indeed  were  made  to  remove  the  defeated  legions 
from  the  neighbourhood  of  Italy.     The  First  Adju- 

1  Tac.  ii.   62,   94,   95.      Cf.    Suetonius,   Vitellius,    3,    10-13.      There   is 
no  reason  to  discredit  these  accounts. 

-   "  Reipublicae  haud  dubie  intererat  Vitellium  vinci,"  Tac.  iii.  86. 


sec. ii  FLAVIAN  INVASION  OF  ITALY    133 

trix  legion,  which  had  fought  gallantly  for  Otho  in 
the  recent  battle,  was  sent  to  Spain.  The  veteran 
Fourteenth  legion  was  known  to  be  in  a  most 
dangerous  temper.  Only  a  detachment  of  the 
regiment  had  taken  an  active  part  in  the  war,  and 
this  had  stood  its  ground  to  the  last  outside  Cremona 
in  the  centre  of  a  ring  of  foes.  The  legion  as  a  whole 
had  not  been  defeated,  and  indignantly  disowned 
a  share  in  the  blame  for  the  defeat.  It  was  promptly 
ordered  to  return  to  its  old  quarters  in  Britain. 
At  the  time  it  lay  at  Turin,  fretting  and  rebellious, 
quarrelling  as  usual  with  the  ferocious  Batavian 
cohorts  attached  to  it.  So  anxious  was  the  Govern- 
ment to  dispose  of  it  without  the  chance  of  further 
friction  of  any  kind,  that  the  Batavians  were  finally 
detached  from  it  and  sent  off  to  Germany,  and  it 
itself  was  bidden  avoid  the  town  of  Vienne  on  its 
march  through  Gaul.  The  townsfolk  of  this  city 
had  always  wished  Vitellius  so  ill  that  it  was  feared 
the  legionaries  might  be  encouraged  to  make  a 
stand  here  and  refuse  obedience  any  more.  Hence 
they  were  made  to  march  by  the  Little  St.  Bernard 
Pass  over  the  Graian  Alps  to  Montmelian  and 
thence,  instead  of  pursuing  the  usual  route  by 
Grenoble  to  Vienne,  to  strike  away  to  Chambery, 
and  so  direct  to  Lyons.1  These  prudent  precau- 
tions were  of  avail,  and  the  legion  arrived  in 
Britain.  It  had  done  no  damage  on  the  way, 
except  that  it  had  left  its  camp-fires  burning  on  the 
night  when  it  marched  from  Turin,  and  by  some 

1   "  Eo   flexa  itineris,"  Tac.   ii.   66.       The  use  of  the  modern  names  is 
perhaps  justified  for  clearness'  sake. 


134    FLAVIAN   INVASION  OF  ITALY   cr.ii 

means  or  other,  thanks  to  this,  part  of  the  unlucky 
colony  was  burnt  to  the  ground.  This  was  a 
small  price  to  pay  for  riddance  of  the  legion. 
Its  Batavian  comrades  also  duly  reached  their 
homeland  on  the  lower  Rhine.  But  fortune  had 
not  separated  the  cohorts  and  the  legion  for  long. 
The  folly  of  the  Roman  Government  had  sent  the 
Batavians,  now  proud  and  experienced  troops,  back 
to  their  tribesmen  to  add  fuel  to  their  discontent 
and  strength  to  their  plots.  The  "  Indian  Mutiny  " 
of  Roman  history  was,  within  a  few  months,  the 
result.  Then  when  the  tide  of  massacres  and 
Roman  defeats  at  last  was  ebbing,  and  Vespasian's 
Government  set  grimly  to  work  to  crush  the 
mutineers,  the  men  of  the  Fourteenth  legion  came 
gleefully  from  oversea  to  take  vengeance  upon 
their  ancient  enemies  and  old-time  false  comrades 
for  all  the  insults  endured  at  their  hands.1 

These  events  were  quickly  to  happen.  But  for 
the  moment  Vitellius  had  rid  Italy  of  two  of 
the  "  conquered  legions."  With  this,  however, 
his  stock  of  wisdom  was  exhausted.  The  Guards 
and  the  Danube  army  had  also  belonged  to  Otho's 
strength.  These  he  now  treated  with  less  prudence. 
The  Guards  were  disbanded,  with  the  exception  of 
two  cohorts  which  had  done  good  service  in  helping 
to  overawe  the  Batavians  while  these  were  still  in 
camp  with  the  Fourteenth  legion  at  Turin.  Though 
the  disbandment  was  well  managed,  the  cohorts 
being  separated  before  the  order  was  issued,  and 
though  the  men  were  given  the  customary  rewards 

1   See  below,  Chap.  III.      Cf.  Tac.  ii.  66. 


sec.  ii  FLAVIAN   INVASION  OF  ITALY     135 

on  retiring  from  the  service1  (a  treatment  indeed 
which  was  more  generous  than  perhaps  they  had 
any  right  to  expect),  yet  they  regretted  the  loss  of 
their  career,  and  gladly  seized  the  chance  of  taking 
up  arms  again,  which  Vespasian's  rising  so  soon 
gave  to  them.  These  Guardsmen  formed  "  the  v 
strength  of  the  Flavian  cause."2  If  Vitellius  had 
been  able  to  retain  them  under  arms  and  attach 
them  by  interest  to  his  service,  they  might  have 
forgiven  and  forgotten  their  own  defeat  and  Otho's 
death.  But  the  new  Emperor  judged  that  he  had 
too  many  troops  of  his  own.  How  then  could  he 
find  room  in  his  army  for  those  who  had  fought 
against  his  cause  ?  Moreover,  there  was  the  risk 
of  treachery  in  case  of  disturbance.  The  problem  of 
dealing  with  Otho's  Guards  was  certainly  a  delicate 
one  for  Vitellius,  but  the  event  did  not  justify  the 
easy  solution  which  pleased  him. 

The  Danube  army  was  differently  treated.  This 
had  consisted  of  seven  legions :  two  in  Pannonia 
(VII.  Galbianaand  XIII.  Gemina);  two  in  Dalmatia 
(XI.  Claudia  and  XIV. Gemina);  and  three  in  Moesia 
(III.  Gallica,  VII.  Claudia,  and  VIII.  Augusta).  The 
Fourteenth  legion  had  now  been  sent  to  Britain. 
Of  the  other  six,  only  the  Thirteenth  legion  from 
Pannonia  had  been  present  at  the  battle  of 
Bedriacum.  The  survivors  of  this  regiment  were 
at  first  set  to  work  to  build  amphitheatres  at 
Cremona  and  Bononia,  in  which  Vitellius  was  to  be 


1  The  "honesta  missio"  probably  carried  with  it  the  pension  paid  by  the 
Aerarium  militare. 

2  Tac.  ii.  67,  "robur  Flavianarum  partium." 


136    FLAVIAN   INVASION  OF  ITALY   ch... 

entertained  when  he  arrived  in  Italy.  Such  work 
was  inglorious,  and  the  strain  was  not  relieved  by 
the  gibing  of  the  townsfolk  who,  at  Cremona  at 
least,  sharpened  their  silly  wits  upon  the  vanquished 
and  labouring  soldiers.  That  merriment  was  pre- 
sently to  be  recompensed,  and  the  men  of  the 
Thirteenth  exacted  the  full  price,  and  more,  when 
autumn  came.  But  now  it  was  summer,  and  the 
soldiers  performed  masons'  work  till  the  buildings 
were  done.  Then  they  were  sent  back  to  their  old 
headquarters  in  Pannonia  at  Poetovio  (now  Pettau 
on  the  river  Drave  *).  Their  comrades  of  the 
Seventh  legion  had  preceded  them  on  their  return 
to  the  province  at  Vitellius'  orders.  There  the 
two  legions  waited,  nursing  wrath  in  their  hearts, 
longing  for  the  hour  of  requital  for  Otho's  defeat 
and  for  their  damaged  reputation.  The  Eleventh 
legion  had  in  the  same  way  returned  to  its  province 
Dalmatia.  It  too  had  had  no  glut  of  fighting,  and 
was  ready  to  strive  again.2 

But  the  three  Moesian  legions,  the  Third, 
the  Seventh,  and  the  Eighth,  were  bolder,  and 
refused  any  parleyings  with  their  triumphant  rivals, 
the  men  of  the  German  army.  They  were  in  full 
strength,  marching  for  Aquileia,  standing,  as  it  were, 
on  the  very  threshold  of  the  war,  at  the  moment 
when  the  battle  was  being  fought  at  Cremona  a  few 
miles  to  the  west.  Had  they  come  so  far  to  find 
the  door  shut  violently  in  their  very  faces  upon  all 
their  hopes  of  merry  battle  and  the  soldier's  sure 
reward  ?     Messengers  came  bringing  the  tidings  of 

1   Tac.  iii.  I.  2  Tac.  ii.  67,  86  ;  cf.  iii.  32. 


SEC.  II 


FLAVIAN  INVASION  OF  ITALY    137 


Otho's  defeat.  They  chased  them  roughly  from 
the  lines,  and  hastened  forward  to  the  frontier  town 
as  if  scornful  of  the  rumour  of  disaster.  Colours 
were  found  bearing  Vitellius'  name.  They  rent 
them  in  pieces.  If  they  had  not  enjoyed  the  fight- 
ing, at  least  they  would  not  forgo  the  plunder. 
They  seized  on  the  legions'  military  chests,  broke 
them  open,  and  divided  up  the  money.  They 
sought  spoils  on  every  hand.  They  were  in  the 
enemy's  land,  and  as  the  enemy's  land  it  should  be 
treated.  Vitellius  was  leading  his  placid  and  gross 
life  at  Rome.  Caecina  and  Valens  were  contend- 
ing there  for  the  prize  of  greater  honour  at  their 
master's  hands.  But  on  the  north-east  frontier  of 
Italy  were  three  legions,  which  still  defied  them 
all,  and  formed  the  centre  for  the  gathering  storm. 
News  reached  the  legionaries  from  the  distant  East. 
The  Empire  was  not  at  peace.  The  provinces  did  not 
all  rest  quiet  under  Vitellius'  rule.  It  was  but  the 
German  army  which  had  won  him  the  victory.  The 
great  army  of  the  Eastern  frontier  would  acquiesce 
no  more  in  a  triumph  so  lightly  won,  would  accept 
no  longer  so  despicable  an  Emperor.  Vespasian  had 
risen.  Mucianus,  Governor  of  Syria ;  Alexander, 
prefect  of  Egypt ;  all  the  kings,  princes,  peoples, 
and  soldiers  of  the  Eastern  Empire  were  leagued 
together  under  his  banner  against  the  glutton,  the 
puppet  nominee  of  the  savage  German  army.  The 
legions  at  Aquileia  heard  the  news  and  embraced 
the  opportunity.  They  had  disowned  allegiance 
to  Vitellius :  this  then  might  be  counted  loyalty 
to  Vespasian    their   Prince.      They  had  plundered 


138    FLAVIAN   INVASION  OF  ITALY   ch.h 

Italy:  this  was  clearly  a  land  hostile  to  his 
cause.  With  speed  they  sent  to  their  com- 
rades in  Pannonia.  The  Seventh  legion,  swayed 
by  desire  for  revenge  and  by  the  promptings  of 
their  legate,  Antonius  Primus,  gladly  answered  to 
the  call.  The  Thirteenth  legion  had  still  better 
reason  to  join  the  growing  army.  The  Dalmatian 
army,  the  Eleventh  legion,  hesitated.  But  the 
Army  of  the  Danube  was  so  far  united  that  five 
legions  were  ready  to  strike  in  Vespasian's  name 
against  their  old  enemies  and  conquerors.  Letters 
were  at  once  sent  to  the  other  remnants  of  Otho's 
army.  Surely  they  would  not  now  hold  back  from 
the  cause  which  had  revived  again.  The  First 
legion  in  Spain,  the  Fourteenth  in  Britain,  received 
from  their  old  comrades  the  news  of  the  great  rising 
and  the  call  to  arms.  Manifestoes  were  scattered 
broadcast  through  Gaul.  The  little  cloud  in  the 
East  had  become  a  rushing  tempest.  "  In  the 
twinkling  of  an  eye  the  flame  of  a  mighty  war  leapt 
up."  The  banner  of  the  Flavian  cause  was  waving 
on  the  north-eastern  frontier  of  Italy.1 

Flavius  Vespasianus,  the  general  then  busy 
with  the  Jewish  war,  was  at  this  time  nearly 
sixty  years  old.  Neither  his  age  nor  his  blunt 
soldierly  character  of  good-humoured  common  sense 
encouraged  him  to  embark  upon  so  desperate  an 
enterprise  as  that  of  challenging  Vitellius,  the 
Emperor  now  recognised  in  Italy  and  Rome,  for  his 
Imperial  power.     Vespasian  had  survived  the  shock 

1   "  Momento  temporis  flagrabat  ingens  belluni,"  Tac.  ii.  86;  cf.   ii.  85. 
Suetonius,   Vespasian,  6. 


VESPASIAN. 
From  a  bust  in  the  Museo  Nazionale,  Naples. 


sec.  ii  FLAVIAN   INVASION  OF  ITALY    139 

which  his  untimely  gift  of  sleep  had  once  given 
to  Nero's  artistic  sensibilities.1  He  had  placidly 
accepted  Galba  as  his  Prince,  and  sent  his  elder  son 
Titus  from  Judaea  to  do  the  old  Emperor  homage  on 
his  father's  behalf.  He  was  not  inclined  to  quarrel 
with  Otho,  and  his  army  in  due  course  took  the 
oath  of  fidelity  to  him.  Even  Vitellius'  victory  had 
been  recognised  in  the  same  way.  Left  to  himself, 
Vespasian  might  well  have  been  content  to  smoke 
out  the  hornets'  nest  in  Judaea,  whatever  prince 
ruled  at  Rome.  The  burning  of  the  Capitol  at 
Rome  might  have  been  spared,  and  the  burning  of 
Jerusalem  have  been  the  chief  glory  of  the  Flavian 
leader,  not  of  Titus  his  son.  But  the  Fates  of  the 
Roman  Empire  pressed  hard  upon  him.  Every 
influence  was  brought  to  bear  to  move  his  caution 
and  provoke  him  to  defy  Vitellius.  His  own 
army  and  the  great  Army  of  Syria  demanded  the 
right  to  challenge  the  insolent  troops  of  Germany 
for  the  mastery.  Mucianus,  Governor  of  Syria,  who 
had  become  his  close  friend  through  Titus'  willing 
offices,  was  instant  in  his  urging  that  not  even  safety 
could  now  be  won  save  by  accepting  the  last  risk, 
and  promised  him  his  powerful  support.  Men 
worked  upon  that  superstition  which  in  him  was  so 
curiously  interwoven  with  a  hardy  scepticism  and 
healthy  vigour  of  thought — a  purple  strand  in  a 
thick  grey  robe.  The  towering  cypress  tree  which 
rose  above  the  ancestral  mountain -farm  in  far-off 
Samnium,  and  in  Vespasian's  youthful  days  had 
fallen   only  to   rise  again   the   following  day  more 

1  See  my  Life  of  Nero >  p.  380. 


140    FLAVIAN   INVASION  OF  ITALY   ch... 

green,  more  beautiful,  and  all  unhurt  ;  the  statue  of 
the  murdered  Julius  at  Rome,  which,  as  Galba's 
star  was  setting,  turned  to  face  the  rising  sun  ;  the 
two  eagles  which,  before  the  armies  clashed 
together  on  the  battle  -  field  of  Bedriacum,  were 
seen  contending  in  the  air,  when,  behold,  from 
the  eastern  quarter  of  the  heavens  a  third  came 
speeding  and  chased  the  victor  from  his  victory, — 
how  could  omens  such  as  these,  remembered  from 
the  past  or  carried  to  his  credulous  ear  by  eager 
faithful  friends,  how  could  they  be  mockeries  with- 
out meaning?  His  Jewish  captive,  Josephus,  was 
always  whispering  promises  of  coming  Empire  to 
his  master.  The  very  gods  of  the  mysterious  land 
which  he  held  in  iron  grip  knew  of  his  coming 
glory.  Again,  there  was  an  altar  built  on  Mount 
Carmel  to  the  unknown,  unseen  God  ;  again,  the 
priest  stood  to  offer  sacrifice  upon  it,  and  the  Deity 
vouchsafed  his  answer,  not  now  by  fire  from  heaven 
to  confound  the  impious,  but  by  quiet  promise 
through  the  priestly  assurance  that  the  Roman 
general  who  stood  with  hidden  thoughts  offering 
the  sacrifice  should  have  full  fruition  of  his  secret 
hopes.1  That  nothing  should  be  lacking  to  rouse 
Vespasian  to  put  his  fate  to  the  touch,  letters  were, 
it  was  said,  brought  to  him,  purporting  to  be 
written  by  Otho  in  the  brief  time  between  his 
defeat  and  death,  commanding  him  to  take  venge- 

1  This  extraordinary  tale  finds  a  place  in  both  Tacitus  and  Suetonius.  It 
evidently  made  a  deep  impression  on  the  Roman  mind,  which  always  loved  to 
toy  unintelligently  with  Jewish  rites  and  mysteries.  The  very  priest's  name — 
Basilides — is  given.  For  the  omens  in  general,  of  which  I  give  only  a  selec- 
tion, cf.  Tacitus,  ii.  78  ;  Suetonius,  Vespasian,  5. 


nam  FLAVIAN  INVASION  OF  ITALY   141 

ance  on  the  victor  for  his  Emperor's  ruin,  and 
imploring  him  to  help  the  State  in  its  bitter  need. 
Still  the  general  hesitated,  counting  the  strength 
and  valour  of  the  legions  of  Germany.  How  could 
he  pit  his  own  less  war-worn  troops  against  those 
tlushed  with  so  notable  a  victory  ?  But  his  friends' 
impatience  brooked  no  longer  delay ;  his  soldiers' 
enthusiasm  for  their  general  cast  for  him  the 
decisive  throw.  Tiberius  Alexander,  prefect  of 
Egypt,  proclaimed  Vespasian  Emperor  to  his  troops 
at  Alexandria  on  July  1.  His  own  army  acclaimed 
him  Emperor  on  the  3rd,  and  the  three  legions  of 
Syria,  with  Mucianus  the  Governor,  saluted  the 
name  and  ensign  of  the  new  Prince  a  fortnight 
later.  How  the  Danube  army  welcomed  the  news 
has  been  related.  The  fire  of  revolt  was  kindled 
again  on  the  frontiers  of  Italy. 

§  3.  Flavian  Plans  of  War 
A.  The  Muster  of  the  Eastern  Army. — Even 
apart  from  the  six  legions  on  the  north-eastern 
frontier  of  Italy  which  had  declared  for  him, 
Vespasian's  army  in  the  East  was  a  truly  formidable 
one.  In  Syria  were  the  three  legions — the  Fourth 
Scythica,  Sixth  Ferrata,  and  Twelfth  Fulminata. 
The  fourth  legion  which  properly  belonged  to  this 
province  was  the  Third  Gallica,  but  this  had 
recently  been  sent  to  Moesia,  and  was  now  with 
the  other  troops  of  Moesia  at   Aquileia.1     It  was 

1  Hence  Tacitus  speaks  loosely  of  the  Syrian  legions  as  four  :  "  Quattuor 
Mucianus  obtinebat  in  pace  "  (ii.  4),  and  counts  the  Third  Gallica  as  one  of  the 
'•novem  legiones  integrae  e  Judaea  et  Suria  et  Aegypto  "  (ii.  76),  for  there 
were  three  legions  in  Judaea  and  two  in  Egypt. 


142    FLAVIAN   INVASION  OF  ITALY    ch.h 

indeed  a  curious  chance  that  a  habit  acquired  by 
this  legion  during  its  stay  in  Syria  should  help 
largely  to  decide  the  issue  of  the  desperate  battle 
which  was  soon  to  be  fought  in  the  plain  of  the  Po.1 
The  Syrian  legions  were  as  devoted  to  Vespasian's 
cause  as  were  his  own  veteran  troops  in  Judaea, 
not  only  because  they  felt  themselves  part  of  the 
whole  Army  of  the  East,  whose  interests  were  not 
divisible,  but  also  because  Mucianus  in  his  guile 
had  warned  them  that  it  was  Vitellius'  intention  to 
remove  them  to  the  German  frontier,  and  to  send 
his  own  legions  of  Germany  to  enjoy  the  climate 
and  luxuries  of  Syria  in  their  stead.  The  mere 
thought  of  the  bleak  north,  of  the  savage  wilds  of 
the  German  marches  and  their  barbarian  inhabitants, 
of  the  black  forests  and  cold,  wind-swept  marshes, 
of  the  unceasing  toil  and  pitiless,  inclement  weather, 
which  were  the  unfailing  lot  of  those  encamped 
upon  the  Rhine,  and  the  contrast  of  it  all  with  their 
own  peaceful,  happy  life  under  the  warm  Syrian 
sun,  amid  the  groves  and  fountains,  the  thronging, 
busy  streets  and  booths,  the  never-ceasing  merri- 
ments and  festivals  of  Antioch,  excited  in  the 
breasts  of  the  legionaries  of  Syria  the  direst  feelings 
of  resentment  against  the  Emperor  at  Rome." 
Very  reasonably  Vespasian  might  count  all  the 
Syrian  legions  as  "  his  own." 

Equally  eager,  splendidly  disciplined,  and  better 
acquainted  with  their  general  were  the  three 
legions  in  Judaea — the  Fifth  Macedonica,  Tenth 
Fretensis,  and    Fifteenth  Apollinaris.     These  had 

1  Tac.  iii.  24.     See  below,  p.  202.  2  Tac.  ii.  80. 


SEC...  FLAVIAN  INVASION  OF  ITALY    143 

borne  with  him  the  burden  of  the  ferocious  Jewish 
war,  a  struggle  stained  by  every  horror  that  the 
savagery  and  brutality  of  the  religious  fanatic  could 
devise.  Weary  marches,  desperate  sieges,  merci- 
less pursuits,  had  led  the  veteran  troops  at  last 
within  sight  of  the  goal  of  the  bitter  enterprise,  and 
only  the  walls  and  precipices  of  Jerusalem  itself 
still  defied  the  Roman  arms.  There  is  small 
wonder  that  the  soldiers  in  Judaea  would  follow  the 
general  who  had  redeemed  the  Roman  honour  and 
redressed  disgrace,  who  had  given  them  unfailing 
victory  and  immeasurable  spoil,  even  to  the  gates 
of  Rome  itself  with  devotion  and  proud  confidence.1 
The  two  legions  in  Egypt — the  Third  Cyrenaica 
and  Twenty  -  second  Deiotariana  —  brought  the 
number  of  the  legions  of  Vespasian's  Army  of  the 
East  to  eight.  But  in  addition  to  the  legionaries 
there  were  to  be  counted  the  auxiliary  forces  of  the 
Roman  provinces  and  subject  princes  of  the  Eastern 
Empire.  Sohaemus,  Prince  of  Sophene  (the  strip 
of  land  which  borders  Upper  Euphrates  on  its 
eastern  bank  and  surrounds  the  sources  of  the 
Tigris),  came  with  his  native  levies.  Antiochus, 
King  of  Commagene  (the  district  on  the  great  river 
wedged  in  between  the  Roman  provinces  of  Syria 
on  the  south  and  Cappadocia  on  the  north),  who 
was  the  richest  of  the  princes  of  the  East  owning 
Roman  overlordship,  offered  Vespasian  the  resources 
of  his  kingdom.  Herod  Agrippa  II.,  ruler  of  Peraea, 
was  secretly  summoned  from  Rome,  to  which  he 
had  gone  in  the  early  part  of  the  year,  and  sailed 

1   "  Miles  ipsi  adeo  paratus,"  Tac.  ii.  74. 


144    FLAVIAN   INVASION  OF  ITALY   ch.h 

swiftly  back  to  Syria,  leaving  Yitellius  ignorant  of 
his  flight  from  the  city.  His  sister  and  Queen,  the 
beautiful  Berenice,  then  "  at  the  height  of  her 
beauty," l  eagerly  embraced  the  Flavian  cause. 
Titus,  still  an  impressionable  youth  at  twenty-eight, 
was  young  enough  to  be  enamoured  of  her  mature 
charms  ;  Vespasian  was  old  and  wise  enough  to  be 
pleased  by  the  magnificence  of  her  gifts.  There 
was,  finally,  no  portion  of  the  Roman  world  from 
Greece  to  Armenia,  from  Egypt  to  the  Black  Sea, 
which  did  not  swear  allegiance  to  Vespasian.  New 
troops  were  levied.  The  veterans  were  recalled  to 
the  standards.  The  mint  at  Antioch  poured  out 
new  gold  and  silver.  Cities  rang  with  the  clank  of 
hammers  and  the  forging  of  arms.  The  rich  con- 
tributed their  wealth  of  free  will  or  compulsion. 
And  the  chiefs  of  the  party,  the  officers  and  more 
experienced  veterans  of  the  army,  the  princes  of 
the  East  with  brilliant  retinues,  gathered  together 
at  Berytus  for  a  council  of  war.  The  massing  of 
the  infantry  and  cavalry,  the  emulous  rivalry  of  the 
royal  pomp  and  trains,  made  the  Syrian  seaport 
indeed  present  the  appearance  of  a  city  of  the 
Imperial  Court.2 

B.  The  Council  of  War  at  Berytus:  the  "Strategy 
of  Exhaustion." — The  council  of  war,  assembled 
at  Berytus  to  discuss  and  choose  a  strategy  for  the 
coming  campaign,  had  to  take  into  account  not  only 
the  distance   between  the   Syrian  army  and   Italy, 

1  Her  first  husband  (and  uncle),  Herod,  Prince  of  Chalcis,  had  died  twenty 
years  before  this.  But  the  Romans  in  the  matter  of  beauty  were  Venetian 
rather  than  Florentine  or  English  in  taste. 

2  Tac.  ii.  8 1. 


sec.  mi  FLAVIAN   INVASION  OF  ITALY   145 

but  also  that  between  it  and  the  Army  of  the 
Danube.  The  Flavian  forces  were  strong  and 
their  resources  adequate  ;  but  they  were  in  two 
widely  separated  halves,  and,  moreover,  the  un- 
finished Jewish  war  could  not  be  neglected.  John 
of  Gischala  and  the  Zealots  were  not  men  to  wait 
upon  Vespasian's  convenience.  But  the  council 
was  confident  that  their  numbers  were  large  enough 
for  both  wars.  Happily,  there  was  no  other  danger 
on  the  Eastern  frontier.  The  kings  of  Armenia 
and  Parthia  alike  were  friendly  to  the  Flavian  cause, 
and  the  latter,  King  Vologeses,  the  hero  of  the 
great  struggle  with  Rome  in  the  days  of  the 
Emperor  Nero,  actually  made  Vespasian  the  offer 
of  forty  thousand  Parthian  cavalry,  the  most  famous 
horsemen  in  the  world,  to  help  him  against  his 
enemies.  Nothing  serves  so  forcibly  to  illustrate 
the  wisdom  of  Nero's  final  solution  of  the  problem 
of  the  Eastern  frontier  as  does  this  peace  on  that 
frontier  during  the  Roman  civil  wars.  It  was 
exactly  the  time  when  the  Parthian  might  have 
been  expected  to  take  advantage  of  the  discord 
which  was  rending  in  twain  the  strength  of  his 
hereditary  enemy.  But  neither  Vologeses,  nor  his 
brother  Tiridates  in  Armenia,  showed  any  desire 
to  break  the  peace  and  friendship  recently  secured 
by  the  Neronian  policy.  The  Flavian  leaders  could 
therefore  devote  part  of  their  forces  in  the  East  to 
quell  the  rebellion  of  the  Jews,  and  could  direct  the 
rest  to  Italy  undisturbed  by  any  fear  of  an  invasion 
of  the  frontier  or  of  a  sudden  attack  upon  their  rear. 
Titus  and  the  greater  part  of  the  Army  of  the  East 

L 


146    FLAVIAN  INVASION  OF  ITALY   as.  ■ 

were  set  apart  to  end  the  Jewish  war.  This  policy- 
determined,  the  council  turned  its  thoughts  to  the 
war  with  Vitellius.1 

Vespasian  himself,  it  was  decided,  should  not 
conduct  the  campaign  in  person.  He  departed  to 
Egypt  to  seize  and  secure  firmly  the  "  keys  of  the 
country,"  Alexandria  and  Pelusium.  No  corn-ship 
could  sail  thence  to  Italy  without  his  pleasure. 
From  Egypt  as  a  base  he  intended  to  proceed  by- 
sea  and  land  against  the  other  granary  of  Rome, 
the  province  of  Africa.  By  this  means  he  thought 
that  the  enemy  in  Italy  could  be  put  to  great 
distress,  and  that  discord  would  be  the  result  of  it, 
even  though  no  single  Flavian  soldier  had  set  foot 
in  Italy.2  Meanwhile,  Mucianus  was  to  march  by 
land  through  Asia  to  the  Bosporus.  The  best  ships 
of  the  Pontic  fleet,  forty  in  number,  were  summoned 
to  Byzantium  to  effect  and  secure  the  passage  of 
his  army.  Mucianus  himself  led  the  van  of  the 
column,  consisting  of  light  armed  troops  ;  but  there 
followed  as  its  main  strength  the  Sixth  legion  and 
thirteen  thousand  veterans  besides.  At  Byzantium 
he  halted,  hesitating  between  two  strategies.2 

Two  plans  of  campaign,  in  fact,  seem  to  have 
been  considered  by  the  Flavian  leaders.  The  first 
was  that  of  offence  pure  and  simple.  In  pursuit 
of    this    strategy,    Mucianus    should    march    from 

1  Tac.  ii.  82;  iv.  51.  Jerusalem  was  not  taken  and  destroyed  by  Titus 
until  September  a.  d.  70. 

2  Tac.  ii.  82  ;  iii.  8,  48. 

3  Tac.  ii.  82,  83 ;  iii.  47.  The  withdrawal  of  the  ships  and  the  Roman 
troops  in  the  province  to  join  Mucianus'  column  gave  the  opportunity  for  a 
local  rising  in  Pontus,  which  was,  however,  easily  suppressed  by  a  small  force 
sent  later  for  the  purpose  by  Vespasian  (iii.  47,  4S). 


sec.  in  FLAVIAN  INVASION  OF  ITALY   147 

Byzantium  by  the  well-known  military  road,  the 
Via  Egnatia,  through  Macedonia  to  the  seaport  of 
Dyrrhachium  on  the  Adriatic.  There,  if  he  had 
the  ships  to  cross  to  Italy,  he  could  threaten  a 
landing  at  any  one  of  the  harbours  within  reach  as 
opportunity  offered.  Not  only  Brundisium,  the 
Dover  of  Italy,  lay  opposite  and  open  to  his  land- 
ing, but  Tarentum,  and  all  the  coast-line  of  Lucania 
and  Calabria  were  equally  exposed  to  a  hostile 
descent.  Vitellius  would  be  in  sore  perplexity,  not 
knowing  how  to  guard  so  long  a  coast -line,  and 
already  threatened  on  his  north-east  frontier  by  the 
Danube  army.  If  he  sent  his  troops  to  defend  the 
line  of  the  Po  or  the  Julian  Alps,  which  lay  to  the 
east  of  Aquileia,  surely  then  Mucianus  could  make 
a  dash  on  Southern  Italy,  even  on  Rome  itself;  for 
the  road  over  the  Apennines  from  Brundisium  to 
Capua  and  the  capital  had  been  used  before  now 
by  many  an  army.  If,  on  the  contrary,  Vitellius 
massed  his  troops  round  Rome,  he  surrendered  all 
North  Italy  to  the  Danube  army,  and  with  it  the 
courage  and  confidence  of  his  men.  If  he  divided 
his  army  and  sent  one  half  northwards,  keeping  the 
other  to  watch  the  coast,  his  resistance  to  the 
vigorous  attack  on  both  sides  was  likely  to  be  but 
an  enfeebled  one.  The  march  to  Dyrrhachium 
from  Byzantium  involved  the  strategy  of  offence 
and  co  -  operation  between  the  Danube  and  the 
Eastern  army. 

But  this  was  not  the  strategy  which  Vespasian 
himself  wished  his  troops  to  adopt.  He  preferred 
to    rely    upon    slower    means    for   exhausting    and 


148    FLAVIAN   INVASION  OF  ITALY   cb.ii 

wearing  out  the  enemy.  If  his  own  strategy  were 
followed,  Mucianus  was  to  march  from  Byzantium 
by  the  valleys  of  the  Moritza  and  Morava  through 
Moesia,  and  up  the  valley  of  the  Drave  through 
Pannonia  and  over  the  Julian  Alps  to  Aquileia.  The 
Army  of  the  Danube  was  to  wait  there  until  Muci- 
anus arrived.  So  the  whole  Flavian  force  would 
be  concentrated  to  threaten  Italy  on  the  north-east, 
and  meanwhile  all  supplies  of  corn  to  the  enemy 
would  be  cut  off  by  Vespasian's  activity  in  Egypt 
and  in  Africa.  Hunger  and  despair,  the  would-be 
emperor  hoped,  would  do  the  work.  The  Vitellians, 
starved  and  desperate,  would  submit  without  fight- 
ing. The  strategy  of  exhaustion  and  combination, 
not  that  of  offence  and  co-operation,  should  be 
employed.  And  at  Byzantium  Mucianus  definitely 
chose  this  plan.  He  sent  bidding  the  Danube 
army  not  to  move  from  its  lines  at  Aquileia,  and 
marched  himself  with  his  whole  force  for  that  city.1 
The  other  strategy,  that  of  offence,  was  indeed 
attractive.  The  plan  of  a  "double  objective" 
always  perplexes  the  enemy,  and  when  skilfully 
used,  as  by  the  Japanese  in  the  war  of  1894,  mav 
always  lead  to  notable  successes.  Such  a  plan  was 
involved  in  this  Flavian  strategy  of  offence.  But 
apart  from  its  greater  risks,  which  might  well  have 
been  ventured,  two  difficulties  in  its  way  were 
serious.  From  the  middle  of  July  to  the  middle 
of  August  the  prevailing  wind  in  the  Adriatic  is 
that  from  the  north-west,  and  this  greatly  hampered 
naval  operations  from  Dyrrhachium  as  a  base.     A 

1  Tac.  ii.  83  ;  iii.  8. 


sec.  ,ii  FLAVIAN  INVASION  OF  ITALY    149 

still  more  weighty  objection  was  the  fact  that  the 
command  of  the  sea  was  as  yet  by  no  means  en- 
sured to  the  Flavians.  About  the  very  time  that 
the  plan  of  invasion  oversea  was  being  discussed  at 
Berytus,  a  Flavian  general,  at  the  council  of  war 
held  by  the  officers  of  the  Danube  army  at  Poetovio, 
pointed  out  that  the  enemy  had  two  fleets,  those  of 
Ravenna  and  Misenum,  and  that  these  might  easily 
take  the  offensive  by  sea,  for  there  was  no  Flavian 
fleet  in  the  Adriatic  to  stop  them.  At  the  outset 
of  the  campaign,  indeed,  the  command  of  the  sea 
rested  with  the  Vitellians.  It  is  true  that  the 
Ravenna  fleet  quickly  proved  treacherous  to  their 
cause  ;  but  its  coming  treachery  was  not  an  element 
in  the  strategical  situation  on  which  plans  could  be 
built,  or  by  which  strategy  could  be  determined, 
either  at  Berytus  or  at  Poetovio.  In  old  days  Sulla 
had  crushed  the  democrats  at  Rome  by  his  strategy 
of  attack  oversea  from  Dyrrhachium.  Later  in  the 
century,  Julius  Caesar,  for  fear  lest  the  like  strategy 
should  be  used  by  Pompey  against  him,  had  been 
forced  to  risk  the  passage  of  the  Adriatic  in  face  of 
the  enemy's  superiority  at  sea,  and  to  seek  out  his 
foe  in  Epirus.  Still  later,  Antony  had  threatened 
Octavian  with  the  same  strategy.  But  both  Sulla 
and  Pompey  had  had  command  of  the  Adriatic. 
Antony  had  been  master  of  a  powerful  fleet. 
Mucianus  neither  possessed  the  command  of  the 
sea,  nor  had  as  yet  any  means  of  gathering  a  fleet 
to  secure  it.  The  familiar  strategy  of  offence  by 
sea  had  therefore  to  be  abandoned  by  him.1 

1  Cf.  Tac.  ii.  98  ;  iii.  2  ;  Annals,  vi.  55.      Pliny,  Nat.  Hist.  ii.  124. 


150    FLAVIAN   INVASION  OF  ITALY   ch.u 

But  it  by  no  means  followed  as  a  consequence 
that  no  strategy  of  offence  could  be  tried.  Ves- 
pasian's proposed  "strategy  of  exhaustion"  deserved 
hearty  condemnation.  Apparently  he  intended  his 
army  at  Aquileia  to  do  nothing,  even  after  Mucianus 
and  his  force  had  joined  the  Danube  legions  there. 
The  Vitellians  were  to  be  starved  into  submission 
and  blockaded  during  the  process  on  the  north-east 
frontier.  From  a  political  point  of  view  it  might 
be  urged  that  this  plan  would  save  bloodshed. 
From  a  military  point  of  view  there  was  nothing  to 
be  said  for  it  at  all.  Many  losses  have  been  caused, 
many  campaigns  well-nigh  ruined,  by  the  interfer- 
ence of  the  politician  with  the  conduct  of  military 
operations.  Virginia,  Natal,  the  Yalu,  have  all 
recently  enforced  again  this  lesson,  which  is  as  old 
as  the  history  of  war.  But  Vespasian  hitherto  had 
been  far  more  of  a  soldier  than  a  statesman,  and  his 
plans  for  the  campaign  against  Vitellius  remain  all 
the  more  a  puzzle ;  for  how  could  it  be  expected 
that  this  "strategy  of  exhaustion"  would  end  the 
war  in  his  favour  ?  Sextus  Pompeius  had  tried  the 
plan  of  starving  Italy  before  now,  but  Octavian  had 
defied  even  this  risk ;  and  Octavian  was  far  less 
popular  with  his  troops  than  was  Vitellius  with  his 
army,  which  was  still  devoted  to  his  interests.  The 
civilian  in  Italy  might  suffer  if  no  corn-ships  came 
from  the  South  ;  but  the  soldier  would  find  a  way, 
even  at  the  civilian's  expense,  to  feed  himself.  And 
not  even  to-day  are  the  issues  of  war  decided  by 
the  clamour  of  civilians  ;  still  less  would  this  have 
been    the    case    during    the    civil   war    of  a.d.    69. 


sec.  in  FLAVIAN   INVASION  OF  ITALY    151 

Neither  was  Italy  so  barren  of  food,  so  dependent 
on  sea-borne  corn,  so  pitifully  unable  to  feed  her 
own  children,  as  England  is  to-day.  The  proleta- 
riate in  Rome  might  have  felt  the  pinch  of  need, 
and  its  anger  was  doubtless  dangerous  enough  to 
an  emperor  in  Rome  ;  but  there  was  no  need 
(other  than  the  excellence  of  the  palace  kitchen) 
for  Vitellius  to  stay  in  the  city  and  listen  trembling 
to  the  howling  of  the  hungry  mob.  There  is  no 
evidence  that  the  corn  from  Egypt  or  Africa  was 
necessary  for  any  place  save  Rome,  or  even  ever 
sent  up  country  at  all.  Italy  fed  herself  even 
though  Rome  starved.  There  was  no  transport 
for  corn,  no  commerce  in  corn,  from  Rome  to  the 
other  districts  of  the  land  any  more  in  a.d.  69  than 
there  was  in  the  days  of  the  Gracchi.  Vespasian's 
"  strategy  of  exhaustion  "  was  not  in  the  least  likely 
to  exhaust  the  Vitellian  army.  Its  ultimate  base  of 
supplies  was  still  the  western  part  of  the  Empire, 
Germany  and  Gaul.  Neither  Vespasian  in  Egypt 
nor  Mucianus  at  Aquileia  threatened  in  any  way 
the  safety  of  the  enemy's  communications  with  the 
lands  which  were  still  the  source  of  their  strength, 
the  place  of  replenishment  for  their  resources.  The 
strategy  of  masterly  inactivity,  if  pursued  in  this  way 
by  the  Flavians,  would  not  discourage  the  temper 
of  the  Army  of  Germany.  They,  after  all,  were  the 
men  in  possession.  Rome  and  Italy  were  the  sign 
of  Empire,  the  crown  of  conquest.  It  must  be  the 
Flavians'  part  to  attack  and  evict,  for  theirs  had 
been  the  challenge.  If  they  rested  quiet,  the  Vi- 
tellian army  in  high  scorn  might  take  the  initiative 


152    FLAVIAN   INVASION  OF  ITALY 

themselves.  They,  by  means  of  the  fleet,  might 
descend  upon  the  motionless  force  at  Aquileia,  land 
in  their  rear,  and  cut  them  off  hopelessly  from  their 
base  and  line  of  retreat,  while  the  Vitellians  could 
always  use  the  fleet  as  a  base.  The  value  of  the 
possession  of  the  strategical  initiative  in  war  cannot 
be  set  too  high.  The  force  which  possesses  the 
command  of  the  sea  is  the  more  likely  to  possess 
this  initiative  if  numbers  are  equal ;  and,  possess- 
ing it,  such  a  force  compels  the  enemy  to  make 
his  dispositions  conform  to  its  own  plans  for 
the  campaign.  But  the  Flavians,  if  the  attacking 
party,  might  seize  at  the  outset  the  strategical 
initiative,  thanks  to  the  fact  that  a  large  army  lay 
already  on  the  frontier  of  the  enemy's  country  when 
the  war  broke  out,  whereas  the  defending  army  was 
slowly  moving  north  from  Rome.  If  they  let  the 
opportunity  slip,  the  initiative  would  naturally  pass 
to  the  Vitellians.  The  "  strategy  of  exhaustion," 
which  would  have  been  no  exhaustion  at  all  of  the 
Vitellians,  wilfully  abandoned  this  the  Flavian  great 
military  opportunity.  It  made  a  present  of  the 
strategical  initiative  to  the  foe.  This  strategy, 
in  fact,  is  appropriate  only  to  the  weaker  side, 
and  then  under  very  special  conditions.  Fabius 
Maximus  used  it  at  a  crisis  and  saved  Rome.  But 
it  was  not  this  strategy  which  gave  Rome  at  last 
the  victory  in  the  second  Punic  war.  Pericles 
used  it  against  the  Peloponnesians,  and  thereby 
ran  the  ship  of  State  hard  on  the  rocks,  whence 
more  skilful  pilots  had  to  rescue  it.  Frederick  the 
Great  saved  his  kingdom  by  its  use,  and  had  indeed 


sec.  in  FLAVIAN  INVASION  OF  ITALY   153 

no  other  choice.  But  the  quarrel  was  not  of  his 
seeking,  and  his  prize  of  victory  was  not  conquest 
but  his  country's  preservation.  Such  a  strategy 
might  have  been  appropriate  to  the  Vitellians  :  the 
choice  of  it  by  the  Flavians,  whose  part  it  was 
to  conquer,  not  to  preserve,  must  have  gone  far 
to  ruin  their  cause.  The  Danube  army  meant  to 
fight.  Were  they  to  be  told  that  fighting  was  too 
dangerous  ?  For  what  other  purpose  were  they 
at  Aquileia  ? — To  wait  attack  by  the  hated  foe  ? 
To  be  cut  off  by  an  enemy  landing  from  the  sea 
in  their  rear,  and  a  chain  of  fortresses  in  a  hostile 
land  to  their  front  ?  Roman  war  is  fought  with 
men,  not  with  automata.  The  Flavian  was  the 
challenger.  He  had  flung  down  the  gauntlet. 
Was  he  to  retire  to  his  tent  until  the  other 
champion  came  to  turn  him  out  with  ignominy, 
scarcely  leaving  him  time  to  mount  his  lazy  steed  ? 

Offensive  strategy  alone  can  quickly  end  a  war.  The  passive 
attitude  may  eventually  induce  the  weary  enemy  to  seek  for 
peace ;  it  can  never  produce  the  same  effect  as  the  offensive 
crowned  with  tactical  victory.  Hence  in  strategy  the  defensive 
should  never  be  assumed  except  as  a  temporary  measure,  or  by 
the  weaker  side,  to  be  changed  to  the  offensive  as  soon  as 
opportunity  offers.  A  nation  which  declares  war  and  acts  on 
the  defensive  shows  that  it  does  not  understand  the  condition 
most  essential  for  success.1 

Whether  in  reality  this  "  strategy  of  exhaustion  " 
would  have  been  pursued  by  Mucianus  after  his 
arrival  at  Aquileia  or  not  cannot  be  determined  ; 
for  despite   his  orders   sent   to   the   Danube   army 

1  James,  Modern  Strategy,  pp.  41,  42. 


i54    FLAVIAN   INVASION  OF  ITALY 

there,  and  long  before  his  arrival,  this  had  taken 
the  bit  in  its  teeth  and  charged  furiously  upon  the 
foe. 

C.  The  Council  of  War  at  Poetovio  :  the  "Strategy 
of  Annihilation." — While  at  Berytus  Vespasian  and 
Mucianus  were  discussing  plans  for  the  war,  the 
officers  of  the  Danube  army  met  at  Poetovio,  the 
town  near  the  western  frontier  of  the  province  of 
Pannonia,  to  deliberate  on  their  own  account.  The 
actual  governors  of  the  three  provinces — Pannonia, 
Dalmatia,  and  Moesia — took  but  a  small  part  in 
the  council  and  in  the  military  operations  which 
followed.  The  Legate  of  Pannonia,  Lucius  Tampius 
Flavianus,  does  indeed  seem  to  have  been  present, 
and  he  presently  accompanied  the  army  of  invasion 
as  far  as  Verona ;  but  he  was  cautious  by  nature, 
old  in  years,  and,  moreover,  a  kinsman  of  Vitellius, 
and  at  first,  when  the  disturbance  began  in  Pan- 
nonia, fled  hastily  to  Italy  to  be  out  of  the  way  of 
danger.  The  persuasions  of  Cornelius  Fuscus, 
procurator  of  the  province,  induced  him  to  return. 
The  procurator  desired  for  his  party  the  prestige 
attaching  to  the  name  of  one  who  had  been  consul ; 
the  governor  hoped  to  pluck  some  profit  from  the 
rebellion.  But  the  troops  naturally  mistrusted  him, 
and  at  Verona  they  made  an  excited  attack  upon 
him,  which  nearly  cost  him  his  life,  and  got  rid  of 
him  away  from  the  army  for  good  and  all.1  Marcus 
Pompeius  Silvanus,  Governor  of  Dalmatia,  was,  like 


1  Tac.  ii.  86  ;  iii.  4,  10.  Cf.  Fliny,  Nat.  Hist.  ix.  26.  He  was  consul 
about  a.d.  46  and  again  under  Vespasian  in  74.  He  seems  to  have  given  his 
name  to  the  ala  I.  Pannoniorum  Tampiana,  C.I.L.  iii.  p.  S64. 


sec.  in  FLAVIAN  INVASION  OF  ITALY   155 

Tampius,  old,  rich,  and  more  of  a  financier  than  a 
general.  He  certainly  was  not  of  any  weight  in 
the  council,  if  present ;  for  the  one  legion  of  his 
province,  the  Eleventh  Claudia,  was  still  hesitating, 
and,  in  fact,  the  legion  and  its  governor  appeared  in 
the  Flavian  camp  only  after  the  first  great  victory 
had  been  won  and  Cremona  had  fallen.  These 
were  incidents  which  relieved  the  anxiety  of  general 
and  private  concerning  the  probable  issue  of  the 
struggle,  but  gave  them  matter  for  anxious  thought 
concerning  their  reception  in  the  victors'  camp  if 
they  delayed  longer  to  join  them.  Silvanus,  a  man 
"  apt  to  waste  in  words  the  hour  for  deeds,  and  a 
sluggard  in  war,"  was  wholly  ruled  by  the  legate 
of  the  Eleventh  legion,  Annius  Bassus,  who  by 
judicious  deference  to  the  old  governor  won  his 
assent  to  all  his  own  plans  and  carried  them  out 
quietly  and  ably.1  Thus  neither  Tampius  nor 
Silvanus  added  any  strength  to  the  Flavian  cause 
except  the  lustre  of  their  names  and  the  benefit 
of  their  actual  insignificance.  The  third  of  the 
three  legates,  M.  Aponius  Saturninus,  Governor  of 
Moesia,  was  absent  from  the  council  of  war.  He 
had  indeed  written  hurriedly  to  Vitellius  the  news 
that  the  Third  legion  in  Moesia  had  mutinied 
against  his  authority.  Later,  however,  he  had  seen 
fit  to  follow  the  lead  of  his  troops,  and  declare  for 
Vespasian,  and  early  in  the  campaign  he  appeared 
in  the  Flavian  camp  with  one  of  the  three  legions 
of  his  province,  the  Seventh  Claudia.      But  some 

1  Silvanus:  consul  a.d.  45.     Cf.  Tac.  ii.  86;   iii.   50;   iv.  47  ;  Annals, 
xiii.  52.      In  74  he  was  consul  again  with  Tampius. 


156    FLAVIAN  INVASION  OF  ITALY   ch... 

letters  which  he  was  supposed  to  have  written  to 
Vitellius  were  one  day  published  in  the  camp,  and 
the  fiery  and  suspicious  troops  indignantly  joined 
in  an  eager  hunt  for  the  traitor  through  the 
gardens  where  he  was  staying.  Aponius  saved 
his  life  by  hiding  promptly  in  the  furnace  of  some 
disused  baths,  and,  when  the  storm  blew  over, 
retired  to  Padua  and  took  no  further  part  in  the 
war.1  It  was  not  to  such  time-servers  as  an 
Aponius,  a  Tampius,  a  Silvanus,  that  Vespasian 
owed  his  Empire.  Three  men  of  lower  rank, 
whose  military  energy  was  spoilt  by  no  politic 
caution,  whose  zeal  on  his  behalf  was  hampered 
by  no  considerations  of  their  own  dignity,  swayed 
the  counsels  of  the  Danube  army  and  led  irresolu- 
tion captive  to  daring. 

Cornelius  Fuscus,  the  procurator  of  Pannonia, 
had  as  a  mere  youth  preferred  the  Imperial  Civil 
Service  to  the  Senatorial  career  which,  as  a  lad 
of  good  family,  he  would  naturally  have  followed. 
Renouncing  the  rank  to  which  his  birth  entitled 
him,  he  chose  the  career  which,  if  of  less  repute, 
gave  greater  opportunities  of  a  fortune.  But  the 
tempestuous  days  of  the  civil  wars  afforded  him 
chances  of  action  which  he  loved  better  even  than 
money-making.  To  him,  now  in  the  vigour  of  his 
early  manhood,  battle  was  a  delight,  risk  more 
joyous  than  certainty,  peril  than  the  rewards  of 
peril.  His  services  in  the  war  won  him  honour 
and  promotion.  Under  Domitian  he  was  prefect 
of  the   Praetorian   Guard,   and   he  perished  at  the 

1  Tac.  ii.  96  ;  iii.  5,9,  II. 


sec.  ,.i  FLAVIAN   INVASION  OF  ITALY   157 

head  of  his  troops,  trapped  and  destroyed  by  the 
barbarians  in  the  second  Dacian  war  of  a.d.  89. x 

Arrius  Varus  was  probably  an  older  man,  but 
the  military  experience  which  he  had  gained  while 
serving  under  Corbulo  in  the  Parthian  war  twelve 
years  before  served  him  in  good  stead.  The 
rapidity  with  which  the  column  of  invasion  swept 
down  upon  North  Italy  showed  that  he  had  not 
forgotten  the  lessons  taught  him  in  the  field  by  his 
old  master  in  the  art  of  war.2  His  very  success 
earned  for  him,  later,  Mucianus'  suspicion,  and  he 
had  to  suffer  degradation  from  the  office  of  Prae- 
torian prefect,  which  had  been  his  reward,  being 
given  instead  the  inferior  position  of  prefect  of  the 
Corn  Supply.3 

Such  energy  was  exhibited  also  in  an  equal, 
if  not  in  still  larger,  measure  by  the  third  of  the 
three  leaders,  M.  Antonius  Primus.  This  officer 
is  the  hero  of  the  successful  Flavian  invasion 
of  Italy.  His  earlier  career,  indeed,  did  him 
little  credit.  Eight  years  before,  he  had  been 
condemned  at  law  as  one  of  the  witnesses  to  a 
forged  will ;  but  Galba  had  restored  him  to  his 
senatorial  rank  and  given  him  the  command  of  his 
new  legion,  the  Seventh  Galbiana.  Antonius  had 
a  happy  confidence  in  his  own  ability,  and  men  said 
that  he  had  written  to  Otho  offering  himself  as 
general-in-chief  for  his  war  with  Vitellius.  That 
Emperor  was  already  plagued  by  too  many  general 
officers,  and  took  no  notice  of  the  offer.      But  now 

1  Tac.  ii.  86  ;  iii.  4,  12,  42  ;  iv.  4.      Suetonius,  Domitian,  6. 

2  Tac.  iii.  6  ;  Annals,  xiii.  9.  3  Tac.  iv.  68. 


158    FLAVIAN   INVASION  OF  ITALY   «*.  u 

at  last  Antonius'  chance  was  come.  He  was  an 
able  speaker,  and,  when  the  news  of  Vespasian's 
rising  reached  Pannonia,  he  at  once  harangued  the 
troops  on  behalf  of  the  Flavian  general,  and  that 
in  no  ambiguous  terms.  Others  might  strive  to 
face  both  ways,  but  Antonius  was  impatient  of  such 
shallow  cowardice.  The  blunt  soldiers  found  in 
him  a  man  after  their  own  heart.  His  unscrupulous 
dexterity  might  bring  disgrace  upon  his  rivals ;  he 
might  rob  with  the  one  hand  and  fling  money 
broadcast  with  the  other ;  but  he  was  a  stalwart 
soldier  who  knew  his  own  mind,  and  if  he  advertised 
his  own  merits,  at  least  they  were  merits  which  his 
fellow-soldier  loved,  and  he  really  possessed  them. 
The  men  of  the  Seventh  legion  knew  that  they 
had  found  a  man  to  lead  them,  however  hazardous 
the  enterprise,  and  were  impatient  of  any  other 
general.1 

When  such  officers  met  in  council  at  Poetovio, 
it  is  not  surprising  that  a  vigorous  strategy  found 
favour.  There  were  indeed  some  who  urged  that 
they  were  bound  to  wait  for  the  arrival  of  Mucianus 
and  their  comrades  of  the  Syrian  army.  They 
dwelt  upon  the  strength,  the  fame,  the  recent 
victory,  of  the  enemy.  Their  own  position,  they 
urged,  could  easily  be  made  impregnable  against 
attack  until  the  reinforcements  came.  The  high 
road  from  Poetovio  by  Emona  and  Nauportus  to 
Aquileia,  a  hundred  and  fifty  miles  away,  had  to 
cross  the  mountain  chain  of  the  "  Pannonian  Alps  " 
to    the    north    of    the    Istrian    promontory.       The 

1  Tac.  ii.  86  ;  iii.  3,11;  Annals,  xiv.  40. 


sec.  in  FLAVIAN  INVASION  OF  ITALY    159 

passes  of  the  ridge,  they  argued,  could  be  blocked, 
and  their  army  would  then  rest  in  safety  under 
cover  of  the  mountains.  "  Conquered  troops,"  they 
asserted,  alluding  to  the  recent  defeat  which  some 
of  their  army  had  suffered  at  Bedriacum,  "  may  talk 
as  boldly  as  they  please,  but  they  have  not  the 
courage  of  their  conquerors  for  all  that." 

A  fiery  speech  by  Antonius  consumed  this 
advocacy  of  delay  in  a  moment.  The  council  of 
war  was  held  in  open  air,  and  the  centurions  and 
even  private  soldiers  came  thronging  up  to  assist 
the  deliberations  of  their  officers.  Antonius'  clear 
loud  voice  rang  through  the  camp,  and  he  carried 
even  the  more  cautious  away  by  his  fierce  eloquence. 
In  bitter  terms  he  described  the  demoralisation  of 
the  Vitellian  soldiery.  "  Scattered  through  the 
townships  of  Italy  and  no  longer  under  arms, 
sluggard  guests  dreaded  only  by  their  hosts,  drain- 
ing the  cup  of  new,  strange  pleasures  with  a  wild 
zest,  a  zest  as  great  as  was  that  rude  ferocity  which 
once  was  theirs  and  was  theirs  no  longer,  these 
erstwhile  soldiers  of  Vitellius  had  lost  their  hardi- 
ness in  the  circus,  the  theatre,  the  allurements  of 
the  capital.  Yes,  but  they  were  soldiers  still. 
Give  them  but  time,  and  the  very  thought  of  war 
would  brace  them  again  to  valour.  Germany  and 
Gaul,  Britain  and  Spain,  Italy  and  Rome  itself, 
would  send  out  new  armies  to  fill  their  ranks. 
Nor  would  their  own  position  be  safe  behind  the 
ramparts  of  the  mountains.  Vitellius'  fleets  com- 
manded the  Adriatic,  and  it  would  be  easy  to  land 
an  army  in  their  rear.     Where  then  would  be  the 


160    FLAVIAN   INVASION  OF  ITALY   ch.h 

service  of  their  delay,  and  where  the  food  and 
money  for  the  troops  if  they  lay  idle  till  next  year's 
summer  came  ?  They  had  not  been  defeated,  but 
tricked  into  submission.  The  day  for  vengeance 
for  this  trickery  was  come.  Their  comrades  of  the 
Moesian  army  had  lost  no  single  man.  What  did 
it  matter  that  they  were  inferior  in  legionaries  ? 
In  discipline,  in  sobriety,  in  very  numbers,  if 
men  of  every  arm  were  counted,  they  had  the 
mastery  of  the  foe.  Above  all,  in  cavalry  lay  their 
own  great  strength.  In  the  battle  lately  fought 
two  little  squadrons  had  charged  and  broken  the 
enemy's  line.  Were  these  defeated  troops  ?  But 
now  sixteen  squadrons  of  horse  would  overwhelm 
with  the  thunder  of  the  onset  of  their  serried  ranks, 
would  bury  beneath  the  rushing  wave  of  their 
furious  charge,  horses  and  horsemen  forgetful  of 
battle.  Keep  back  the  legions,"  he  cried,  turning 
scornfully  to  the  advocates  of  caution  ;  "  keep  back 
the  legions,  you  who  risk  nothing  by  defeat,1  and 
give  me  the  cohorts  only.2  I  have  planned,  and  I 
will  carry  out  the  plan.  You  will  be  glad  enough 
to  follow  in  my  steps  when  the  victory  is  won."3 

After  such  a  speech  there  was  no  room  left  for 
moderate  counsels.  The  soldiers  cheered  "  their 
one  and  only  leader  "  to  the  echo,  and  the  council 
broke  up  intent  on  an  immediate  advance  into  Italy. 
A  message  was  sent  to  Aponius  Saturninus  bidding 
him   hasten   to   bring  the   Moesian   army   into   the 

1   "  Quibus  fortuna  in  integro  est,"  i.e.  who  have  not  compromised  them- 
selves fatally  with  Vitellius,  but  still  hope  to  be  able  to  sit  on  the  fence. 
-  I.e.  the  auxiliaries. 
3  Tac.  iii.  1-3. 


sec.  in  FLAVIAN   INVASION  OF  ITALY   161 

field.1  Certain  precautions  also  were  taken  to 
ensure  the  safety  of  the  northern  frontier  when  the 
provinces  south  of  the  Danube  were  stripped  bare 
of  troops.  There  was,  in  fact,  danger  all  along  the 
Danube  frontier.  West  of  Pannonia  lay  the  dis- 
trict of  Noricum,  which  was  threatened  by  the 
procurator  of  Raetia,  who  was  firmly  loyal  to 
Vitellius.  On  the  north  of  Pannonia,  occupying 
the  district  north  of  the  Danube  between  the  rivers 
March  and  Waag,  lay  the  tribe  of  the  Suebi,  ruled 
by  two  princes  jointly,  Sido  and  Italicus.  Though 
vaguely  in  the  sphere  of  Roman  influence  the 
Suebi  were  practically  independent  of  Rome.  The 
Danube,  after  flowing  in  an  easterly  direction  for 
many  miles,  turns  abruptly  to  the  south  at  a  point 
about  a  hundred  and  forty  miles  east  of  Vienna, 
and  continues  on  the  southerly  course  for  not  far 
short  of  two  hundred  miles.  This  reach  of  the  river 
formed  the  eastern  boundary  of  Pannonia.  Some 
fifty  or  sixty  miles  to  the  east  of  it  the  Theiss  flows 
parallel  to  the  Danube,  and  joins  it  from  the  north 
after  the  greater  river  has  turned  eastwards  again. 

1  Tacitus,  whose  account  of  the  P'lavian  invasion  is  far  more  satisfactory 
than  that  of  the  war  of  Otho  and  Vitellius,  leaves  us,  however,  in  great 
perplexity  as  to  the  actual  position  of  the  three  Moesian  legions  at  this  time. 
In  April  they  have  already  "entered  Aquileia"  (ii.  46),  and  are  there  when 
they  refuse  allegiance  to  Vitellius,  "  Aquileiam  progressae "  (ii.  85).  But 
the  council  of  war  at  Poetovio  sends  bidding  Saturninus  "  cum  exercitu 
Moe.^ico  celeraret "  (Hi.  5),  and  Antonius'  first  act  is  to  occupy  Aquileia 
with  his  auxiliaries  (iii.  6).  When  he  moves  on  to  Verona,  it  is  some  days 
before  the  Moesian  army  joins  him  there  (iii.  9),  and  then  they  arrive  in  two 
detachments,  the  Seventh  legion  first,  and  then  the  Third  and  Eighth  (iii. 
9,  10).  It  is  possible  that  they  were  at  Aquileia  all  the  time,  as  Vespasian 
appoints  this  town  for  the  general  rendezvous  (iii.  8).  But  iii.  5  seems  to 
imply  that  they  had  withdrawn  again  to  Moesia.  Tacitus,  however,  never 
tells  us  this  directly,  and  this  is  characteristic  of  the  looseness  of  his  military 
narrative. 

M 


162    FLAVIAN   INVASION  OF  ITALY   ch.h 

The  strip  of  land  between  the  Danube  and  the 
Theiss,  some  two  hundred  miles  therefore  in 
length,  was  occupied  by  a  tribe  of  Sarmatian  stock, 
the  Jazyges,  which  always  maintained  its  independ- 
ence of  the  Roman  Empire  even  after  Trajan's 
conquest  of  Dacia  many  years  later.  A  tribe  of 
hardy  horsemen,  they  could  defy  any  attempt  of 
the  slow-moving  legionary  to  subdue  them  in  their 
native  wilds.  East  of  the  Jazyges,  north  of  the 
Danube  and  the  province  of  Moesia,  lay  the  power- 
ful and  restless  tribesmen  of  the  Dacians,  and  the 
lower  course  of  the  river  to  the  sea  had  roving 
Sarmatian  tribes,  such  as  the  Roxolani,  upon  its 
northern  bank. 

So  far  as  was  possible,  the  Flavian  leaders 
secured  the  safety  of  all  this  vast  length  of  frontier 
before  they  directed  the  army  which  garrisoned  it 
upon  Italy.  A  special  expeditionary  force  was  sent 
under  an  able  officer  to  assist  the  native  levies  of 
Noricum  to  defend  the  line  of  the  river  Inn  against 
attack  from  Raetia.  This  force,  consisting  of  eight 
cohorts  of  auxiliary  infantry  and  one  squadron  of 
Spanish  horse,  the  ala  I.  Hispanorum  Auriana, 
under  Sextilius  Felix,  was  unmolested  by  the 
enemy  over  the  river.1  Sido  and  Italicus,  princes 
of  the  Suebi,  with  a  cavalry  bodyguard  of  their  own 
people,  actually  joined  the  column  of  invasion,  and 
fought  for  Vespasian's  cause  at  Cremona.-  In  like 
manner  the  chiefs  of  the  Jazyges  offered  their  ser- 
vices and  those  of  the  horse  and  foot  of  the  tribe. 
The  latter   offer  was    declined,   since    the   Flavian 

1  Felix,  cf.  Tac.  iv.  70.  -  Ibid.  iii.  21. 


sec  n,  FLAVIAN   INVASION  OF  ITALY   163 

leaders  could  not  trust  such  allies'  loyalty,  if  it 
should  be  tempted  by  bribes  from  the  enemy.  But 
they  prudently  secured  pledges  for  the  peace  of 
that  section  of  the  frontier  by  taking  the  chiefs 
themselves  with  them.  By  such  measures  the 
safety  of  Pannonia  was  guaranteed  in  the  absence 
of  its  garrison ;  but  the  longer  Moesian  frontier 
was  left  dangerously  denuded  of  Roman  troops. 
The  Governor  of  Moesia  may  have  relied  upon  the 
effect  of  the  crushing  blow  which  had  in  the  pre- 
vious winter  been  dealt  to  the  raiders  of  the 
Roxolani  in  the  province.1  But  the  Dacians  were 
but  eagerly  watching  for  their  opportunity.  As 
soon  as  the  legions  marched  for  Italy,  they  crossed 
the  river  and  fell  upon  the  Roman  camps  on  the 
southern  bank.  Happily  for  the  Roman  province, 
Mucianus  was  already  in  Moesia  on  his  march  to 
Italy,  and  sent  off  the  Sixth  Ferrata  legion  of  the 
Syrian  army  in  hot  haste  to  the  rescue.  For  a 
short  time  the  Dacians  were  repelled,  but  the 
situation  on  the  lower  Danube  grew  more  and 
more  ominous.2 

But,  meanwhile,  Italy  had  been  the  scene  of 
fighting.  Antonius  and  his  fellow-generals  had 
taken  such  precautions  as  seemed  to  them  neces- 
sary or  possible  to  guard  the  frontier  from 
Passau  to  the  sea.  But  for  the  main  enterprise 
every  legionary,  despite  Antonius'  vaunt,  must  be 
called  to  the  war.  Antonius  himself,  with  Arrius 
Varus  at  his  right  hand,  led  the  advance  of  the 
invading    column.       If   the    orders    from    Syria    to 

'   See  above,  Chap.  I.  pp.  31-32.  -  Tac.  iii.  46. 


1 64    FLAVIAN  INVASION  OF  ITALY   ch.h 

await  Mucianus'  arrival  ever  reached  him,  these 
were  blandly  disregarded.  At  the  head  of  a  picked 
band  of  auxiliaries  and  part  of  the  cavalry  Antonius 
and  Varus  crossed  the  Pannonian  Alps  and  swept 
down  upon  Aquileia,  leaving  the  rest  of  the  army 
and  the  legionaries  to  follow  with  what  speed  they 
could.  They  seized  Aquileia,  and  pressed  at  once 
on  westwards.  The  strategy  of  instant  attack  had 
carried  the  day.  The  Flavians  would  seek  out  the 
enemy  to  annihilate  them,  if  it  might  be  so,  in 
battle.  While  their  supreme  leaders  in  far-off 
Eastern  lands  were  devising  schemes  of  "  exhaus- 
tion "  and  devious  strategies  of  war,  the  Danube 
army  flung  caution  to  the  winds  and  rushed  to  the 
attack.1 

Thus  the  Flavian  invasion  of  Italy  led  to  a 
struggle  once  more  between  the  old  enemies  of  the 
war  in  the  spring  of  the  year — the  Rhine  army  and 
the  Danube  army.  In  April  the  Danube  army  had 
had  but  a  part  of  its  strength  engaged,  but  it  had 
enjoyed  the  co-operation  of  the  Army  of  Italy,  upon 
which  indeed  the  chief  brunt  of  the  fighting  had 
fallen.  In  April  the  Rhine  army  had  been  the 
army  of  invasion.  In  October  the  Danube  army 
was  at  full  strength,  but  there  was  no  friendly  army 
marching  from  the  south  to  combine  with  it.  The 
relics  of  the  former  Army  of  Italy  were  either 
enlisted  in  its  ranks  or  scattered  to  the  four  winds, 
and  some  were  even  fighting  for  the  enemy.  The 
Rhine  army  had  become  the  army  defending  Italy 
against  invasion.      It  had  itself  suffered  in  strength, 

1  Tac.  iii.  5,  6. 


sec.  iv  FLAVIAN  INVASION  OF  ITALY    165 

thanks  to  Vitellius'  discharge  of  many  of  its  troops, 
and  in  efficiency,  thanks  to  six  months'  loose  living 
in  Italy.  But  it  had  added  on  to  it  all  available 
soldiers  in  Rome  and  Italy,  and  the  two  victorious 
generals  of  the  first  campaign,  Valens  and  Caecina, 
were  once  more  in  command  of  it.  The  strategy  of 
the  October  campaign  is  more  simple  and  there- 
fore less  interesting  than  that  of  April ;  but  the 
struggle  was  a  fiercer  one,  even  to  the  death.  In 
April  the  seeds  were  abundantly  sown  of  bitterness 
and  passion.  Now  the  late  harvest-time  had  come, 
and  the  furious  soldiery  were  the  reapers. 


§  4.    The  Strategy  of  the  Defence 

The  tidings  of  the  mutiny  in  Moesia  against  his 
authority  had  first  reached  Vitellius  at  Rome  in  a 
letter  from' Aponius  Saturninus,  governor  of  the  pro- 
vince. But  Saturninus  sent  word  only  that  the 
Third  legion  had  revolted,  and  flattering  friends  in 
the  Imperial  Court  made  light  of  the  whole  disturb- 
ance. Trustworthy  news  was  indeed  difficult  to 
obtain  as  soon  as  the  Danube  army  had  blocked 
the  road  over  the  Pannonian  Alps,  for  if  the  north- 
west wind  in  the  Adriatic  hampered  any  naval 
movements  on  part  of  the  enemy,  it  at  least  also 
hindered  the  coming  of  despatches  from  the  East 
overseas.  Vitellius,  however,  sent  to  Britain,  Spain, 
and  Germany  for  reinforcements.  But  his  sum- 
mons were  tardy,  nor  were  they  very  urgent. 
Vespasian's  spies  and  agents  were  everywhere,  and 
men    had    no    confidence    in    Vitellius'   chances    of 


1 66    FLAVIAN   INVASION  OF  ITALY   ch.h 

victory.  Three  of  the  four  legions  in  Britain  did 
indeed  send  detachments  in  time  to  take  part  in  the 
struggle.1  The  fear  of  troubles  in  the  country  and 
the  half-heartedness  of  its  governor  restrained  the 
rest.  The  legates  of  the  three  legions  in  Spain 
with  one  consent  held  all  their  troops  back.2  They 
would  not  help  to  prop  a  falling  cause.  In  Upper 
Germany  was  an  old  and  timorous  governor, 
Hordeonius  Flaccus,  whom  Vitellius  had  left  there 
to  guard  the  bank  of  the  Rhine.  But  he  by  this 
time  was  alarmed  at  the  signs  of  revolt  against 
Rome,  which  were  now  but  too  clear  lower  down 
the  river,  and  wisely  kept  the  few  troops  at  his 
disposal  in  the  province.  Only  the  one  legion  in 
Africa  3  and  the  provincial  troops  here  were  ready 
and  eager  to  fight  for  their  old  governor  Vitellius, 
but  their  legate,  Valerius  Festus,  wavered,  and 
they  did  not  cross  the  sea.  When,  therefore, 
the  instant  and  alarming  approach  of  the  danger 
soon  put  its  reality  beyond  question,  Vitellius  had 
to  confront  it  with  such  troops  as  he  had  in  Italy, 
together  with  the  timely,  if  weak,  reinforcements 
from  Britain.  Alike  the  West  and  the  South 
were  happy  to  be  spectators  of  the  combat,  and 
were  ready  to  applaud  the  victor  heartily  enough. 
Self-interest  in  such  a  civil  war  was  bound  to  be 
men's  ruling  instinct. 

But     no    such     reproach     of     timorousness     or 

1  Viz.    II.  Augusta,  IX.  Hispana,  XX.  Valeria  Victrix.     The  Fourteenth 
legion    was,    of  course,    hostile   to  Vitellius,   and   sent  no  aid.     Cf.    above, 

P-  133- 

2  Viz.  I.  Adjutrix,  VI.  Victrix,  X.  (Temina. 

3  Viz.  II T.  Augusta. 


sec.  .v  FLAVIAN   INVASION  OF  ITALY    167 

indifference  belongs  to  the  Emperor's  old   troops, 
the  soldiers  of  the  former  German  army.      They 
from    first    to    last    were    loyal    and    devoted    to 
Vitellius.      Nor,  indeed,  were  their  numbers  small. 
Their    old    generals,     Caecina    and    Valens,    were 
ordered    to  march  at   once   to   the   seat   of  war  in 
North   Italy  as  soon  as  the  general  revolt  of  the 
Danube    army    was    beyond    doubt.       Valens    was 
handicapped  by  illness,  and  unable  to  leave  Rome 
at  once.      But  Caecina,  after  taking  an  affectionate 
farewell    of  the   Emperor,    whom    he    intended    to 
betray,   marched   for  the  north   at   the  head  of  an 
imposing  column  of  infantry,  preceded  by  a  cavalry 
detachment.     In  the  van  of  the  foot  there  marched 
the   veterans   of   four  legions  —  the  First,   Fourth 
Macedonica,    Fifteenth  Primigenia,  and  Sixteenth. 
The  centre  consisted  of  the  Fifth  Alaudae  legion 
and  the  Twenty-second  Primigenia.      In  the  rear 
of  the   column   came  the   First   Italica  legion,  the 
Twenty  -  first   Rapax,   and    the    detachments    from 
the  three  legions  of  Britain — the  Second  Augusta, 
N  inth  H  ispana,  and  Twentieth  Valeria  Victrix.    The 
whole  number  can  hardly  have  fallen  short  of  forty 
thousand   legionaries.     Four  of  these   legions   had 
belonged  to  Valens'  old  command  in  Lower  Ger- 
many,1 and   he   sent   bidding  them  wait   upon   the 
march  until  he  could  overtake  them.     But  Caecina 
overruled    the    order,  and   he    had    every   possible 
military  justification  for  so  doing. 

The  writers  of  the  principate  of  Vespasian  loved 
to  paint  in  gloomy  colours  the  appearance  which  the 

1  Viz.  I.  Italica,  V.  Alaudae,  XV.  Primigenia,  XVI. 


1 68    FLAVIAN  INVASION  OF  ITALY   ch.  n 

soldiers  of  his  defeated  rival  presented  as  they 
marched  to  the  front  from  Rome  by  the  great  north 
road,  and  the  Roman  historian  of  a  later  age  had  no 
choice,  it  seems,  but  to  tell  their  story  over  again. 
If  these  writers  dared  to  ascribe  honourable  motives 
to  men,  Tacitus'  critical  faculty  was  at  once  aroused  ; 
but  the  mere  record  of  supposed  facts,  if  they 
were  picturesque,  excited  no  suspicion  in  his  mind. 
Yet  the  soldier's  trade  is  war,  and  many  an  army 
has  marched  out  to  defeat  as  cheerfully  and  made 
as  brave  a  show  as  have  the  coming  victors  in  the 
battle.  When  the  defeat  is  history  of  the  past  then 
the  curious  scribes  find  presage  for  it  in  the 
imagined  demeanour  of  the  troops  as  they  marched 
for  the  front.  The  change  in  the  bearing  of  the 
German  army,  says  the  Roman  writer,  was  indeed 
great  as  it  left  the  city.  There  was  no  strength  in 
their  bodies,  no  fire  in  their  hearts.  The  column 
rolled  heavily  along,  sluggish  and  scattered,  the 
weapons  dull  with  long  neglect,  the  very  horses 
moving  listlessly.  The  soldier,  grumbling  at  the 
sun,  the  dust,  the  weather,  shirked  his  duty,  and 
made  up  for  it  by  quarrelling  with  his  comrade.1  It 
is  a  sombre  military  picture.  Yet  this  is  the  army 
which,  after  a  march  of  three  hundred  miles  in 
summer  heat,  is  ready,  in  spite  of  the  desertion  of 
its  general,  to  march  some  thirty  miles  on  a  short 
October  day,  and,  without  taking  rest,  fight  strenu- 
ously in  hand-to-hand  combat  with  the  enemy  the 
whole  of  the  following  night  until  the  sun  rose  ! 
Moore's   Englishmen,   Blucher's    Prussians,    young 

1  Tac.  ii.  99  :  cf.  ii.  ior. 


sec.  iv  FLAVIAN  INVASION  OF  ITALY    169 

General  Bonaparte's  ragged  troops,  happy  in  their 
desperate  victories,  happier  still  in  their  generals, 
would  yet  find  comrades  to  their  heart  in  the 
Romans  of  Caecina's  leaderless  army — soldiers,  in 
truth,  for  all  their  grumbling,  and  no  craven,  no 
undisciplined,  mob  of  recreants. 

The  force  of  circumstances  compelled  the  Vitel- 
lians  to  adopt  the  strategical  defensive.  To  this 
Caecina  now  chose  to  add  the  tactical  defensive  as 
well.  He  contented  himself  with  the  occupation  of 
a  strong  position  guarding  the  line  from  Cremona 
to  the  sea.  The  thrust  of  the  assailants  would 
come  from  the  north-east.  Caecina  made  due  pre- 
parations to  repel  this  attack  upon  his  lines.  North 
of  the  main  stream  of  the  river  Po,  between  the 
marshes  of  Mantua  and  the  lagoons  of  Maestra,  two 
rivers  at  no  great  distance  apart  flow  through  the 
level  plain-.  The  first  of  these,  the  Tartaro,  is 
distant  some  five  miles,  at  an  average,  from  the 
Po  for  the  greater  part  of  its  course.  North  of 
this  again  the  splendid  stream  of  the  Adige,  which 
comes  foaming  down  from  the  Brenner  Pass  and 
sweeps  in  a  magnificent  semicircle  round  the 
fortress  of  Verona,  leaves  the  hills  at  that  city, 
and  for  the  rest  of  its  course  cuts  through  the 
marshy  level  to  the  sea  by  Chioggia  with  a  quieter 
Mood.  In  Roman  days  there  was  no  bridge  over 
the  Po  from  Hostilia  for  over  forty  miles  to  the 
east.  From  the  passage  of  the  river  at  Hostilia 
the  main  road  ran  north-east  through  Ateste  (Este), 
Patavium  (Padua),  and  Altinum  (Altino)  to  Aquileia. 
This  road  crossed  first  the  Tartaro  at  three  miles' 


170    FLAVIAN   INVASION  OF  ITALY   ch.h 

distance  from  Hostilia,  and  then  the  Adige  eight 
miles  from  its  crossing  of  the  Tartaro.  Here, 
guarding  the  passage  of  the  Adige,  is  to-day  the 
fortress  [city  of  Legnago,  one  of  the  four  great 
fortresses  of  the  "  Quadrilateral,"  famous  in  the 
times  of  the  Austrian  domination  of  Italy.1  In  the 
first  century  of  our  era  a  little  Roman  market-town, 
by  name  Forum  Alieni,  lay  on  the  site  of  the 
fortress  of  Legnago.2 

The  advance  guard  of  cavalry  was  sent  on  at 
once  to  occupy  Cremona,  and  it  was  followed  to 
the  town  by  the  First  Italica  legion  and  the 
Twenty-first  Rapax.  The  rest  of  the  large  army 
was  directed  straight  upon  Hostilia.  Caecina 
himself  turned  aside  to  visit  the  naval  station  at 
Ravenna,  where  lay  the  fleet  under  the  command 
of  its  prefect,  Sextus  Lucilius  Bassus.  Ravenna, 
indeed,  was  necessary  to  the  completeness  of  the 
defence.  For  although  the  invaders  had  no  fleet 
able  to  cope  with  the  Vitellian  fleet  at  Ravenna, 
yet  there  seems  to  have  been  a  coast  road  leaving 
the  main  road  from  Aquileia  to  Padua  some  miles 
short  of  this  latter  city,  and  striking  due  south  by 
Adria  to  Ravenna ;  and  if  the  enemy  had  chosen 
this  road,  the  defenders'  position  from  Cremona  to 
Hostilia  would  have  been  outflanked  and  turned. 
But  the  presence  of  a  friendly  fleet  at  Ravenna 
would   make   any  such  scheme  of  advance  far  too 

1  The  four  fortresses  were :  N.W.  Peschiera;  N.E.  Verona;  S.W.  Mantua; 
S.E.  Legnago. 

-  There  cannot  be  reasonable  doubt  as  to  this  identification.  The  sug- 
gestion that  Ferrara  is  Forum  Alieni  is  impossible  in  view  of  the  military- 
operations  of  this  campaign. 


sec.  iv  FLAVIAN   INVASION  OF  ITALY    171 

dangerous  to  adopt.  Thus  the  line  of  defence  ran 
from  Cremona  to  Hostilia,  from  Hostilia  to 
Ravenna.  And  thrown  forward  at  Forum  Alieni, 
guarding  the  bridge  over  the  Adige,  was  a  small 
outpost,  consisting  of  three  cohorts  of  auxiliary 
infantry  and  a  squadron  of  Gallic  cavalry,  the 
ala  II.  Gallorum  Sebosiana.  The  breadth  of 
the  river  and  the  single  bridge  should  make  its 
defence  possible  even  by  so  small  a  force  if  it 
observed  the  elementary  duties  of  an  outpost, 
constant  scouting  by  the  cavalry,  and  watchful 
pickets  at  night. 

The  first  blow  dealt  to  the  defence  fell  upon  this 
force  at  Legnago,  which  the  enemy  successfully 
rushed  at  dawn.  The  camp  was  completely  sur- 
prised, and  many  were  cut  down  before  they  could 
reach  their  weapons.  This  disgrace  was  the  result 
only  of  inexcusable  carelessness  on  part  of  the 
sentries,  or  of  the  commander,  if  no  sentries  were 
posted. 

Men  on  the  line  of  defence  cannot  sleep  at  ease  at  night, 
or  kindle  fires  to  warm  themselves.  The  night  is  the  time 
when  they  must  be  most  vigilant  and  wide  awake.  The  patrols 
on  the  picket  line  and  the  scouts  far  in  front  must  try  to  take  in 
everything.  However  tired  they  may  be  from  their  day's  work, 
at  night  they  must  not  even  allow  a  singing  insect  or  a  flying 
bird  to  pass  unnoticed.1 

The  outposts  of  the  camp  on  the  Adige,  if  there 
were  such,  incurred  great  dishonour.  But  the 
reverse    was  partially   redeemed    by   some    of  the 

1  Human  Bullets,  by  Lieut.  Sakurai,  p.  67.  The  Japanese  siege  of  Tort 
Arthur  puts  most  warfare,  ancient  and  modern,  to  shame. 


i;2    FLAVIAN  INVASION  OF  ITALY   ch.  n 

troops  who,  though  surprised,  held  their  ground 
long  enough  to  allow  the  destruction  of  the 
bridge,  and  by  this  means  checked  for  the  time 
the  pursuit  and  advance  of  the  invaders.  But 
when  the  news  of  this  reverse  reached  Caecina,  and 
his  outposts  also  came  into  touch  with  the  enemy's 
skirmishers,  who  presently  crossed  the  Adige,  the 
Vitellian  general  moved  his  main  camp  at  Hostilia 
a  few  miles  forward,  and  entrenched  a  strong  posi- 
tion on  the  northern  bank  of  the  river  Tartaro. 
To  its  rear,  therefore,  lay  this  river,  crossed  by  a 
bridge  ;  on  both  flanks  the  marshes  of  that  muddy 
stream  safely  guarded  it.1  With  its  front  only 
exposed  to  attack  and  this  strongly  fortified,  secured 
on  the  west  of  the  whole  line  by  Cremona,  on  the 
east  by  the  fleet  at  Ravenna,  the  camp  on  the 
Tartaro  might  surely  defy  the  assaults  of  the 
enemy.2 

While  Vitellius'  legions  lay  upon  the  Tartaro 
or  at  Cremona,  maintaining  a  strictly  defensive 
attitude,  the  Emperor's  second  general,  Fabius 
Valens,  in  due  course  left  Rome  and  moved  slowly 
northwards  along  the  northern  highway,  following 
in  the  steps  of  the  army  towards  Ariminum  (Rimini) 
which   lies  on  the  road  thirty  miles  south-east  of 

1  This  position  is  described  by  Tacitus  in  iii.  9  as  "inter  Hostiliam  et 
paludes  Tartari  fluminis."  From  this  it  would  appear  as  if  it  lay  on  the 
south  bank  of  the  Tartaro  and  that  the  "  flumen  "  at  its  rear  (loc.  cit.)  is  the 
Po.  But  when  in  iii.  14  the  troops  evacuate  it,  Tacitus  describes  the 
movement  as  "  relictis  castris,  abrupto  ponte  Hostiliam  rursus,  inde 
Cremonam  pergunt."  Therefore  the  bridge  broken  down  lay  between  the 
camp  and  Hostilia,  and  as  this  latter  place  lay  on  the  north  bank  of  the  Po, 
it  can  only  have  been  a  bridge  over  the  Tartaro.  As  this  bridge  lay  in  rear 
of  the  camp  (iii.  9),  the  camp  must  have  been  on  the  north  bank  of  this 
latter  river. 
fe&  -  Movements  of  the  Vitellians  :  cf.  Tac.  ii.  99,  100  ;  iii.  6,  9. 


sec.v    FLAVIAN  INVASION  OF  ITALY    173 

Ravenna,  and  one  hundred  miles  from  the  camp  on 
the  Tartaro.  In  his  train  he  seems  to  have  brought 
more  women  than  soldiers,  more  eunuchs  than 
legionaries.  The  march  of  so  soft  a  column  was 
naturally  slow.  Valens  possessed  military  ability  ; 
but,  at  a  time  when  there  was  crying  need  for  it  at 
the  front,  he  preferred  to  postpone  its  exercise  to 
the  gratification  of  an  unbridled  and  horrible  lust 
with  which  he  amused  himself  at  his  frequent  halts 
along  the  road.  The  army  at  Cremona  and  the 
Tartaro  could  look  for  small  reinforcements  to 
arrive  with  this  general  when  he  came,  and  might 
look  for  his  coming  for  long  in  vain.1  Caecina's 
army  was,  in  fact,  the  Emperor's  one  hope.  It  had 
marched  swiftly  to  the  north.  Now  it  lay  sullenly 
in  its  lines  waiting  attack  by  the  foe.  To  the 
strategy  of  attack,  chosen  by  Antonius  Primus  and 
the  Flavian  generals  of  the  Danube  army,  the 
Vitellians  opposed  a  strategy  of  defence.  Such  a 
strategy  may  be  executed  by  a  tactical  offensive  as 
well  as  by  a  tactical  defensive.  But  Caecina  chose 
the  latter,  and  destroyed  his  Emperor  by  his 
choice. 


§  5.    The  Strategies  compared 

The  comparison  between  the  advantages  of  the 
offensive  and  the  defensive  in  strategy  is  a  favourite 
theme    with    military    scientists,    and    Clausewitz's 

1  Tac.  iii.  40.  Tacitus'  account  of  Valens'  actual  movements  is  the 
vaguest  and  worst  possible.  See  below,  Note  F,  "  Valens'  March  to  the 
North." 


174    FLAVIAN   INVASION  OF  ITALY   ch.  n 

expressed  preference  for  the  latter  has  produced 
a  rich  crop  of  explanations,  interpretations,  even 
apologies.  Certain  advantages  of  the  attack  are 
indeed  evident,  and  are  as  visible  in  strategy  as 
they  are  in  tactics.  The  attacking  army  is  the 
more  likely  to  be  keen,  even  enthusiastic ;  its 
confidence  is  probably  greater ;  its  sense  of  daring 
stimulates  courage  and  at  the  same  time  enforces 
discipline.  "  The  greater  vitality  resides  in  the 
attack."1  The  invasion  of  a  hostile  country  especi- 
ally fires  the  imagination  and  stimulates  the  vigour 
of  the  soldier.  An  advantage  in  numbers  over 
the  enemy  is  indeed  greatly  to  be  desired  by  the 
invader.  Lines  of  communications  have  to  be 
guarded,  and  these  are  always  increasing  in  length. 
Fortresses  have  to  be  seized  and  garrisoned,  or 
blockaded  ;  important  strategical  points  have  to  be 
secured.  Supplies  are  obtained  with  greater  diffi- 
culty in  a  hostile  than  in  a  friendly  land.  Losses 
in  battle  are  made  good  less  easily.  Stragglers 
are  cut  off  and  cannot  rejoin.  The  sick  and 
wounded  cannot  be  left  to  the  care  of  the  in- 
habitants, but  must  be  tended  and  guarded  by  the 
invading  army.  To  supply  these  many  demands 
for  men,  and  yet  to  retain  a  force  strong  enough  to 
push  ever  deeper  and  deeper  into  the  heart  of  the 
enemy's  country,  and  able  to  defeat  the  foe  when 
these  choose  to  stand  their  ground  rather  than  to 
surrender  still  more  of  the  homeland  to  their  foe, — 
these  requirements  make  a  superiority  of  numbers 
on  part  of  the  invader  and  his  constant  reinforce- 

1   Von  der  Goltz. 


secv    FLAVIAN  INVASION  OF  ITALY    175 

merit  well-nigh  essential.1     "  Armies  acting  on  the 
offensive  melt  like  fresh  snow  in  spring."2 

It  is  thus  plain  that  the  offensive  is  only  possible  when  large- 
numbers  enable  a  leader  to  overcome  the  difficulties  it  offers, 
and  good  organisation  ensures  the  rapidity  necessary  for  carrying 
it  out.  But,  given  these,  there  can  be  no  doubt  of  its  advantage. 
The  moral  gain  is  great ;  the  soldier  feels  he  is  superior  to  his 
adversary  when  led  with  determination  against  him ;  and  this 
mental  attitude  leads  more  than  half-way  along  the  road  to 
victory.3 

There  are  few  generals  who  would  not  prefer 
to  conduct  rather  than  to  resist  an  invasion  ;  who 
would  not  choose  to  attack  rather  than  to  defend. 
Even  though,  on  the  actual  field  of  battle,  the 
lot  of  the  defender  may  seem  to  have  fallen  in 
pleasanter  places,  to  act  on  the  defensive  in  the 
whole  theatre  of  the  war  is  but  a  gloomy  busi- 
ness. There  are  indeed  no  generals  who  would 
not  desire  to  have  for  their  invasion  or  attack  an 
efficient  army  larger  than  that  of  the  defender. 
And  yet  the  brilliant  genius  of  the  commander 
has  in  times  past  more  than  made  good  even  an 
inferiority  of  numbers  possessed  by  the  invading 
army  when  it  crossed  the  enemy's  frontier.  Hannibal 
in  Italy,  Cromwell  in  Scotland,  Lee  and  Jackson  in 
Maryland,  took  no  account  of  the  general  rule  that 

1  Of  the  drain  in  men  suffered  by  an  invading  army  there  are  stock 
examples  in  the  military  text-books  :  Napoleon  in  1812  crossed  the  frontier 
of  Russia  with  over  three  and  a  half  million  troops :  at  Moscow  he  had  barely 
a  hundred  thousand;  in  1877,  450,000  Russians  crossed  the  Danube:  43,000 
arrived  outside  Constantinople;  the  Germans  in  1870  invaded  France  with 
three  and  a  half  million  men  :  six  weeks  later  they  had  only  half  the  number 
before  Paris.  Cf.  James,  Modern  Strategy,  p.  37  ;  Von  der  Goltz,  The  Nation 
in  Arms,  p.  258. 

-  Von  der  Goltz. 

;i  James,  Modern  Strategy,  p.  3S. 


176    FLAVIAN   INVASION  OF  ITALY   ch.  n 

the  invader  must  greatly  outnumber  the  field  army 
of  the  defender.1  But  the  world  sees  few  com- 
manders such  as  these.  And  even  so,  alike  in  Italy 
and  in  America,  the  brutal  weight  of  numbers  had 
in  the  end  its  revenge. 

The  Flavian  invading  army  was  far  from  en- 
joying any  such  superiority  of  numbers.  But 
the  leader  of  the  invasion,  Antonius  Primus, 
was  a  commander,  as  the  event  showed,  who 
won  the  admiration  and  devotion  of  his  troops  to 
a  high  degree,  and  inspired  them  with  his  own 
self-confidence  and  energy.  The  troops  which  he 
led  across  the  frontier  were  spirited  and  ready  for 
any  desperate  enterprise.  No  invading  army  could 
in  temper  have  been  better  fitted  for  its  work. 
Neither  did  the  general  lack  ability  or  a  keen  insight 
into  the  possibilities  of  a  military  situation.  The 
German  general  would  have  his  brother-officers  go 
to  school  of  Goethe's  Mephistopheles  : — 

An  Kiihnheit  wird  's  euch  auch  nicht  fehlen, 
Und  wenn  ihr  euch  nur  selbst  vertraut, 
Vertrauen  euch  die  andern  Seelen.'2 

It  was  just  this  supreme  confidence  in  himself 
which,  added  to  his  courage  and  "  dash,"  endeared 
Antonius  to  his  men,  and  seemed  to  make  him  an 
ideal  leader  of  an  invading  army. 

1  Of  course  this  rule  applies  only  when  the  hostile  nations  and  their 
armaments  are  of  similar  character  and  their  troops  display  similar  qualities. 
It  does  not  hold  good  in  contests  between  European  troops  and  most  Asiatic 
nations,  or  barbarians  ;  e.g.  an  Alexander  or  a  Caesar,  a  Cortez  or  a  Clive,  is 
not  an  example  to  the  contrary  of  this  rule.  Nor  does  it  apply  to  mere  raids 
across  the  frontier  where  no  fighting  or  occupation  of  the  enemy's  land  is 
intended. 

2  I  take  the  quotation  from  Von  der  Goltz,  The  Nation  in  Arms,  p.  388. 


sec.v    FLAVIAN   INVASION  OF  ITALY   177 

The  invasion  of  Italy  by  the  Danube  army  was 
clearly  a  daring,  even  a  perilous,  strategy  to  adopt. 
But  when  invasion  is  conducted  by  such  a  general, 
when  in  itself  it  has  such  advantages  over  the 
defence,  surely  the  plan  might  seem  justified  ? 

In  the  zeal  to  inflict  injury  upon  the  enemy,  a  resolution  must 
not  aim  at  the  unattainable,  though  it  should  venture  to  go  to 
the  extreme  limit  of  the  permissible.  In  war,  nothing  rational 
must  be  considered  impossible  as  long  as  it  has  not  been  tested  ; 
and  we  may  dare  everything  we  believe  we  can  carry  out.1 

"  In  war,  nothing  rational  must  be  considered 
impossible."  The  Flavian  invasion  of  Italy  as  con- 
ducted in  the  autumn  of  a.d.  69  had  no  rational 
prospect  of  success.  It  ought  to  have  been  hurled 
back  in  ruinous  defeat  and  panic-stricken  disgrace 
over  the  Pannonian  Alps. 

For  the  Flavian  forces  struggled  over  the 
mountains  down  to  the  plain  of  North-east  Italy  in 
widely  separated  detachments.  Those  who  arrived 
first  flung  themselves  forward  regardless  of  any 
co-operation  with  those  who  were  to  come  after 
them.  Here  came  a  band  of  auxiliaries  ;  then,  after 
a  gap,  came  a  legion  ;  then,  after  a  pause  of  some 
days,  other  legions.  The  whole  conduct  of  the 
enterprise  in  face  of  a  foe  who  knew  how  to  use  his 
opportunities  was  mad.  Even  when  concentrated, 
the  invading  army  was  hardly  equal  in  numbers  to 
the  Vitellian  forces  on  the  river.  If  the  defenders 
had  taken  the  tactical  offensive  as  soon  as  the  foe 
appeared  in   the  plain  south  and  east  of  Aquileia, 

1  Von  der  Goltz,  op.  cil.  p.  383. 

N 


178    FLAVIAN   INVASION  OF  ITALY   ch.h 

there  was  no  hope  for  the  invaders  but  that 
they  would  have  been  swallowed  up  piecemeal. 
The  legions  which  came  tardily  down  over  the 
mountains,  because  encouraged  by  the  unopposed 
progress  of  Antonius  and  the  advance  guard, 
would  have  hesitated  longer  had  a  few  frightened 
fugitives  come  speeding  back  to  them  for  refuge 
from  the  pursuit  of  a  victorious  foe.  Antonius' 
numbers  for  the  first  few  days  after  he  had  come 
into  touch  with  the  enemy  were  contemptible. 
But  the  legions  of  Vitellius  lay  passively  in  their 
entrenchments,  looking  dully  at  the  stream  of 
the  enemy  which  flowed  past  their  front  in  inter- 
mittent waves.  The  procession  of  the  Flavian 
troops  passed  gaily  along  day  after  day,  always 
exposing  their  flank.  They  seized  town  after  town, 
fortress  after  fortress,  a  few  miles  away  from  the 
torpid  Vitellian  army.  They  concentrated  undis- 
turbed, unopposed,  in  high  spirits,  and  at  leisure,  at 
the  powerful  fortress  of  Verona.  Even  after  this, 
their  whole  line  of  communications  lay  open  to 
attack.  If  the  enemy  fell  on  them  they  must  of 
necessity  have  formed  front  to  a  flank  outside 
Verona's  walls,  and  their  defeat  would  have  meant 
annihilation  for  them.  During  the  weary  days  while 
they  were  mustering,  a  single  victory  (and  it  was 
impossible  that  the  Vitellians  should  not  gain  it) 
would  have  ruined  the  whole  scheme  of  invasion — 
at  least  until  Mucianus  arrived,  and  would  have 
given  even  him  much  cause  for  thought.  So  sensi- 
tive is  the  barometer  of  men's  inclinations  in  days  of 
civil  war  to  the  storms  of  failure  or  the  sunshine  of 


sec.v    FLAVIAN   INVASION  OF  ITALY    179 

success  that  the  mere  rout  of  the  advance  guard  of 
the  Danube  army  might  possibly  have  wrecked 
Vespasian's  whole  enterprise.  This,  perhaps,  would 
have  been  too  much  to  expect.  Hardrada  is  routed 
at  Stamford  Bridge,  but  the  Norman  still  lands  at* 
Pevensey  and  Harold  falls  at  Senlac.  But,  even 
so,  the  Northman  did  not  invade  on  Duke  William's 
behalf.  At  least  the  frontier  of  North-east  Italy 
might  have  been  securely  guarded  had  Vitellius' 
men  quitted  their  stagnant  lines  and  advanced  to 
battle.  Mucianus'  heart  might  well  have  failed  him, 
or  his  discouraged  troops  have  refused  to  follow 
him ;  and  then  the  victorious  army  might  have 
marched  south  again,  as  did  Harold's  men,  and  have 
defied  the  Eastern  invader  to  make  good  his  landing 
from  overseas,  or,  if  he  landed,  have  fought  him  to 
the  death.  Antonius'  invasion  would  have  remained 
on  record  but  as  a  monument  of  rash  folly.  The 
real  struggle  for  Empire  might  once  more  have 
been  waged  outside  the  walls  of  Rome.  And  who 
then  would  with  surety  forecast  the  victory  for 
Vespasian  ? 

It  was  treachery  which  defeated  Vitellius'  soldiers 
and  their  Emperor,  not  the  strategy  of  the  enemy. 
Two  men  were  faithless  to  the  cause.  Caecina,  the 
general,  of  subtilty  kept  his  splendid  army  idle  in 
its  entrenchments.  Lucilius  Bassus,  prefect  of  the 
fleet  at  Ravenna,  sought  to  entice  the  sailors  from 
their  loyalty.  Both  men's  motives  were  despicable. 
However  unworthy  their  Emperor,  it  was  not  for 
these  men  who  had  received  honours  at  his  hands  to 
plot  craftily  against  him.      Happy  indeed  was  Rome 


180    FLAVIAN  INVASION  OF  ITALY   ch.h 

and  fortunate  the  Empire  which  lost  Vitellius  to 
gain  Vespasian  as  Emperor  ;  but  the  traitor's  taint  is 
not  therefore  sweet-scented.  Bassus,  a  mere  cavalry 
captain,  had  hoped  for  the  prefecture  of  the  Guards 
at  Rome.  Preferred  to  the  lower  post  of  Admiral  of 
the  Fleet,  he  sought  in  a  dastardly  perfidy  the  remedy 
for  his  disappointment.  Caecina,  vain  and  ambitious, 
ever  craving  popularity,  secretly  resentful  at  the 
greater  fame  which  his  colleague  had  won  in  the 
recent  campaign,  believed  that  Valens  enjoyed  the 
greater  share  of  Vitellius'  esteem.  The  two  had  in- 
dulged in  envious  rivalry  of  pomp,  parade,  and  self- 
advertisement  during  the  last  few  months  at  Rome. 
Caecina  had  yielded  himself  a  slave  to  indolence  and 
luxury.  His  ambition  waxed  as  his  self-control 
waned.  Envy  preyed  upon  him,  jealousy  mastered 
him.  The  man  brooded  over  his  wrongs  until,  as 
often  happens,  he  lost  his  sense  of  honour.  He 
who  first  should  make  terms  secretly  with  the 
Flavians  would  doubtless  receive  the  greater  rewards 
at  their  hands.  Without  shame  and  without  scruple 
Caecina  and  Bassus  conspired  together  at  Padua  to 
bring  fleet  and  army  over  to  the  enemy.1 

There  was  little  difficulty  with  the  fleet.  The 
sailors,  loyal  to  Otho,  had  accepted  Vitellius'  rule 
with  chagrin.  Many  of  them  were  drawn  from  the 
provinces  of  Dalmatia  and  Pannonia,  now  in  arms 
for  Vespasian.  In  a  night  the  mutiny  was  accom- 
plished. The  Ravenna  fleet  declared  for  the 
Flavians  with  Bassus'  full  approval,  and  chose 
Cornelius    Fuscus    as    their    new    prefect.      Bassus 

1  Tac.  ii.  56,  92,  93,  100,  101. 


sec.v    FLAVIAN  INVASION  OF  ITALY    181 

gained  little  by  his  treachery  save  the  loss  of  his 
command  and  a  short  stay  in  prison  at  Adria. 
Later  he  was  employed  by  the  victors  again  as 
cavalry  captain  on  petty  operations  in  Campania. 
His  ambition  had  sorely  overleapt  itself,  and  he  had 
this  excuse  for  his  treachery  that  he  had  failed  to 
profit  by  it.1  But  Caecina  had  a  harder  task  with 
the  army.  The  general  could  urge  with  truth  that 
the  mutiny  of  the  fleet  had  made  their  position  on 
the  Tartaro  untenable,  and  that  Vitellius'  cause  had 
suffered  a  grievous  blow.  With  some  success  he 
laid  insidious  siege  to  the  loyalty  of  the  centurions 
and  a  few  of  the  soldiers,  who  at  last  allowed  his 
arguments  and  their  fears  to  prevail  over  their 
fidelity.  But  though  for  a  moment's  space  Vespasian 
was  proclaimed  in  the  Vitellian  camp,  the  bulk  of  the 
soldiers  and  the  higher  officers  held  firmly  by  their 
Emperor.  "  What  did  the  miserable  fleet  count," 
they  passionately  asked,  "in  comparison  with  eight 
legions  ?  Were  they,  the  proud,  victorious  army  of 
Germany,  to  be  handed  over  to  an  Antonius  Primus 
as  so  many  cattle,  so  many  slaves,  for  sale  ?  Caecina 
and  Bassus  might  seek  to  rob  the  army  of  its 
Emperor ;  but  how  could  they,  soldiers  who  had  in 
this  campaign  tasted  nothing  as  yet  of  bloodshed, 
make  answer  to  their  enemies  or  look  them  in  the 
face  when  asked  tauntingly  of  their  victories  or 
defeats  ?  " 

The  Fifth  legion  Alaudae  overthrew  Vespasian's 
standards  with  bitter  indignation  ;  the  others  swiftly 
followed  suit.     Caecina  was  put  by  the  men  in  irons. 

1  Tac.  ii.  101  ;  iii.  12  ;  iv.  3. 


iS2    FLAVIAN   INVASION  OF  ITALY   ch.ii 

The  army  on  the  Tartaro  was  loyal  to  its  Emperor. 
Its  inactivity,  its  hopeless  reliance  upon  the  tactical 
defensive,  had  been  forced  on  it  by  guile  and  by 
treachery.  Its  opportunity  for  avenging  the  folly 
of  the  invader  on  his  head  was  indeed  lost.  The 
mutiny  of  the  fleet  must  force  them  to  retire  from 
their  useless  lines.  Their  general,  too,  was  lost  to 
them.  But  honour  was  not  lost,  and  they  were 
still  eight  legions  with  arms  in  their  hands  and 
burning  anger  in  their  hearts.1 

The  strategy  of  invasion  should  have  been  met 
and  defeated  by  the  tactics  of  offence  employed 
by  the  strategical  defensive.  Success  crowned 
Antonius'  rashness.  If  he  had  reason  to  hope  that 
Caecina  would  play  the  traitor,  the  prize  was  worth 
the  risk,  and  Antonius  who  ventured  it  was  truly  a 
great  general.  But  if,  when  he  descended  from  the 
mountains  upon  Aquileia,  he  knew  nothing  of  the 
temper  of  the  sailors  at  Ravenna  or  of  Caecina's 
meditated  treachery  (and  the  scanty  evidence  points 
to  this  conclusion),  he  cannot  be  acquitted  of  an 
impulsive  rashness  which  properly  deserved  defeat. 
For  in  this  case  he  was  presuming  upon  a  degree 
of  sloth,  ignorance,  and  incapacity  on  part  of  the 
enemy  which  it  was  incredible  that  they  should 
display.  Judged  by  results,  the  Flavian  strategy 
of  invasion  was  a  notable  success.  Yet  there  are 
victories  in  war  as  in  games  which  rightly  give 
small  satisfaction  to  the  victors.  Every  general 
makes  mistakes ;  but  it  is  not  scientific  strategy 
which  is  built  upon  nothing  but  the  expectation  of 

1  Tac.  ii.  ioi  ;  iii.  13,  14. 


sec.v    FLAVIAN   INVASION  OF  ITALY    183 

the  foe's  mistakes.  When  Antonius  rushed  to 
knock  his  head  against  the  enemy's  wall,  he  deserved 
a  headache  rather  than  the  discovery  that  the  wall 
was  lath  and  plaster. 

And  on  the  other  side,  Valens,  so  far  as  he  was 
able,  completed  the  ruin  begun  by  the  traitor 
general  and  admiral.  When  the  Vitellian  army 
quitted  the  position  on  the  Tartaro  and  concentrated, 
ably  enough,  at  Cremona,  it  lacked  nothing  even 
then  but  a  general.  One  fatal  error  gave  the 
hard-won  victory  to  the  foe.  From  this  Valens' 
ripe  wisdom  would  surely  have  saved  the  army. 
But  Valens  was  not  at  Cremona.  While  still  on  his 
slow  march  to  the  north,  he  received  the  news  of 
the  mutiny  of  the  fleet.  He  was  then  already 
probably  north  of  Ariminum,  not  very  far  from 
Ravenna,  to  which  city  he  was  marching  to  co- 
operate with  the  fleet.1  The  harbour  now  was 
hostile,  and  he  could  not  venture  to  march  forward. 
His  disorderly  rabble  was  not  an  army.  But  still 
he  might  have  turned  aside  by  a  cross  road  to  the 
main  road  from  Ariminum  to  Bononia,  and,  if  he 
travelled  with  great  speed,  have  perhaps  reached 
Caecina  on  the  Tartaro  in  time  to  dissuade  that 
wavering  general  from  his  treachery.  Bassus' 
action  seems  to  have  befallen  before  Caecina 
expected  it,  and  the  general  was  still  hesitating. 
Or,  Valens  might  certainly  have  reached  Cremona, 
had  he  pushed  on  fast,  before  the  critical  battle  was 
fought.  But  Valens  never  at  any  time  in  his  career 
showed  resolution   or  rapidity   of  movement.      He 

1  See  Note  F,  "  Valens'  March  to  the  North." 


1 84    FLAVIAN  INVASION  OF  ITALY   ch... 

halted  his  column  in  miserable  indecision,  and  finally 
sent  to  Rome  begging  for  reinforcements.  A  poor 
little  force  was  sent  him,  of  no  avail  for  any  useful 
purpose.  These  he  despatched  to  Ariminum.  He 
himself  gave  up  all  thoughts  of  taking  any  part  in 
the  campaign  on  the  Po,  and  crossed  the  Apennines 
to  reach  the  Arno  valley  and  the  sea  at  Pisa.  Great 
schemes  of  future  warring  at  the  head  of  the  forces 
of  Gaul  and  Germany  floated  through  his  mind,  and 
he  took  ship  at  Pisa  for  the  coast  of  Gaul.  This 
plan  ended  lamentably.  At  Monaco  on  his  voyage 
he  heard  that  all  the  coasts  had  declared  for 
Vespasian.  Most  of  his  comrades  promptly  followed 
the  example  set  them,  and  Valens  himself  with  but 
ten  companions,  setting  out  upon  the  unfriendly  sea, 
was  driven  by  a  gale  upon  the  Stoichades  islands.1 
There  he  was  captured  and  sent  as  a  prisoner  of 
war  to  the  Umbrian  hill-city  of  Urbinum.- 

Long  ere  this,  the  Vitellians  in  North  Italy 
had  fought  their  final  battle  there  without  a 
general  to  lead  them.  When  so  many  were  the 
blunders  of  the  defence,  when  the  defending  army 
had  for  its  generals  a  Caecina  and  a  Valens,  the 
traitor  and  the  faint-heart,  there  is  small  wonder 
that  the  strategy  of  offence  and  invasion  prevailed 
against  it.  A  combination  by  the  Vitellians  of  the 
tactical  offensive  with  the  strategical  defensive  must 
have  given  them,  at  first  at  least,  the  victory. 
The  Imperial  army  was  wilfully  sacrificed  by  in- 
competence and  treachery. 

1  The  lies  d'Hyeres,  off  Toulon  rather  than  (as  Tacitus  describes  them)  off 
Marseilles.  2  Tac.  iii.  41-43  ;  cf.  Note  F. 


sec.  vi  FLAVIAN   INVASION  OF  ITALY    185 

§  6.    The  Second  "  Battle  of  Bedriactim  " 

A.  The  Flavian  Advance  to  Verona. — Antonius 
led  the  van  of  his  invading  army  rapidly  down  upon 
North-east  Italy.  The  towns  which  lay  upon  the 
roads  leading  west  and  south-west  from  Aquileia, 
far  from  resisting  him,  even  welcomed  his  coming. 
Opitergium,  the  modern  Oderzo,  forty-five  miles 
from  Aquileia,  and  Altinum  (Altino)  to  the  south  of 
it,  fifty-five  miles  from  Aquileia,  opened  their  gates 
to  him.  Altinum  was  a  position  of  importance,  as 
a  few  miles  to  the  west  of  it  the  road  from  Ravenna 
which  crossed  the  Po  at  Adria  joined  the  road  from 
Padua  and  Hostilia  to  Aquileia.  It  was  Antonius' 
intention  to  press  rapidly  forward  down  this  latter 
road.  While  then  the  fleet  at  Ravenna  belonged 
to  the  enemy,  there  was  a  danger  lest  a  hostile 
force  should  advance  by  the  former  road  from  the 
seaport  and  throw  itself  astride  of  the  road  by  which 
the  Flavian  vanguard  had  come  after  these  had 
passed  by.  In  this  event,  their  van  would  be  cut 
off  from  the  rest  of  their  army  and  from  their 
communications.  Its  position  would  be  perilous. 
Antonius  knew  nothing  as  yet  of  the  intention  of 
the  fleet  to  desert  from  the  enemy.  He  therefore 
left  a  garrison  in  Altinum  to  guard  the  communica- 
tions, and  pressed  on  with  the  main  body  to  Padua, 
and  beyond  it  to  Ateste,  the  modern  Este.  Both 
of  these  towns  admitted  him.  At  Ateste  he  was 
but  some  seventeen  miles  from  the  enemy's  outpost 
which  was  guarding  the  bridge  over  the  Adige  at 
Forum  Alieni.      This   outpost,  as  has  been   seen, 


1 86    FLAVIAN   INVASION  OF  ITALY   ch.h 

was  successfully  rushed  at  dawn  and  dispersed. 
But  it  found  time  to  destroy  the  bridge  and  check 
the  pursuit.1  4. 

While  Caecina  the  enemy's  general,  upon  news 
of  this  reverse,  moved  his  army  from  Hostilia  forward 
to  the  camp  on  the  Tartaro  and  entrenched  himself 
there,  Antonius  lay  quiet  at  Padua.  The  first  wave 
of  the  invasion  had  swept  forward  eighty  miles,  and 
had  for  the  moment  spent  its  force.  But  the  news 
of  the  advance  and  of  the  success  gained  on  the 
Adige  was  quickly  carried  back  to  Poetovio,  and 
the  two  legions  of  Pannonia,  the  Seventh  and  the 
Thirteenth,  started  out  forthwith  with  much  greater 
confidence,  and  joined  Antonius  in  due  course  at 
Padua,  meeting  with  no  difficulty  on  the  way. 
Here  they  were  given  some  days'  rest  to  recover 
from  the  march. 

Antonius  and  the  other  Flavian  leaders  had  now 
to  consider  their  next  movements.  The  enemy  lay 
quiet  in  their  lines  beyond  the  Adige,  and  showed 
no  sign  of  advancing  themselves  to  the  attack. 
This  left  the  initiative  still  comfortably  in  the 
invader's  keeping.  To  assault  the  position  on  the 
Tartaro,  however,  seemed  most  unwise.  It  was  an 
exceedingly  strong  one,  and,  moreover,  was  held 
by  a  veteran  army  which  was  far  stronger  in 
numbers  than  were  the  troops  who  had  up  to 
that  time  arrived  at  Padua.  It  was  also  becom- 
ing evident  that  Caecina's  heart  was  not  towards 
Vitellius  any  longer,  and  it  was  far  better  to 
give    him     time    for    quiet     thought    and    a    fur- 

1   See  above,  p.  171. 


sec. vi  FLAVIAN   INVASION  OF  ITALY    187 

tive  correspondence  with  the  Flavians  than  to 
hurry  him  into  a  loyal  resistance  by  an  ill-timed 
attack.  And  two  other  thoughts  had  weight  with 
Antonius.  The  strength  of  his  force  lay  largely 
in  its  cavalry.  If  Caecina  should  make  up  his  mind 
after  all  to  advance  his  standards  against  the  invader, 
Antonius  needed  a  battle-ground  where  he  could 
deploy  all  his  horse  to  best  advantage.  There  was, 
indeed,  plenty  of  level  ground  round  Padua  at  the 
foot  of  the  green  Euganean  Hills.  But  these  hills 
might  screen  the  advance  of  the  enemy,  and,  besides, 
to  stay  at  Padua  would  seem  a  confession  of  fear. 
The  advance  must  continue.  The  second  con- 
sideration which  influenced  the  Flavian  general  was 
the  very  lively  dread  lest  large  reinforcements 
should  reach  the  enemy  from  Germany  by  way  of  the 
Brenner  Pass.  Raetia,  on  the  northern  side  of  the 
Alps  where  this  pass  began,  was  loyal  to  Vitellius, 
and  the  Flavian  forces  already  sent  to  the  line  of  the 
Inn  might  indeed  defend  Noricum  against  an  attack 
from  Raetia,  but  were  too  weak  to  intercept  the 
coming  of  reinforcements  to  Italy  over  the  Brenner.1 
If  there  were  an  army  on  the  march  by  this  road — 
and  Vitellius  had  certainly  sent  to  Germany  for 
help — and  if  it  reached  Hostilia,  little  would  be 
heard  henceforward  of  Caecina's  wavering.  One 
town,  however,  guarded  the  issue  of  the  Brenner 
Pass  from  the  mountains,  the  strong  and  important 
city  of  Verona,  across  the  river  Adige,  some  forty 
miles  up  stream  above  Legnago.  To  the  south  of 
Verona  was  a  plain  suited  in  every  way  for  cavalry 

1  See  above,  p.  162. 


188    FLAVIAN   INVASION  OF  ITALY   ch.h 

manoeuvres.  If  he  seized  this  town,  Antonius 
would  have  not  only  the  prestige  of  holding  one  of 
the  greatest  cities  of  North  Italy,  but  also  a  suit- 
able base  for  his  future  operations.  Moreover,  at 
Verona  he  would  be  able  to  intercept  any  German 
reinforcements  coming  to  the  enemy  from  the 
north ;  and  Verona  was  half-  way  from  Padua  to 
Cremona.  Poised  at  this  central  point  he  could 
swoop  down  upon  any  part  of  the  enemy's  long 
line  of  defence,  which  reached  from  Cremona  to 
Hostilia  and  beyond. 

As  soon,  therefore,  as  his  men  were  ready  again 
for  marching  Antonius  marched  from  Padua  to 
Verona.  On  the  road  lay  the  town  of  Vicetia,  the 
modern  Vicenza  under  the  Monti  Berici.  To-day 
this  busy  city  numbers  nearly  half  as  many  in- 
habitants as  Verona  herself;  but  in  Roman  days 
it  was  a  small  place  of  little  importance.  It  gave, 
however,  peculiar  pleasure  to  the  Flavians  to  take 
possession  of  the  town,  since  it  happened  to  be 
Caecina's  birthplace,  and  they  gleefully  thought 
how  the  news  would  be  spread  abroad  that  "  the 
enemy's  general  had  been  despoiled  of  his  native 
land."1  The  town,  however,  was  of  no  military 
importance,  and  Caecina,  who  was  pondering  matters 
of  greater  moment,  was  not  likely  to  be  greatly 
moved  by  so  trivial  an  incident.  From  Vicetia  the 
army  moved  forward  to  Verona,  thirty  miles  away, 
and  the  town  gladly  received  them.  Entrenchments 
were  thrown  up  and  a  halt  of  some  time  was 
called. 

1   "  Patriam  hostium  duci  ereptam,''  Tac.  iii.  8. 


sec. vi  FLAVIAN  INVASION  OF  ITALY    189 

From  this  position,  says  the  Roman  historian, 
Caecina  could  doubtless  have  hurled  them,  had  he 
so  chosen.  But  the  traitor  kept  his  army  quiet, 
and  soon  the  coming  of  the  Moesian  army  relieved 
Antonius  from  the  more  instant  peril  of  defeat 
First  the  Seventh  Claudia  legion  arrived  under 
command  of  a  stalwart  tribune,  Vipstanus  Messalla, 
and  accompanied  by  the  Governor  of  Moesia  him- 
self, Aponius  Saturninus.  The  tribune  had  taken 
direct  command  himself  of  the  legion,  since  its 
legate  was  at  this  time  fleeing  for  his  life  over  the 
Balkans  from  the  private  vengeance  of  the  governor. 
Messalla,  says  Tacitus,  "  was  the  one  and  only  man 
who  brought  an  honest  heart  to  that  war."  Later 
he  wrote  the  history  of  it,  and  Tacitus  has  certainly 
thanked  him  gracefully  for  the  use  which  he  himself 
made  of  his  history. 

After  the  Seventh  legion  came  the  other  two, 
the  Third  and  Eighth.  By  this  time  at  last  the 
Flavian  leaders  had  assembled  a  truly  powerful 
army  at  Verona.  It  cannot  have  been  greatly,  if  at 
all,  weaker  than  the  enemy's  force  on  the  Tartaro. 
One  thing  was  lacking  to  its  strength,  namely,  per- 
fect discipline.  And  it  enjoyed  too  many  possible 
generals.  The  one  evil  remedied  the  other.  The 
turbulence  of  the  men  scared  away  the  governors 
of  Moesia  and  Pannonia  from  the  camp.  Antonius 
was  left  in  undisputed  command,  so  far  as  any  man 
had  command  over  the  unruly  spirits  of  the  soldiers, 
and  he  could  rely  on  Arrius  Varus  to  help  him.1 

B.    The  Race  for  Cremona. — Then  to  the  Flavian 

1  Tac.  iii.  6-1 1. 


i9o    FLAVIAN  INVASION  OF  ITALY 

camp  at  Verona  there  came  exciting  news.  Antonius 
had  for  some  time  past  been  in  correspondence 
with  Caecina.  The  latter  had  doubtless  assured 
the  Flavian  commander  that  he  could  induce  his 
army  to  follow  his  example  in  renouncing  its  allegi- 
ance to  Vitellius.  Caecina  tried  and  failed.  His 
indignant  troops  put  him  in  chains  ;  but  they  wisely 
judged  their  position  on  the  Tartaro  to  be  no  longer 
tenable.  The  Ravenna  fleet  threatened  their  rear ; 
the  Flavian  army  at  Verona,  within  a  few  miles  of 
their  front,  was  now  strong,  and  might  advance 
against  them  when  they  were  in  confusion  and  had 
no  leader  to  inspirit  their  defence.  To  the  west 
at  Cremona,  however,  lay  two  legions  and  a  force 
of  cavalry  of  their  comrades.  The  troops  resolved 
to  march  to  join  them  at  once.  The  direct  road  to 
Cremona  from  Hostilia  lay  north  of  the  Po,  and  ran 
by  Mantua  and  Bedriacum.  By  this  route  Cremona 
was  not  quite  sixty  miles  away  from  Hostilia.  But 
if  they  marched  by  this  route  they  would  be  peril- 
ously near  the  Flavians  at  Verona,  and  their  right 
flank  would  be  exposed  to  attack  by  these  at  any 
point  along  the  road.  If  the  Verona  army  marched 
down  upon  their  column  (and  Antonius  would 
certainly  not  miss  such  an  opportunity)  they  would 
have  hurriedly  to  deploy  into  order  of  battle  by  the 
right  with  the  enemy's  cavalry  rushing  down  upon 
them,  and  they  would  fight  with  flanks  unprotected, 
without  cavalry  of  their  own,  and  with  their  backs 
to  the  broad  and  unfordable  stream  of  the  Po. 
Though  they  now  lacked  Caecina's  guidance,  the 
legate  of  the   Fifth  legion  and  the  Camp  Prefect, 


sec. vi  FLAVIAN  INVASION  OF  ITALY    191 

to  whom  the  soldiers  had  entrusted  the  command, 
were  able  to  realise  that  such  a  position  must 
mean  ruinous  defeat.  The  march  by  Mantua  was 
impossible.  Their  only  alternative  was  to  cross  to 
the  south  of  the  Po,  and  march  by  Mutina,  Parma, 
and  (possibly)  Placentia.  It  was  a  terribly  long 
detour  and  a  long  and  trying  march  of  a  hundred  and 
ten  miles.1  But  at  least  their  right  flank  would  be 
safe,  and  distance  would  be  their  only  enemy.  They 
must  trust  to  speed  of  marching  to  bring  them  to 
Cremona  in  time.  And,  in  truth,  never  did  troops 
merit  better  the  praise  which  belongs  to  the  Roman 
soldier  than  do  these  betrayed  and  leaderless  men 
of  the  Vitellian  army.  Placed  in  so  disheartening 
and  critical  a  position,  the  modern  European  soldier 
might  but  too  easily  lose  heart  entirely,  or  lack  the 
initiative  and  foresight  which  the  Roman  at  this 
time  displayed.  There  have  been  few  troops  in  the 
world  to  equal  those  of  Rome. 

The  Vitellians  abandoned  their  camp  on  the 
Tartaro  and  fell  back  to  Hostilia.  They  crossed 
the  Po  there,  broke  down  the  bridge  behind 
them,2  and  disappeared  entirely  from  the  range  and 
ken  of  Antonius'  scouts.  The  crisis  of  the  cam- 
paign evidently  was  hard  at  hand.3 

1  Distances  :  Hostilia  to  Mutina,  30  miles  ;  Mutina  to  Parma,  30  ;  Parma 
to  Placentia,  35  ;  Placentia  to  Cremona,  15. 

2  An  obvious  inference  from  military  requirements.  The  broken  bridge  in 
Tac.  iii.  14  is,  of  course,  that  over  the  Tartaro. 

3  Tacitus  has  not  the  least  interest  in  the  march.  He  merely  remarks, 
"Abrupto  ponte  Hostiliam  rursus,  inde  Cremonam  pergunt."  That  they 
marched  vid  Parma,  as  Mommsen  suggests,  is  shown  by  the  fact  that 
Antonius,  on  their  departure,  makes  a  forced  march  of  two  days  from 
Verona  to  Bedriacum  (some  thirty-three  miles),  but  never  gets  in  touch  with 
the  Vitellian  retreating  army,  and  he  arrives  outside  Cremona  after  a  day's 
fighting  as  soon  as  they  do.     And  these  have  marched  with  desperate  haste, 


192    FLAVIAN   INVASION  OF  ITALY   ch.h 

Antonius  understood  at  once  the  meaning  of 
the  empty  camp.  There  was  now  to  be  a  race  for 
Cremona  and  its  small  garrison  between  his  own 
and  the  Hostilia  army.  The  two  divisions  of  the 
enemy  must  not  be  allowed  to  join.  Cremona  must 
be  taken  before  the  eastern  division  could  march  to 
it,  and  before  it  could  be  reached  by  Valens,  who 
surely  (so  judged  the  soldier  Antonius)  would  hasten 
to  join  the  enemy's  army  on  hearing  of  Caecina's 
betrayal.  Vitellian  reinforcements,  too,  were  said 
to  be  mustering  from  Britain,  Gaul,  and  Germany. 
He  had  the  shorter  march  by  fifty  miles.  But  the 
garrison  of  Cremona  was  strong  enough  to  offer 
a  sturdy  resistance.  Without  hesitation  or  delay, 
Antonius  led  his  entire  force  south  from  Verona. 
The  race  for  Cremona  had  begun. 

By  the  evening  of  the  second  day  Antonius  and 
his  army  had  marched  thirty-five  miles  from  Verona 
to  Bedriacum,  where  the  road  from  Mantua  to 
Cremona  joined  their  own.  Cremona  itself  lay 
twenty-two  miles  away  by  the  Via  Postumia  to  the 
west.1  There  had  been  no  sign  of  the  enemy  on 
the  march.  Evidently  he  had  not  risked  the  direct 
route  from  Hostilia  to  Cremona. 

Next  day,  October  27,  a.d.  69,  Antonius  left 
his  legions  at  Bedriacum  with  orders  to  com- 
plete the  defences  of  that  camp.  He  himself 
rode    out    at    the    head    of    four    thousand    cavalry 

thirty  miles  on  the  day  before  the  night  of  the  final  battle.  Verona  to 
Cremona  is  fifty-five  miles.  If  the  Vitellians  found  a  road  striking  off  straight 
to  Cremona  at  some  point  on  the  main  road  short  of  Placentia,  the  distance 
from  Hostilia  to  Cremona,  vid  Parma,  may  be  reckoned  at  about  ninety  to 
ninety-five  miles. 

1  For  the  site  of  Bedriacum-Calvatone  see  Note  C. 


sec.  vi  FLAVIAN  INVASION  OF  ITALY    193 

eight  miles  along  the  road  in  the  direction  of 
Cremona.  His  auxiliary  infantry  were  sent  out 
on  both  sides  of  the  road  to  plunder  and  taste 
beforehand  the  sweets  of  victory,  which  could 
only  be  fully  theirs  when  they  had  won  the  battle. 
This  strange  order  could  not  be  justified  save  by 
the  character  which  civil  war  breeds  in  the  troops  ; 
and,  in  fact,  it  was  quickly  punished.  At  eleven 
o'clock  a  mounted  scout,  who  had  been  sent  forward 
by  Antonius,  came  riding  back  in  hot  haste  with 
the  news  that  the  foe  were  advancing  along  the 
road.  He  had  himself  seen  but  a  small  body  of 
their  horse,  but  "  the  noise  and  the  movement 
of  the  enemy  could  be  heard  far  and  wide."  This 
somewhat  vague  report  *  caused  Arrius  Varus,  with 
part  of  the  cavalry,  to  ride  on  to  investigate  its  truth. 
Antonius,  with  greater  foresight,  halted  and  drew  up 
the  remainder,  leaving  room  on  the  road  to  receive 
Varus  and  his  troopers  into  the  shelter  of  the  centre 
should  they  be  driven  back  by  weight  of  numbers. 
Messengers  were  at  once  sent  off  to  bid  the  legions 
march  forward  from  Bedriacum,  and  signals  were 
displayed  to  summon  the  plundering  auxiliaries  to 
the  spot.  Very  soon  Varus  and  his  company  were 
seen  galloping  back  at  full  speed  along  the  road, 
hotly  pursued  by  superior  numbers  of  the  enemy's 
horse.  The  garrison  of  Cremona  had  boldly  sallied 
out  to  the  attack.  The  runaways  plunged  madly 
into  the  midst  of  the  'Flavian  battle  -  line.  All 
Antonius'  precautions  were  vain,  and  the  whole 
mass  of  his  four  thousand  cavalry  fled  back  along 

1  It  is  a  report  more  worthy  of  the  historian  than  of  the  scout. 

O 


194    FLAVIAN   INVASION  OF  ITALY   ch.ii 

the  road  in  hopeless  panic  and  confusion,  their 
general  swept  along  in  the  rout,  vainly  protesting, 
imploring,  reproaching.  Their  flight  was  checked 
only  by  a  small  stream  with  steep  banks  and  danger- 
ous to  ford,  which  crossed  the  line  of  the  road.1  The 
bridge  over  it  had  been  broken,  and  Antonius  seized 
the  last  chance  of  rallying  the  fugitives  here.  A 
standard-bearer  came  wildly  riding  in  flight  down 
upon  him.  He  thrust  the  coward  through  with 
a  lance,  grasped  the  standard,  and  himself  turned 
it  to  face  the  pursuing  foe.  A  handful  of  his  men, 
not  more  than  a  hundred,  saw  the  sight  and  made 
a  stand  round  their  general  upon  the  streamlet's 
bank.  The  mass  of  fugitives,  their  flight  checked, 
rallied  round  them.  Discipline  had  already  been 
restored  and  some  kind  of  order  formed  when 
the  pursuers  came  in  sight,  following  the  rout 
recklessly  in  a  long  scattered  line.  They  were 
roughly  handled,  and  the  tide  of  pursuit  turned. 
Back  towards  Cremona  hurried  the  Vitellian  horse, 
and  in  their  wake  followed  the  Flavian  army,  now 
continually  reinforced  by  legionaries  from  the  camp 
and  auxiliaries  from  the  fields.  A  confused  mass 
of  horse  and  foot,  like  a  muddy  tide,  rolled  along 
the  Postumian  way  for  at  least  ten  miles,  until  it 

1  Heraeus,  quoting  one  Rycke,  calls  this  stream  "  the  Delmona,  a  tributary 
of  the  Oglio."  There  is  to-day  a  "  Dugale  Delmona  "  south  and  east  of 
Piadena.  But  the  Tacitean  stream  may  be  a  brook  more  to  the  north,  running 
(sic)  from  near  Drizzona  to  the  Oglio  near  Isola  Dovarese,  which  probably 
crossed  the  line  of  the  ancient  road  from  Calvatone  westwards.  Precise 
identification  is  impossible,  owing  to  the  modern  drainage  channels.  There 
is  a  marsh  north  of  San  Lorenzo  dei  Picenardi  and,  a  mile  to  the  west,  a 
moated  castello,  and  lakelet  at  Torre  dei  Picenardi,  and  there  may  well  have 
been  a  stream  here  in  old  days.  The  "  Delmona  "  (sic)  is  too  far  to  the  east 
and  much  too  near  to  Calvatone. 


sec.v,  FLAVIAN   INVASION  OF  ITALY    195 

presently  dashed  upon  the  enemy's  infantry.  The 
two  legions  had  marched  out  under  cover  of  their 
cavalry's  advance,  and  halted  four  miles  from 
Cremona.  Their  steadfast  array  and  standards 
glittering  in  the  sun  gave  promise  of  a  firm  resist- 
ance as  of  a  line  of  stubborn  cliffs  against  the  flood. 
But  the  Flavians  were  not  to  be  denied.  Their 
camp  lay  eighteen  miles  behind  them  ;  they  had 
fought  and  pursued  breathlessly  the  last  ten  miles. 
But  the  Vitellians  had  no  general  to  marshal  them 
to  admit  their  fleeing  horse  within  their  ranks,  or  to 
take  advantage  of  the  enemy's  weariness.  They 
stood  stolid  and  perplexed,  and  when  Vipstanus 
Messalla  hurled  himself  upon  them,  they  broke 
and  fled.  Shelter  was  near.  Outside  Cremona's 
city  walls  lay  their  own  fortified  camp.  They 
quickly  found  refuge  in  it,  and  the  road  up  to  its 
very  gates  remained  in  possession  of  the  Flavian 
troops. 

The  evening  shadows  were  falling  when  the 
mass  of  Antonius'  army  was  gathered  on  the  scene 
of  the  victory.  His  troops  clamoured  to  be  led 
to  the  final  assault.  Thoughts  of  the  sack  and 
gluttonous  rapine  of  a  helpless  city  when  night 
covered  every  deed  of  darkness  spurred  on  the 
infuriated  soldiery.  Their  general  knew  well  that 
to  assault  the  enemy's  position  that  night  was  a 
task  fraught  with  the  greatest  peril.  His  men  were 
tired  with  the  busy  day's  fighting  and  pursuit;  they 
had  no  siege  implements  ;  they  knew  nothing  of  the 
nature  of  the  fortifications  which  confronted  them. 
While  he  opposed  their  demands,  he  sent  the  least 


196    FLAVIAN   INVASION  OF  ITALY    ch.h 

wearied  of  his  horsemen  back  to  the  camp  at  Bedri- 
acum  to  bring  supplies  and  a  siege  train  up  to  the 
front  as  speedily  as  possible.  But  that  night  he 
would  advance  no  further.  All  Antonius'  entreaties 
and  arguments,  however,  could  not  prevail  over  the 
impatient  ardour  of  his  troops.  They  were  on  the 
point  of  advancing  madly  forward  when  scouts 
were  seen  speeding  back  from  the  direction  of  the 
city.  The  tidings  which  these  brought  hushed 
the  clamour  in  a  moment  and  gave  the  soldiers 
food  indeed  for  saner  thought.  The  whole  Hostilia 
army,  the  scouts  reported,  was  even  now  within  the 
city  walls  and  preparing  to  march  out  at  once  to 
the  attack. 

The  veterans  of  Caecina's  army  had  indeed 
accomplished  a  feat  well-nigh  beyond  the  possible. 
In  four  days1  they  had  covered  a  hundred  miles. 
That  morning  they  had  quitted  camp  thirty  miles 
away  from  Cremona,  and  now,  as  evening  fell,  they 
marched  into  the  city,  as  their  comrades  came 
fleeing  back  under  its  shelter.  The  whole  army 
was  at  last  united.  Once  more  the  scale  of  victory 
seemed  to  incline  towards  the  Vitellians.  They 
needed  but  a  general  of  their  own  to  throw  his 
sword  into  the  scale  and  the  day  was  theirs. 

Had  that  night  been  allowed  to  pass  without 
fighting,  the  Flavian  army  must  on  the  morrow  have 
been  in  sorry  plight.  Hungry  and  stiff  and  anxious, 
encamped  all  night  long  upon  the  open  road  within 

1  I  allow  a  day  for  the  news  of  the  evacuation  of  the  camp  to  be  brought 
to  Verona  and  for  Antonius  to  make  his  dispositions  for  the  march  to  Cremona. 
Not  even  Vitellians  could  have  marched  a  hundred  miles  in  three  days  and 
fought  on  the  evening  of  the  third. 


sec.  vi  FLAVIAN   INVASION  OF   ITALY    197 

striking  distance  of  the  enemy,  constantly  alarmed 
and  ever  under  arms,  without  food  or  fires  or 
entrenchments,  they  would  have  passed  the  night 
in  as  miserable  a  state  as  that  of  an  army  expecting 
attack  could  well  be.  Meanwhile  the  Vitellians, 
safe  within  their  lines,  warm  and  with  abundant 
supplies,  would  have  rested  from  their  long  day's 
march,  and  risen  for  the  battle  on  the  coming  day 
with  a  fresh  energy  and  confidence  equal  to  their 
courage  and  determination  to  conquer.  One  charge 
of  theirs,  it  might  be  thought,  would  have  broken 
the  cold  and  wavering  Flavian  line,  and  then  there 
remained  for  them  but  the  grimmest  and  the  most 
savage  of  pursuits. 

What  spell  of  Fortune's  weaving  was  it  that 
bewitched  the  men  of  the  German  army  ?  Surely 
Heaven  was  resolute  that  Vitellius'  rule  must  end, 
though  his  foes  in  the  field  sought  by  their  mistakes 
to  maintain  it.  Once  Antonius  had  offered  the 
enemy  victory  on  the  plains  by  Padua  and  Verona, 
and  had  been  saved  by  treachery.  Now  on  the 
morrow  he  promised  them  easy  victory  again. 
They  had  only  to  wait  for  it.  But  now  Folly  came 
to  Treachery's  aid  and  finished  the  work.  The 
Hostilia  army,  weary  and  footsore,  impatiently 
brooked  no  single  night's  delay.  They  had  no 
general  to  compel  them  to  be  wise.  At  nine  o'clock 
that  night  the  whole  Vitellian  force  marched  de- 
fiantly out  from  their  camp  under  the  walls  of 
Cremona  and  challenged  their  hated  foe  to  the  final 
struggle.  The  hour  for  which  they  had  been 
waiting  so  long  had  come  at  last  ;   the  prize  they 


198    FLAVIAN   INVASION  OF  ITALY   ch.h 

had  toiled  so  heavily  to  gain  seemed  at  last  within 
their  grasp.  A  night's  delay  might  let  it  slip.  But 
now  there  should  be  no  escape  for  the  enemy.1 

C.  The  Battle  of  Cremona. — Thus  "  indigus  rec- 
toris,  inops  consilii,"  the  Vitellian  army  marched 
out  and  drew  up  in  order  of  battle.  On  the  left  of 
the  raised  Postumian  way  were  stationed  the  men 
of  the  Twenty-second,  Sixteenth,  and  First  legions  ; 
on  the  road  itself  the  Fifth  and  Fifteenth  legions, 
with  the  detachments  of  the  Second,  Ninth,  and 
Twentieth  behind  them  ;  on  their  right  flank  stood 
the  Fourth  legion.  Men  of  the  First  legion  Italica 
and  the  Twenty-first  legion  were  scattered  along 
the  entire  line.  No  precise  position  is  given  for 
the  cavalry  and  auxiliaries,  but  the  former  at  least 
were  doubtless  posted  on  the  extreme  wings.  The 
Flavians  were  already  drawn  up  in  line  of  battle  to 
meet  them.  On  the  road  in  the  centre  was  the 
Thirteenth  legion.  North  of  the  road,  forming  the 
Flavian  right  wing,  and  drawn  up  along  a  cross 
road,  were  the  Eighth  legion  next  the  main  road  ; 
then  the  Third  legion,  distributed  in  the  intervals  of 
thick  brushwood ;  and  next  to  them  the  band  of 
Otho's  old  Praetorians  who  had  joined  Antonius' 
standard.  On  the  left  wing  were  the  men  of  the 
Seventh  Galbiana  legion  next  the  road,  beside  whom 
stood  those  of  the  Seventh  Claudia  legion,  whose 
front  was  protected  by  a  ditch.  The  auxiliaries 
were  placed  beyond  the  legionaries  on  both  wings, 
beyond  whom  again  were  some  of  the  cavalry,  while 
the   rest   were   kept   in   reserve  in  the   rear.     The 

1   Tac.  iii.   12-21. 


FLAVIAN   INVASION  OF  ITALY    199 


battle   plan    can   therefore    be   represented   by  the 
following  diagram  : — 


M 

—    <n 

bi  -o 

•3-5 


PVM 


>>   O    d     O  :=     O    >     O    •- 


rL  r- 


S      ■?     °  » 


iW 


41  — 


E±3.E±] 

>    o  > 


cb  cbcb 

.*     :=      x 
bi      M     S 

J     J     1 


£        3 


>        ;=        - 


.>   6  rt    6  >. 


200    FLAVIAN   INVASION  OF  ITALY   ch.h 

Fighting  began  about  nine  o'clock  at  night,  and, 
as  always  happens  in  battles  by  night,  was  confused. 
Order  was  quickly  lost,  and  hand-to-hand  conflicts 
were  waged  all   over  the   field.     The  two  armies 
were  armed  alike  ;  the  watchwords  quickly  became 
known   to   the   men   on   both   sides ;    and  captured 
standards  displayed  here  and  there   by  both   com- 
batants increased  the  perplexity  and  disarray.     The 
Flavian   left   was    hard  pressed,   and   the    Seventh 
Galbiana  legion  lost  men  quickly.      Its  very  eagle 
was  all  but  taken,  and  rescued  only  by  the  desperate 
valour  of  a  centurion,  who  died  to  save  it  from  the 
enemy.     Antonius  summoned  the  Praetorians  from 
the  right  wing  to  strengthen  the  wavering  line,  and 
the  battle,  now  restored,  swayed  to  and  fro  in  alter- 
nate advance  and  retreat.     The   Vitellian  artillery 
had  at  the  beginning  of  the  fight  been  scattered  up 
and  down   the  line  of  battle,  and  its  missiles  had 
gone  hurtling  among  the  bushes  opposite  without 
doing  great  hurt  to  the  enemy.      But  later  all  the 
engines   were   massed    together  on   the   high-road, 
and  their  fire,  concentrated  on  the  clear  space  in 
front    of    them,    made    the     Flavian    centre    suffer 
heavily.       Here    again    the    tide    of    war    seemed 
setting  against  Antonius,  when  two  of  his  soldiers 
found  a  remedy.     Their  names  are  not  known,  but 
their  deed  is  not  forgotten.     Snatching  up  shields 
from  two  of  the  enemy's  dead,  they  made  their  way 
undetected   over   to  the    hostile   line,   and   cut   the 
ropes   of  the   engines.     At  once  they  fell,  pierced 
with  wounds  ;   but  they  had  saved  their  comrades 
and  their  general,  for  now  the  enemy's  artillery  was 


sEcv.  FLAVIAN   INVASION  OF  ITALY    201 

useless.  Presently,  late  at  night,  the  moon  rose 
in  the  east,  and  shone  full  upon  the  faces  of  the 
Vitellians.  The  moonlight,  disabling  their  own 
sight,  exposed  them  to  the  sure  aim  of  the  foe, 
while  they  themselves  smote  vainly  at  the  shadows 
which  the  dark  figures  of  the  soldiers  opposite  cast 
far  on  the  ground  before  them.  Ever  and  again 
clouds  drifted  over  the  face  of  the  moon,  and  then, 
as  by  common  consent,  the  fighters  drew  apart  and 
rested,  leaning  on  their  weapons,  until  the  moon 
shone  out  full  again.  Women  came  out  from 
Cremona,  some  themselves  to  plunge  into  the  battle 
and  be  slain,  fighting  fiercely  for  the  cause ;  some 
to  carry  food  and  drink  to  the  soldiers  of  their 
army.  The  Vitellians  ate  and  drank,  and  offered 
of  the  fare  also  to  the  enemy.  "  Come,  comrades," 
they  cried.  "  Here  is  meat  and  drink  :  take 
and  eat ;  take  and  drink ;  that  we  may  slay  and 
be  slain,  but  strong  and  not  fainting."  Then 
arms  were  grounded,  and  the  men  ate  and  drank 
together.  But,  the  short  rest  over,  they  fell 
again  to  fighting  with  bitterness  and  anger  all  the 
greater. 

All  through  the  long  autumn  night  the  battle 
raged  with  unabated  fury.  Here  son  slew  father 
unawares ;  here  brother  cut  down  brother.  Men 
shuddered  at  such  sights,  and,  hastening,  did  the 
like.  The  Flavian  general  was  to  be  seen  every- 
where in  his  battle  -  line,  encouraging,  taunting, 
rebuking,  cheering  his  soldiers  on  to  yet  stronger 
blows  and  a  more  stubborn  stand.  "On  that  same 
battle-field,  yet  cumbered  with  the  relics  of  their 


202    FLAVIAN   INVASION  OF  ITALY   ch.ii 

dead,  the  Pannonian  legions  must  redeem  their 
honour  from  the  stain  of  the  defeat  which  they  had 
once  suffered  there.  The  men  of  Moesia  had  been 
bold  enough  of  speech  against  the  foe :  could  they 
not  show  the  deeds  to  match  ?  Dared  the  men  of 
the  Third  to  shame  the  records  of  the  regiment  ? 
Had  it  not  fought  under  Mark  Antony  in  Parthia, 
under  Corbulo  in  Armenia?  Had  it  not  but  newly 
crushed  the  wild  Sarmatian  invader  and  saved  their 
province  ?  Why  above  all,  he  fiercely  demanded, 
were  the  Guards  hanging  back  in  the  final  hour  of 
trial  ?  Had  they  not  even  yet  drained  ignominy 
to  the  full  ?  Boors  and  peasants  that  they  were, 
soldiers  no  more,  did  there  remain  for  them  yet 
another  Emperor,  another  camp,  to  shelter  them  ? 
Their  standards,  their  arms,  were  with  the  enemy  ; 
for  them  death  alone  was  guerdon  of  defeat." 
Everywhere  the  men  wildly  cheered  their  fiery 
leader  as  he  rode  up  and  down  the  line,  and  grimly 
they  held  their  ground,  until  at  last  the  sun  rose 
upon  the  scene. 

Then  the  Third  legion,  lately  come  from  Syria, 
saluted  it,  as  was  their  wont,  and  the  chance  salute 
decided  at  last  the  day.  The  word  ran  fast  down 
the  Flavian  line,  that  Mucianus  and  their  comrades 
of  the  Eastern  army  had  come  at  last.  Their 
hopes  rose  high.  The  enemy  caught  the  rumour 
and  wavered.  In  one  final  heave  of  massed 
column  the  Flavians  thrust  desperately  at  the 
Vitellian  line,  now  ragged,  thin,  despairing.  The 
line  bent  and  gave.  There  was  no  rally.  Ensnared, 
inextricably  involved,  among   the  broken  engines, 


sec.v.  FLAVIAN   INVASION  OF  ITALY    203 

the  waggons,  the  heaps  of  slain,  the  Army  of  Ger- 
many broke  up  into  a  rout  of  fugitives,  and  the 
enemy's  horse,  cutting,  hewing,  butchering,  drove 
them  to  their  camp.  The  battle  on  the  open  field 
was  ended. 

The  tide  of  victory  surged  up  against  the  gates 
and    ramparts    of    the    camp.      The    troops    had 
marched  and  fought  for  twenty  miles   and  twenty 
hours.       Still    Antonius    gave    them    no    rest,    but 
called  on  them  for  the  last  great  effort,  and,  as  one 
man,  they  answered  to  the  call.     A  very  storm  of 
missiles  raged  for  some  time  on  either  side.     Then 
two  columns  of  assault  rushed  at  the  ramparts  and 
the  gates  on  the   eastern  and  the  northern  roads, 
towards    Bedriacum    and    Brixia.     The  men   were 
hurled  back.     Antonius  flung  himself  among  them. 
With   significant  gesture  he  pointed  to  the   city  : 
Cremona  was  theirs  to  sack,   if  they  would   rally. 
Himself  at  the  head  of  the  storming  column,  he  led 
the   Third  and   Seventh   legions  again   up   to   the 
Bedriacum   gate.      Down   crashing   on   their  heads 
came  the  great  engine  of  war  itself,  hurled  by  the 
desperate  defenders,  and  they  recoiled  once  more. 
It  was  but  for  a  moment.     The  engine's   fall  had 
torn    away    with    it    part    of  the    rampart.      Fresh 
assailants  swarmed  to  the  breach,  the  men  of  the 
two  legions  vying  with  one  another  in  eager  regi- 
mental rivalry.      The  gate  was    hewn    down   with 
axes  and  with  swords.     Volusius  of  the  Third  was 
the  first   man  in.      The    others    poured    over   and 
through   the    defences.      The   Vitellians    leapt  de- 
spairing  from  the  ramparts  as  the   foe   rushed   in. 


204    FLAVIAN   INVASION  OF  ITALY   ch.  » 

The   camp   was   cleared  of  the   living   among   the 
enemy  up  to  the  city's  walls.1 

D.  The  Sack  of  Cremona.  —  The  actual  town 
itself  seemed  still  to  defy  assault.  It  was  crowded 
with  citizens  and  many  strangers  who  had  gathered 
there  for  the  fair,  which  had  chanced  upon  those 
very  days ;  and  many  of  the  defeated  troops  had 
escaped  within  the  city  and  thronged  its  lofty  walls 
and  towers,  menacing  the  foe.  But  Antonius  never 
hesitated.  Soon  the  inhabitants  saw  the  fairest  of 
their  buildings  outside  the  walls  in  a  blaze  of  fire, 
and  others,  which  overtopped  the  ramparts,  crowded 
with  soldiery.  A  rain  of  missiles  and  flaming 
brands  again  began  to  descend  upon  the  walls,  and 
under  its  cover  the  legionaries  were  seen  moving 
to  the  assault.  For  nearly  three  centuries  the  proud 
and  stately  city  had  been  queen  of  the  valley 
of  the  Po.  In  her  earliest  years  the  fierce  Gallic 
tribes  had  raged  round  her  walls  in  vain.2  Temples 
and  palaces  gave  her  beauty ;  walls  and  iron-clamped 
gates  glorified  her  strength  ;  the  river  bestowed 
riches  on  her  fields  and  prosperity  on  her  citizens. 
Now  at  last  an  enemy  sterner  than  the  Gaul,  fiercer 
than  the  barbarian,  was  at  her  gates.  And  the 
garrison  played  her  false.  The  officers,  hopeful  of 
mercy  for  themselves,  surrendered  the  city.  The 
soldiers  sullenly  allowed  it,  or,  careless  of  the  end, 
roamed  through  the  streets,  plundering  or  fighting. 
Caecina,  who  had  been  hurried  there  by  the  army 
which  once  owed  him  obedience  and  cast  into  the 

1  Tac.  iii.  21-29;  Dio  lxv.  12-14. 
-  200  B.C.     The  city  was  founded  as  a  Latin  colony  in  218. 


skc.v,  FLAVIAN   INVASION  OF   ITALY    205 

city  prison,  was  released  from  his  dungeon,  clad  in 
the  robes  and  decorated  with  the  insignia  of  his  con- 
sular office,  and  the  men  begged  him  humbly  that 
he  should  plead  for  them  with  the  victors.  "  It  was 
the  last  of  evils  that  so  many  valiant  men  should 
implore  the  traitor's  aid."  The  gates  were  thrown 
open  ;  the  garrison  laid  down  their  arms,  and 
marched  out  between  lines  of  jeering  troops.  But 
soon  mockery  was  changed  to  pity.  The  victors 
had  been  vanquished  by  the  vanquished  of  to-day, 
and  as  they  had  received  mercy  at  the  others' 
hands,  so  now  it  should  be  shown  these  in  return. 
Only  when  Caecina  came  proudly  out  from  the 
gate,  glittering  with  his  train  of  lictors,  did  a  fierce 
cry  of  scorn  and  hate  arise.  Antonius  checked  it, 
and  sent  the  traitor  under  guard  to  Vespasian. 

For  the  moment  the  city  itself  was  spared,  but 
only  for  the  moment.  The  soldiers  had  not  for- 
gotten the  insults  which  the  citizens  a  few  months 
earlier  had  heaped  upon  them.1  And  never  did  city 
promise  richer  plunder.  At  the  crisis  of  the  struggle 
the  general  had  spurred  his  troops  on  by  the  thought 
of  spoiling  it.  He  should  not  restrain  them  now. 
Already  the  flames  were  spreading,  and  one  chance 
word  did  the  rest.  Antonius  hastened  to  the  baths 
to  wash  off  the  blood  and  grime  which  covered 
him.  The  water  was  cold.  "  Were  not  the  fires 
lit  ?  "  he  impatiently  demanded.  An  anxious  slave 
hastened  to  him  with  the  assurance  :  "  It  will  soon 
be  warm."  Question  and  answer  ran  from  mouth 
to  mouth.     The  time  had  come  to  light   the  fires 

1  See  above,  p.  136. 


206    FLAVIAN   INVASION  OF  ITALY   ch.h 

of  rapine  :  this  was  their  general's  meaning.  Forty- 
thousand  armed  men,  and  a  yet  larger  and  more 
horrible  army  of  sutlers,  camp  followers,  the  refuse 
and  sweepings  of  the  vilest,  broke  into  the  city. 
For  four  days  it  was  given  up  to  their  maddened 
lust  and  rage.  The  chapter  in  which  the  Roman 
historian  tells  the  story  of  the  sack  equals  in  ghast- 
liness  the  tale  of  the  sack  of  Rome  by  the  forces  of 
the  Constable  de  Bourbon,  of  heretic  Antwerp  by 
the  fanatic  Spaniards.  When  four  days  had  passed, 
fifty  thousand  had  perished  by  the  sword  and 
torture,  by  fire  and  by  lust.  Flames  consumed  the 
city.  Only  a  solitary  temple,  that  of  Mefitis  out- 
side the  city  walls,  remained  untouched  by  them. 
The  very  spoilers  were  driven  to  encamp  three 
miles  away  by  the  reek  of  the  blood  which  rose 
from  the  poisoned  soil.  In  this  way  Cremona  came 
to  its  end.1 

There  are  wars,  even  civil  wars,  which  inspire 
devotion  and  self-sacrifice  ;  this  struggle  of  a.d.  69 
displayed  the  horrors  of  war  in  all  their  nakedness. 

§  7.    The  Advance  to  Rome 

A.  The  Halt  at  Fano. — The  sack  of  Cremona 
ended  on  October  31  ;  there  were  still  nearly  two 
months  to  pass  before  the  end  of  the  war  came. 

News  of  the  victory  was  sent  at  once  by 
Antonius  to  the  western  provinces,  Spain,  Gaul, 
and     Britain.       All    three    presently    declared    for 

1  Tac.  iii.  30-35  ;  Dio  lxv.  15.      The  city  was  rebuilt  under  Vespasian,  but 
the  disaster  remained  proverbial. 


sec. vii  FLAVIAN  INVASION  OF  ITALY  207 

Vespasian,  the  First  Adjutrix  legion  in  Spain 
setting  the  example.  This  legion  had  never  for- 
given Vitellius'  rise  to  power  and  Otho's  fall.1 
Next,  the  defeated  army  had  to  be  sent  away  to  a 
safe  distance,  lest  it  should  still  take  a  part  in  the 
resistance  which  the  Emperor  would  offer.  The 
men  were  sent  to  the  Danube  provinces,  save  for 
a  few  cavalry  who  took  service  with  the  Flavians. 
Distributed  skilfully  through  Dalmatia,  Illyricum, 
and  Moesia,  they  gave  after  this  no  cause  for 
anxiety,  and  in  Moesia  were  of  excellent  service 
against  the  marauding  Dacians.2  There  then 
remained  only  the  fear  lest  Germany  should  still 
send  men  to  Vitellius'  aid.  Antonius  therefore 
at  once  sent  troops  to  occupy  the  Alpine  passes. 
The  mutiny  against  Rome  which  shortly  afterwards 
broke  out  on  the  Rhine  was  already  so  far  afoot 
(under  the  guise  of  a  war  in  Vespasian's  interests 
against  the  Vitellian  troops  on  the  river)  that  the 
army  in  Germany  was  but  too  busily  occupied,  and 
not  a  man  was  sent  to  cross  the  Alps.  It  was 
Antonius  who,  by  letters,  had  provoked  this  rising 
on  Vespasian's  behalf,  and  his  scheme  was  so  far 
magnificently  successful.3  But,  in  very  truth,  to 
encourage  those  who  were  little  better  than  bar- 
barians to  rise  against  the  Romans,  even  though 
these  last  might  be  of  the  opposite  faction,  was 
nothing  else  than  to  play  with  fire,  and  brought 
quickly  a  terrible  retribution  in  its  train.  It  was 
as  if  the  English  had  let  loose  the  Basutos  upon 

1  Tac.  iii.  35,  44.  2  Tac.  iii.  35,  46. 

3  Tac.  iv.  13  ;  see  below,  Chap.  III.,  for  the  history  of  this  rising. 


208    FLAVIAN   INVASION  OF  ITALY 

their  enemy  in  the  recent  Transvaal  war.  To  this 
extent  Antonius  lacked  the  true  Imperial  feeling 
which,  if  it  delayed  sorely  the  coming  of  peace,  yet 
gave  us  the  chance  of  goodwill  when  peace  at  last 
did  come. 

No  such  thoughts  troubled  Antonius  the  Roman. 
The  immediate  military  need  was  his  only  care, 
and  for  this  at  least  he  had  made  most  wise  pro- 
vision. That  he  had  opened  the  floodgates  to 
rebellion  and  savagery  in  the  far  North,  and  that 
the  flood  would  not  hereafter  be  arrested  at  his 
word,  he  refused  or  was  unable  to  perceive.  At 
least  he  had  stayed  the  coming  of  all  reinforcements 
from  any  quarter  of  the  Empire  to  Vitellius.  And 
with  this  he  was  well  content. 

Having  taken  these  precautions,  the  general 
turned  his  thoughts  to  the  enemy  in  the  South. 
The  Emperor  at  Rome  might  be  inert  and  torpid, 
but  at  least  he  would  not,  like  Otho,  save  his  foes 
the  need  of  further  fighting  by  slaying  himself, 
because  his  army  had  been  vanquished  on  the 
banks  of  the  Po.  Moreover,  he  still  commanded 
troops  of  excellent  quality  and  by  no  means 
contemptible  in  numbers.  Twenty-five  thousand 
infantry,  and  most  of  these  the  veterans  whom  he 
had  made  his  Guards,  could  not  be  played  with  as 
if  they  had  been  a  toy  army  or  a  rabble  of  recruits. 
The  campaign  had  opened  for  the  Flavians  well 
indeed,  but  much  work  yet  remained  to  be  done. 
The  invading  army  must  of  necessity  advance 
towards  the  capital. 

Yet  such  an  advance  promised  many  difficulties. 


sec. vii  FLAVIAN   INVASION  OF  ITALY  209 

From  Cremona  to  Ariminum  the  road  was  easy  all 
through  its  length  of  one  hundred  and  fifty  miles.  It 
ran  over  a  perfectly  flat  plain,  skirting  the  mountain 
chain  upon  the  right  hand.  But  after  Ariminum  the 
troubles  began.  The  great  highway  to  Rome,  the 
Via  Flaminia,  crossed  the  ridge  of  the  Apennines 
at  its  lowest  point.  From  Ariminum  it  ran  along 
the  sea-shore  to  Fanum  Fortunae  (Fano),  and  there 
struck  inland  up  the  stony  winding  channel  of  the 
Metaurus,  entering  the  hills  at  Forum  Sempronii, 
a  name  corrupted  to-day  into  that  of  Fossombrone, 
the  last  comfortable  village  of  the  lowlands.  From 
that  point  the  climb  began,  by  Cales  (Cagli)  and 
the  wild  ravine  of  Cantiano  to  the  top  of  the  pass, 
which  lies  at  a  height  of  2400  feet  above  the  sea. 
The  rise  to  this  was  very  steep.  Thence  it 
dropped  to  Nuceria  (Nocera  Umbra)  and  Fulginium 
(Foligno).  From  Fulginium  it  crossed  the  five- 
mile  expanse  of  level  plain  to  Mevania  (Bevagna) 
opposite,  and  the  chief  natural  difficulties  of  the 
road  were  ended.  But  if  at  places  this  road  runs 
through  a  fair  and  smiling  land,  adorned  with  fields 
and  lanes,  flowers  and  fruit-trees,  worthy  of  the 
county  of  Devon,  at  others  it  pierces  rocky  ravines, 
crawls  up  through  gorges  and  rocky  mountain  sides 
black  with  oakwoods  or  bare  to  all  the  blasts  of 
heaven.  The  mountains  shoot  steeply  up  first  on 
the  right  hand,  then  on  the  left,  rugged,  inhospit- 
able, cleft  by  great  red  ravines  and  strewn  with 
broken  rocks  and  screes.  The  hamlets  are  squalid 
and  miserable,  the  mountaineers  in  appearance  a 
rude   and  lowering   race.      Other   tracks   cross  the 

p 


210    FLAVIAN   INVASION  OF  ITALY   a.... 

central  ridge,  but  there  is  none  which  can  so  easily 
be  traversed  as  this  of  the  main  highway.  Many 
miles  to  the  south  the  Via  Salaria  crosses  the 
central  heights  of  the  Abruzzi  from  Ascoli  to  Rieti, 
but  the  difficulties  of  the  Flaminian  way  are  as 
nothing  to  the  toils  which  await  the  traveller  who 
plunges  by  this  route  into  the  heart  of  Italy's  wildest 
mountains.  More  than  a  hundred  and  eighty  miles 
separate  Ariminum  and  Rome  ;  for  the  first  hundred 
miles  the  road  is  mountainous.  And  to  Antonius 
and  his  army  the  late  season  of  the  year  increased 
the  difficulties.  Already  in  November  snow  had 
fallen  on  the  mountains,  and  bad  weather  had  set 
in.  There  was  little  food  to  be  obtained  along  the 
road  until  he  reached  the  great  central  Umbrian 
valley  at  Foligno,  and  December  would  be  upon  him 
first,  even  if  the  enemy  made  no  effort  to  block  the 
way.  Even  Vitellius  could  hardly  fail  to  seize  this 
great  chance  which  the  winter  offered  him.  And  if 
the  pass  were  blocked  by  the  Emperor's  army,  with 
a  strong  force  at  its  summit,  and  fifteen  thousand 
men  in  camp  at  Foligno,  Antonius  could  never  force 
the  passage  of  the  mountains  in  that  year. 

Other  causes  also  made  the  general  hesitate. 
There  was  dissension  in  his  staff,  some  urging  the 
advance,  some  bidding  him  delay  till  Mucianus 
came.  Mucianus  himself  wrote  to  him  in  ambigu- 
ous terms.  He  doubtless  desired  himself  to  lead 
the  victorious  army  into  Rome.  Antonius,  some- 
what of  a  braggart,  a  veritable  soldier  of  fortune, 
it  might  seem  to  the  other,  had  gained  success 
enough.     The  two  men  hated  and  distrusted  each 


sec. vn  FLAVIAN   INVASION  OF  ITALY  211 

the  other,  and  on  Antonius'  staff  were  many  who 
looked  to  the  greater  man  for  their  promotion,  and 
sought  now  to  thwart  the  general  who  had  led 
them  to  victory.  If  Antonius  gave  the  order  to 
advance,  all  the  penalties  of  failure  would  be  visited 
on  his  head.  Mucianus  clearly  washed  his  hands 
of  all  responsibility.  The  very  troops,  knowing 
well  the  quarrels  in  the  staff,  were  turbulent  and 
clamorous  for  "  shoe-money."  Jealousy  and  insub- 
ordination, difficulties  of  supply,  perils  of  the  road, 
defiance  by  the  enemy, — all  these  troubles,  actual 
or  possible,  pressed  heavily  on  Antonius. 

He  moved  forward  to  Ariminum.  The  town 
was  still  in  the  enemy's  hands,  but  the  fleet  under 
command  of  Cornelius  Fuscus  had  by  this  time 
closely  invested  it  by  land  and  sea.  There  was 
therefore  no  army  upon  the  eastern  side  of  the 
mountains  and  down  the  Adriatic  coast  to  oppose  his 
advance.  And  the  timely  capture  of  Valens  on  the 
Riviera  was  in  every  way  most  fortunate.1  Yet  even 
the  march  to  Fanum  Fortunae,  twenty-eight  miles 
beyond  Ariminum,  was  not  without  its  difficulties. 
Heavy  autumn  rains  had  swollen  the  Po  and  its 
tributary  streams,  and  the  low  country  of  the  valley 
at  foot  of  the  mountains  was  flooded  to  such  an 
extent  that  the  heavy  baggage  of  the  soldiers  had 
to  be  left  behind.  Commissariat  troubles,  too,  began 
early.  Antonius  failed  to  keep  his  troops  in  hand, 
and  they  indulged  in  indiscriminate  plunder  on  the 
way.  This  of  course  at  once  doubled  the  difficulties 
of  supply.     The  force  which  arrived  at  last  at  Fano 

1  See  above,  p.  184;  cf.  Tac.  iii.  44  for  its  effect. 


212    FLAVIAN   INVASION  OF  ITALY   ch.h 

was  neither  a  strong  one  nor  in  good  temper.  All 
the  sick  and  wounded  had  been  left  behind  at  the 
base,  Verona,  but  also  the  greater  part  of  the  legion- 
aries remained  there,  and  Antonius  at  Fano  mustered 
only  picked  troops  from  these,  together  with  auxiliary- 
infantry  and  cavalry.  The  mountains  rose  before 
him,  and  he  halted.  Further  advance  was  not 
possible  until  his  force  was  strengthened,  supplies 
were  collected,  and  the  country  in  his  front  was 
explored.  Antonius  set  his  hand  resolutely  to  all 
three  tasks.  The  legions  were  summoned,  supplies 
hurried  up  by  sea,  and  cavalry  scouts  sent  forward 
to  discover  if  the  enemy  had  occupied  the  pass.1 
But  all  this  involved  delay,  and  winter  was  fast 
approaching.  Vitellius  had  his  opportunity,  if  he 
had  the  wit  to  use  it. 

B.  Movements  of  the  Vitellian  Forces. —  The 
Emperor  had  meanwhile  been  waiting  on  circum- 
stance, and  this  at  last  had  roused  him  to  action. 
At  first,  as  soon  as  his  army  had  marched  for  the 
north  under  Caecina,  and  when  Valens  had  pre- 
sently followed  after  it,  Vitellius  took  no  further 
interest  in  the  war.  In  the  pleasant  shade  of  his 
gardens,  or  under  the  trees  of  the  woods  which 
clothe  thickly  the  sides  of  the  Alban  hills  at 
Aricia,  a  few  miles  south  of  Rome,  the  ruler  of 
the  Empire  dozed  the  days  away,  heavy  with  food 
and  slumberous,  torpid  as  a  fat  and  well-fed  toad.2 

1  Tacitus  explains  that  he  sent  the  cavalry  on  to  explore  the  whole  of 
Umhria  :  "si  qua  Appennini  juga  clementius  adirentur,"  iii.  52.  That  the 
Via  Flaminia  was  the  easiest  route  could  hardly  be  in  question. 

2  "  Ut  ignava  animalia  quihus  si  cibum  suggeras  jacent  torpenlque,"  Tac. 
iii.  36. 


sec. vn  FLAVIAN  INVASION  OF  ITALY  213 

The  news  of  the  mutiny  of  the  Ravenna  fleet 
scarcely  moved  him.  Valens'  urgent  request  for 
reinforcements  was  answered  by  the  sending  of  a 
petty  force  which  was  far  too  weak  to  be  of  any 
use.1  Only  at  last  the  tidings  of  Caecina's  treachery 
and  the  troops'  loyal  requital  of  it  woke  him  from 
his  slumbers  in  the  greenwood.  "  With  that  dull 
soul  joy  had  a  greater  weight  than  trouble."2 
Vitellius  came  to  Rome  and  harangued  both  Senate 
and  people.  When  the  news  came  of  the  battle 
of  Cremona  the  orator's  powers  abruptly  failed  him. 
Every  one  at  Court  went  about  silently,  and  no  one 
made  any  allusion  to  so  unfortunate  an  incident. 
They  whispered  in  the  anterooms  and  streets,  but 
in  the  Emperor's  presence  no  one  had  heard  any- 
thing of  the  battle.  The  Government  ordered 
silence  on  the  topic,  and,  if  it  were  possible,  the 
disaster  was  magnified  in  consequence.3 

The  spies  sent  out  were  courteously  welcomed 
by  the  enemy  and  escorted  round  their  camp. 
Vitellius  blandly  shut  his  ears  to  their  reports.  At 
last  a  brave  centurion,  Julius  Agrestis,  convinced 
his  Emperor  that  it  was  time  to  be  up  and  doing. 
Allowed  at  his  own  request,  so  ran  the  tale,  to  go 
out  from  Rome  to  discover  the  actual  facts  of  the 
situation,  he  went  openly  to  Cremona  and  straight 
to  Antonius,  avowing  his  mission.  That  general 
readily  gave  him  guides  and  showed  him  every- 
thing— his  army,  the  battle-field,  the  prisoners  of 
war.       Back   came   Agrestis    to   Vitellius  and    told 

1  See  above,  p.  184. 

"  "Plus  apud  socordem  animum  laetitia  quam  cura  valuit,"  Tac.  iii.  36. 

3  Tac.  iii.  36,  37,  54. 


2i4    FLAVIAN   INVASION  OF  ITALY 


CH.   II' 


him  all  the  truth.  The  Emperor  refused,  as  usual, 
to  believe  a  word  of  it,  and  suggested  that  he  had 
been  bribed  to  bring  such  news.  The  centurion 
was  indignant.  "  Since,"  he  cried,  "  you  have  need 
of  a  sure  proof,  and  have  no  longer  any  other  use 
whether  of  my  life  or  death,  I  will  give  you  proof 
verily  to  believe."  With  these  words  he  hastened 
from  Vitellius'  presence  and  slew  himself.  Then 
at  last,  "  as  one  roused  from  sleep,"  the  Emperor 
took  measures  of  defence.1 

His  available  forces  consisted  of  sixteen  cohorts 
of  Praetorian,  and  four  of  Urban,  Guards — twenty 
thousand  men  in  all.  To  these  was  added  a  new 
legion,  hurriedly  levied  from  among  the  sailors  of 
the  fleet  at  Misenum,  which  may  have  numbered 
five  thousand  more.'2  Besides  these  he  had  a  small 
force  of  cavalry  at  his  disposal.  With  such  forces 
he  still  might  defend  Italy,  at  least  unless  Vespasian 
came  at  the  head  of  all  the  armies  of  the  East.  And 
of  him  there  was  no  sign. 

The  greater  part  of  this  force  was  sent  north- 
wards along  the  Flaminian  road  to  defend  the 
ridge  of  the  Apennines.  Fourteen  of  the  sixteen 
Praetorian  cohorts,  the  new  legion,  and  the  cavalry 
marched  out  promptly  as  far  as  Mevania  (Bevagna), 
eighty  miles  from  Rome,  near  the  issue  of  the  road 
from    the    northern    mountains.       The    other    two 

1   Tac.  iii.  54. 

-  This  '•  legio  e  classicis  "  is  of  course  not  Legio  I.  Adjutrix,  which  was  then 
in  Spain  (Tac.  ii.  67,  86  ;  iii.  44),  but  the  nucleus  of  the  legion  later  formally 
enrolled  by  Mucianus  in  the  name  of  Vespasian  under  the  title  of  Secunda 
Adjutrix.  There  is  a  military  diploma  of  March  7  a.d.  70,  applying  to 
some  who  have  seen  service  with  this  legion.  Cf.  Dio  lv.  24  ;  C.T.L.  iii. 
849,  907.      Hardy,  Studies  in  Romaji  History,  p.  207. 


s.x.vii  FLAVIAN  INVASION  OF  ITALY  215 

Praetorian  cohorts  with  the  four  cohorts  of  Urban 
Guards  were  left  in  Rome,  under  the  Emperor's 
brother  Lucius,  to  garrison  the  city.  Vitellius  him- 
self still  dallied  for  some  time  in  Rome,  but  at  last, 
at  his  army's  urgent  entreaty,  joined  the  camp  at 
Mevania.  No  enemy  as  yet  had  been  seen  upon 
the  pass.  He  had  even  men  and  time  enough  to 
cross  it  and  descend  upon  the  scattered  enemy  from 
the  hills,  hurling  them,  if  fortune  served,  back  in 
rout  from  Fano  towards  the  flooded  valley  of  the 
Po.  The  Roman  historian  himself  maintains  this 
strategy  to  have  been  the  right  one  for  him  : — 

It  was  open  to  Vitellius  to  cross  the  Apennines  with  the 
vigour  of  his  army  unimpaired,  and  to  fall  upon  the  foe  while  these 
were  weary  with  the  winter's  cold  and  hunger.  But  he  divided 
up  his  strength  and  scattered  it ;  he  gave  over  to  slaughter  and 
captivity  troops  of  the  keenest  courage  and  faithful  to  the  last. 
Though  the  most  skilful  of  his  centurions  opposed  his  plan  and 
would  have  told  him  the  truth  had  he  but  inquired  of  them,  his 
friends  held  them  back  from  coming  to  his  presence.1 

Not  only  did  the  Emperor  refuse  to  advance 
over  the  mountains,  but  very  soon  "he  divided  up 
his  strength  and  scattered  it."  Dire  omens,  indeed, 
were  seen  at  Mevania,  but,  as  the  historian  grimly 
says,  "  Vitellius  was  his  own  worst  portent.  .  .  . 
Ignorant  of  soldiering,  improvident  of  counsel,  here 
asking  one  concerning  the  drill  of  marching  order, 
there  another  concerning  a  scout's  duties ;  here 
questioning  whether  it  were  well  to  hasten  on  the 
final  issue,  there  whether  to  delay  it ;  in  his  face 
and  limbs  alike  making  manifest  his  fear  when  each 

1  Tac.  iii.  56. 


216    FLAVIAN  INVASION  OF  ITALY   ch.  n 

new  messenger  arrived ;  and  at  the  last  reeling 
drunken  round  the  camp  " — such  was  Vitellius  the 
Emperor  among  his  troops. 

He  alone 
Dealt  on  lieutenantry  and  no  practice  had 
In  the  brave  squares  of  war. 

The  very  camp  became  wearisome  to  him  ; 
doubtless  the  camp  kitchen  pleased  him  ill ;  and 
when  one  more  message  of  disaster  reached  him 
he  left  Mevania  and  returned  to  Rome.  The  fleet 
at  Misenum  had  mutinied  against  him.  The  rebels 
had  seized  the  city  of  Tarracina,  where  the  road 
creeps  round  between  the  sea  and  the  sheer  cliff 
which  towers  many  hundred  feet  above  it  and  all 
but  bars  its  passage.  The  city's  walls  and  strong 
position  made  it  a  fortress  all  too  hard  to  storm. 
And  now  it  too  was  in  an  enemy's  hands,  and 
Campania,  south  of  Rome,  in  a  ferment. 

The  Emperor  at  this  lost  the  last  portion  of 
military  wisdom  which  was  his.  In  that  desperate 
situation  one  with  cool  head  would  have  seen  that 
Campania  mattered  very  little.  The  enemy  there 
were  still  but  a  sorry  band,  and  the  fierce  local 
jealousies  between  its  cities  preserved  the  loyalty 
of  some  and  thereby  hampered  the  hostility  of 
others.  A  very  small  force  sent  from  Rome  would 
have  been  enough  to  keep  the  rebels  of  Campania 
in  check.  The  true  danger  lay,  as  always,  north  of 
the  Apennines. 

But  now  Vitellius  issued  his  last  and  most  fatal 
orders  for  the  redistribution  of  his  troops.  The 
Emperor  himself  took  seven  cohorts  and  part  of  the 


sec. vii  FLAVIAN   INVASION  OF  ITALY  217 

cavalry  with  him  to  Rome.  A  poor  seven  only, 
with  the  legion  and  part  of  the  cavalry  to  help 
them,  were  left  to  defend  Italy  against  the  attack 
from  the  north.  On  arriving  at  Rome,  Vitellius 
sent  his  brother  with  six  cohorts  and  five  hundred 
horse  to  Campania,  keeping  with  him  in  the  city 
three  Praetorian  cohorts,  and  probably  the  Urban 
cohorts  as  well.1 

Thus,  instead  of  concentrating  his  army  where 
it  was  above  all  needed,  the  Emperor  made  three 
divisions  of  it ;  instead  of  advancing,  as  a  bold 
general  might  perhaps  have  advanced,  over  the 
pass  to  search  for  the  enemy,  he  recalled  the 
division  now  left  at  Mevania  back  to  Narnia,  thirty 
miles  in  the  rear,  and  only  some  fifty  miles  from 
Rome  ;  instead  of  at  least  attempting  to  block  the 
pass  by  which  the  Flaminian  road  crossed  the 
mountains,  he  left  it  bare  of  all  defence,  opposing 
the  foe's  advance  over  it  by  nothing  save  by  the 
snow  which  had  fallen  upon  it.  "  Fortune,"  com- 
ments the  historian,  "  helped  the  Flavian  leaders 
not  less  often  than  did  their  own  counsels."2  The 
cup  of  Vitellius'  blunders  was  now  indeed  full  to 
the  brim. 

1  The  figures  work  out  as  follows:  there  were  16  cohorts  of  Praetorians 
(Tac.  ii.  93);  of  these  14  go  to  Mevania  (ibid.  iii.  55),  thus  2  remain  in 
Rome.  Later  there  are  6  with  L.  Vitellius  in  Campania  (ibid.  iii.  58),  and 
3  at  Rome  storm  the  Capitol  (ibid.  iii.  78).  If  these  3  were  Praetorian  and 
not  Urban  cohorts,  which  is  probable,  this  leaves  7  only  for  the  force  at 
Narnia.  These  would  be  the  "pars  copiarum  Narniae  relicta"  (ibid.  iii. 
58).  Of  the  original  14  at  Mevania,  therefore,  7  go  back  to  Rome,  making 
with  the  2  left  here  9.  Of  these  9,  6  are  sent  to  Campania  and  3  stay 
behind  in  Rome,  where  also  the  Urban  cohorts  probably  remain  and  take  part 
in  the  defence  of  the  city. 

a  "  Fortuna  quae  Flavianis  ducibus  non  minus  saepe  quam  ratio  admit," 
Tac.  iii.  59. 


2iS    FLAVIAN   INVASION  OF  ITALY   ch.ii 

The  diminished  army  of  defence  now  amounted 
to  some  twelve  thousand  infantry  and  a  handful  of 
cavalry,  some  four  hundred  in  number.  Certainly, 
now  that  the  pass  was  surrendered  to  the  enemy 
and  Mevania  evacuated,  the  position  occupied  at 
Narnia  was  the  best  possible  for  defence.  The 
river  Nar  here  tears  through  a  narrow  rocky  ravine, 
hurrying  south  to  join  the  Tiber,  and  the  Flaminian 
way  spanned  the  valley  by  a  great  bridge  of  three 
enormous  arches.  An  army  posted  above  the 
ravine  might  easily  make  the  passage  of  the  bridge 
most  hazardous  to  an  enemy.1  From  Mevania 
again  two  roads  led  to  the  south — one  the  Flaminian 
way  itself;  one  a  longer  road  running  up  the  vale 
of  the  Clitumnus  to  the  picturesque  little  fortress 
city  of  Spoletium  (Spoleto)  at  the  head  of  it,  and 
thence  crossing  a  low  ridge  to  Interamna  (Terni), 
a  few  miles  higher  up  the  Nar  than  Narnia.  From 
Interamna  it  ran  down-stream  to  join  the  Flaminian 
road  at  this  city.  Thus  an  army  retreating  from 
Mevania  was  bound  to  retire  as  far  as  Narnia  before 
it  made  a  stand  again,  or  its  flank  could  be  turned 
and  its  retreat  intercepted  by  a  force  which  followed 
the  longer  road.  The  cavalry  of  the  Vitellians  were 
pushed  forward  up  this  road  as  far  as  Interamna; 
the  infantry  remained  at  Narnia.  In  this  position 
the  small  army  which  was  the  Emperor's  last  hope 
awaited  the  coming  of  the  enemy. 

C.  The  Capture  of  Rome. — That  coming  was  not 
long  delayed.     Antonius'  cavalry  scouts,  whom  he 

1  So   the  actual   position   taken  up   was  above  the   "  subiectos    Narniae 
campos,"  Tac.  iii.  63. 


SEC.  VII 


FLAVIAN   INVASION  OF  ITALY  219 


had  sent  forward  to  explore  the  pass,  returned  to 
him  at  Fano  with  the  welcome  if  unexpected  news 
that  it  was  clear  of  the  enemy.  The  general  there- 
upon ordered  an  advance,  and  his  troops  made  their 
way  over  the  ridge,  encountering  no  worse  foe  than 
the  snow  of  mid-December  which  lay  upon  the  pass. 
There  was  still  the  fear  lest  the  unoccupied  pass 
should  be  a  snare  of  the  enemy's  setting,  and  that,  as 
the  Flavians  emerged  exhausted  from  the  mountains, 
the  Vitellians  would  fall  upon  them  from  Mevania.1 
But  no  such  danger  was  encountered.  Without  any 
opposition  the  army  crossed  the  pass,  came  down 
into  the  central  Umbrian  plain,  and  advanced  clown 
the  Flaminian  way  as  far  as  the  town  of  Carsulae.J 
Here  the  army  was  halted.  Ten  miles  away  the 
enemy  were  reported  to  be  holding  a  strong  posi- 
tion in  their  front  at  Narnia.  Carsulae  served  the 
Flavians  very  well  for  a  place  of  encampment. 
From  it  two  roads  sloped  gently  down  hill  to  the 
valley  of  the  Nar,  the  one  on  the  right  hand  to 
Narnia,  the  other  on  the  left  to  Interamna,  and  the 
position  commanded  a  wide  and  uninterrupted  view. 
Moreover,  the  countryside  was  friendly  to  them. 
The  retreat  of  the  Vitellians  from  Mevania  had 
convinced  the  flourishing  little  Umbrian  towns  that 
the  Emperor's  cause  was  a  losing  one.  Prosperous 
cities  on  the  line  or  on  the  flanks  and  rear  of  the 
Flavian  advance,  such  as  Foligno,  Spello,  Assisi, 
Todi,  hastened  to  send  supplies.  There  seemed  to 
the  Flavian  leaders  no  need  for  hurry.     The  moun- 

1  For  this  strategical  method  for  defending  a  mountain  ridge  see  above, 
pp.  43-44.  J  Now  iuins  on'y- 


220    FLAVIAN   INVASION  OF  ITALY   ch.h 

tains  had  been  crossed.  Vitellius  might  submit 
without  a  blow,  and  surely  there  had  been  enough 
of  bloodshed  and  rapine.  The  sack  of  Cremona,  it 
was  felt,  had  sorely  besmirched  their  fair  name. 
Heaven  forfend  that  Rome  herself  should  run  any 
such  risk,  if  Vitellius  would  spare  both  her  and 
himself  by  making  terms  with  the  conqueror  while 
there  was  yet  time.  Indeed  there  seemed  a  good 
hope  that  the  war  might  end  in  this  way.  Ves- 
pasian's elder  brother,  Flavius  Sabinus,  had  all 
through  these  months  of  war  stayed  unmolested  in 
Rome.  He  had  been  thirty-five  years  in  the  public 
service,  of  which  seven  were  spent  by  him  as 
Governor  of  Moesia,  twelve  as  prefect  of  the  city. 
A  man  much  esteemed  and  honoured,  he  was  far 
from  ambitious,  and  would  welcome  a  peaceful  and 
friendly  settlement.1  In  Rome  also  was  Vespasian's 
younger  son  Domitian,  now  a  lad  of  eighteen 
summers.  He,  too,  might  serve  as  a  pledge  of 
friendliness.  Negotiations  between  the  Emperor 
and  Sabinus  were  already  afoot.  Vitellius'  lethargy 
seemed  now  not  unlikely  to  be  his  salvation.  The 
Flavians  had  little  fear  of  him  personally,  nor  was 
he  himself  unwilling  to  lay  aside  the  cares  of 
Empire  and  the  dangers  of  an  unstable  princedom 
for  a  modest  competence  which  should  secure  him 
comfort  and  good  fare  for  the  rest  of  his  life  in 
some  luxurious  Campanian  country-house.  It 
would  not  assist  the  hopes  of  so  genial  a  settlement 
if  the  Flavian  army  advanced  in  hostile  guise  up  to 
the  city.     But  the  Emperor  must  be  made  to  realise 

1  Cf.  Tac.  iii.  75. 


sec. vir  FLAVIAN   INVASION  OF  ITALY  221 

that  he  had  no  chance  of  prolonging  a  successful 
resistance.  "  Out  of  the  whole  world  nothing  was 
left  to  him  save  the  land  which  lay  between  Narnia 
and  Tarracina."  l  Yet  still  he  might  be  tempted  to 
put  trust  in  his  armies.  It  was  indeed  hard  to 
drive  sheer  facts  home  into  a  brain  so  dull. 

For  some  days  Antonius  remained  in  camp  at 
Carsulae.  He  was  waiting  for  the  arrival  of  the 
main  legionary  force,  which  was  following  in  the 
steps  of  his  advance  guard  over  the  pass.  Until 
this  arrived,  he  had  no  wish  to  provoke  a  fight. 
Its  coming  made  his  army  strong  enough  for 
any  enterprise  and  alarmed  the  enemy  at  Narnia. 
The  hearts  of  the  defending  force  sank  within 
them.  The  sailor  legion  was  raw  and  had  seen 
no  fighting ;  the  Praetorians  must  have  known  that 
events  in  Rome  were  tending  towards  peace. 
Theirs  is  indeed  a  hard  fate  who  without  need  or 
gain  die  upon  the  last  battle-field  in  the  war — 
fighting  when  all  reason  for  fighting  is  ended,  slain 
in  the  darkness  of  the  valley  when  the  day  of  peace 
is  already  dawning  on  the  mountain-tops.  It  needed 
but  a  little  to  turn  the  scale  of  the  defenders'  waver- 
ing. From  the  hill  at  Carsulae,  Varus  and  his 
cavalry  rushed  upon  the  enemy's  horse  at  Interamna 
and  dispersed  it  utterly.  The  infantry  at  Narnia 
were  isolated.  Antonius  again  and  again  sent 
offers  of  welcome  and  good  treatment  to  the  officers 
if  they  would  submit,  and  these  passed  over  one 
by  one  to  the  Flavian  camp.  Still,  however,  the 
common  soldier  remained   stubbornly  loyal   to  his 

1  Tac.  iii.  60. 


222    FLAVIAN   INVASION  OF  ITALY   ch.h 

Emperor.  His  own  prefects  and  officers  might 
desert  him,  but  he  trusted  yet  that  his  old  general 
Valens  with  a  new  army  from  Germany  would 
suddenly  appear,  coming  down  the  ridge  to  his 
succour.  To  this  hope  Antonius  made  a  grim 
reply.  He  sent  and  beheaded  Valens  in  his  prison 
at  Urbino.  As  the  general  Claudius  Nero  in  old 
days  had  hurled  into  Hannibal's  camp  the  head  of 
Hasdrubal  his  brother,  so  now  Antonius  sent  the 
head  of  Valens  to  the  force  at  Narnia.  Then  the 
Vitellian  soldiers,  long  hoping  against  hope,  saw  at 
last  that  such  hopes  were  vain.  They  surrendered, 
but  with  honour.  Proudly,  in  military  array,  with 
standards  and  colours  and  all  the  panoply  of  war, 
they  marched  down  from  their  lines  on  the  hillside 
to  the  plain  beneath,  where  the  Flavian  army  was 
drawn  up  in  battle  order  to  receive  them  between 
their  lines.  Antonius  spoke  kindly  to  them,  and 
bade  them  remain  at  Narnia  and  Interamna.  He 
left  with  them  some  of  his  troops — as  many  as  would 
be  able  to  suppress  any  rising  on  their  part,  yet  not 
so  many  as  could  terrify  or  maltreat  them.  For  the 
remainder  of  his  force  the  path  to  Rome  lay  open. 

With  this  he  now  moved  forward  a  few  miles 
farther  down  the  Flaminian  way  to  Ocriculum 
(Otricoli),  a  place  in  the  Tiber  valley,  thirty-five 
miles  north  of  Rome.  The  days  of  the  Saturnalia, 
the  great  December  festival  of  the  Romans,  were 
at  hand,  and  Antonius  determined  that  his  army 
should  celebrate  the  feast  at  this  small  town.  The 
news  from  Rome  was  promising  :  the  Emperor,  as 
soon   as   he  had  heard  of  his  army's  surrender  at 


sec. mi  FLAVIAN  INVASION  OF   ITALY  223 

Narnia,  had  practically  consented  to  resign,  and  to 
entrust  the  government  at  Rome  to  Sabinus  until 
Vespasian  should  come  himself.  The  soldiers  were 
better  kept  away  from  the  city,  and  the  war  seemed 
ended.  But  though  Antonius  himself  kept  the 
infantry  in  camp  at  Ocriculum,  he  sent  a  thousand 
cavalry  forward  under  the  command  of  a  kinsman 
of  Vespasian,  O.  Petilius  Cerialis.  Cerialis  was  a 
tried  soldier,  who  eight  years  before  had  seen  service 
in  Britain  as  legate  of  the  Ninth  legion.  He  had 
just  escaped  from  Rome  disguised  as  a  rustic, 
and  had  joined  Antonius  during  the  march  over 
the  mountains.  As  an  officer  of  experience  (whose 
chief  fame,  however,  was  speedily  to  be  won  in  the 
far  north),  he  could  be  trusted  to  lead  the  cavalry 
forward.  But  so  small  seemed  the  need  for  haste, 
that  Antonius  bade  him  ride  by  cross-roads  to  the 
Via  Salaria  and  enter  Rome  by  the  Colline  gate,  to 
which  this  road  led,  instead  of  following  the  direct 
road  by  the  Flaminian  way.1  By  passing  through 
this  gate  the  cavalry  might  attract  less  notice,  and 
the  risk  of  opposition  or  disturbance  be  lessened. 
The  greater  time  which  this  route  would  require 
seemed  not  worth  consideration,  when  all  at  Rome 
was  said  to  be  so  quiet. 

The  mistake  made  by  the  Flavian  commanders 
led  to  the  gravest  and  most  terrible  results.  Had 
they  arrived  outside  the  gates  of  the  city  forty-eight 

1  Perhaps  Cerialis  was  sent  off  earlier,  from  Narnia,  when  he  could  reach 
the  Via  Salaria  by  marching  to  Interamna,  climbing  the  height  by  the  mag- 
nificent falls  of  Terni  to  the  valley  of  the  Velinus,  the  "  Rosy  Vale,"  and  so 
to  Rieti,  at  the  great  bend  of  the  Salarian  way.  This  road  crosses  the 
Apennines  from  Ascoli  on  the  Tronto  to  Antrodoco  and  Rieti.  It  pursues  a 
mountainous  route,  and  is  rough  going  as  far  as  Rieti. 


224    FLAVIAN   INVASION  OF  ITALY   ch.h 

hours  earlier,  their  strength  would  have  been  enough 
to  overawe  the  numerous  partisans  of  Vitellius  within 
the  walls,  and  the  indignation  which  these  in  ever- 
increasing  measure  felt  for  the  Emperor's  pusillan- 
imity would  without  doubt  have  been  checked. 
The  stormy  day  would  have  sunk  to  rest  in  a  calm 
and  tranquil  evening.  But  the  Fates  willed  that 
Vitellius'  sun  should  set  in  gloom  and  raging  storm, 
in  a  consuming  fire  of  slaughter  and  grim  vengeance. 

Antonius  and  Cerialis  had  failed  to  realise  the 
passionate  anger  of  Vitellius'  soldiers  in  the  city, 
their  dogged  determination  of  despair.  On  the 
morning  of  the  19th  of  December  a  messenger 
rode  at  full  speed  into  the  Flavian  camp  at  Ocricu- 
lum,  demanding  to  see  Antonius  at  once.  He  was 
the  bearer  of  tidings  sent  the  evening  before  from 
Rome,  thirty-five  miles  away.  From  him  Antonius 
learnt  that  the  day  before,  while  he  and  his  men 
were  keeping  jollity,  the  soldiery  and  mob  in  Rome 
had  risen,  compelling  Vitellius  to  do  after  their 
own  pleasure  :  that  Sabinus,  Vespasian's  brother, 
Domitian,  his  son,  and  a  little  band  of  adherents, 
were  blockaded  and  besieged  in  the  Capitol  by  a 
Vitellian  rabble,  howling  for  their  blood.  If  he 
would  save  them  from  massacre,  Antonius  must 
march  at  full  speed  to  the  rescue.  Even  so,  it 
might  be  too  late. 

Not  a  moment  was  lost.  The  Flavian  general 
issued  orders  for  an  instant  march,  and  he  and  his 
army  hastened  at  top  speed  for  Rome.  There  was 
life  to  save  and  treachery  to  be  requited.  The  fall 
of  evening  never  stopped  them,  and  it  was  deep 


sbc. vii  FLAVIAN  INVASION  OF  ITALY  225 

night  when  they  reached  a  point  on  the  road  known 
as  "  The  Red  Rocks,"  six  miles  only  from  Rome. 
They  had  marched  nine  and  twenty  miles  without 
ceasing.  But  all  their  speed  was  vain.  There 
the  news  came  to  them  that  all  was  over.  The 
Capitol  had  been  stormed  that  day,  and  its  garrison 
cut  to  pieces.  Sabinus  had  been  taken  and 
butchered  in  cold  blood  at  Vitellius'  very  feet. 
The  great  temple  itself,  the  glory  of  Rome,  tower- 
ing up  to  heaven  on  its  sheer  rock  with  the  busy 
Forum  at  its  feet,  the  home  of  the  greatest  of  the 
gods  of  Rome,  had  been  destroyed  by  fire.  Glutted 
with  blood  and  fury,  the  Vitellians  had  manned  the 
walls  and  held  the  city. 

The  army  of  rescue  became  an  army  of  venge- 
ance. Swift  and  keen  as  a  beast  of  prey  terrible 
in  his  wrath,  the  soldiers,  when  morning  dawned, 
leapt  upon  the  doomed  city.  Cerialis,  too,  had 
heard  the  news,  and,  pushing  faster  forward,  had 
been  ensnared  among  the  gardens  and  orchards 
at  foot  of  the  Pincian  hill  by  the  enemy's  horse 
and  foot,  routed  with  loss,  and  pursued  back  for 
some  miles  as  far  as  Fidenae.  But  when  the 
pursuit  drew  off,  he  rallied  his  troopers  and  ad- 
vanced again  towards  the  city.  There  was  not  a 
man  in  the  Flavian  army  who  would  lightly  now 
withhold  his  hand  from  the  work  to  be  done. 
Messengers  came  out  from  the  city  to  Cerialis  and 
Antonius,  speaking  of  terms  to  be  agreed  upon. 
Cerialis  sent  them  hurrying  back,  answering  with 
scorn  and  insult,  Antonius  with  courtesy  but  with 
equal  firmness.      No  truce  was  henceforth  possible. 

Q 


226    FLAVIAN  INVASION  OF  ITALY   ch.h 

A  Stoic  philosopher  judged  it  the  time  to  preach 
to  the  troops  of  mercy  and  of  peace.  His  "  un- 
timely wisdom  "  came  near  to  costing  him  his  life, 
and  he  was  contemptuously  brushed  aside.  The 
Vestal  Virgins  came  in  procession  from  the  city  to 
bear  to  the  Flavian  general  a  letter  from  Vitellius, 
begging  the  respite  of  a  single  day.  He  sent  the 
Vestals  back  with  all  honour,  but  instantly  refused 
the  Emperor's  request.  The  murder  of  Sabinus, 
the  burning  of  the  Capitol,  had  made  the  war  a 
"  truceless  war  "  for  ever. 

The  army  swept  forward  over  the  Mulvian 
Bridge1  which  crossed  the  Tiber.  Antonius  here, 
it  is  said,  would  have  halted  them  awhile,  but  his 
men  brooked  henceforth  no  restraint.  They  moved 
to  the  assault  in  three  columns  of  attack,  the 
cavalry  leading  and  driving  back  the  Roman  mob 
before  them.  The  centre  column  advanced  by  the 
main  road  upon  the  gate  under  the  Pincian  hill.2 
To  its  right,  another  column  moved  along  the  bank 
of  the  river  to  storm  the  wall.  To  the  left  the 
third  column  moved  round  outside  the  wall  to  the 
Salarian  way  to  assault  the  Colline  gate,  the  scene 
of  Sulla's  desperate  battle  a  century  and  a  half 
ago.3  The  Vitellians  defended  gates  and  walls 
with  the  fierce  courage  of  men  who  knew  that  for 
them  there  was  henceforth  no  pity.  From  the 
garden  walls  of  the  "  Hill  of  the  Gardens,"  to-day 
the  pleasure  resort  of  the  Romans,  the  defenders 

1  Now  the  Ponte  Molle,  two  miles  from  the  gate. 

2  Now  the  busy  Porta  del  Popolo. 

3  Hard  by  the  Porta  Pia,  the  scene  of  the  far  more  famous  entry  of  the 
troops  of  the  kingdom  of  Italy  into  Rome  on  September  20,  1870. 


sec. vn  FLAVIAN  INVASION  OF  ITALY  227 

hurled  javelins  and  stones  upon  the  Flavian  troops, 
who  struggled  in  a  network  of  lanes  upon  its  outer 
slope  and  suffered  heavy  loss.  Here  the  assailants 
gained  no  ground  till  late  in  the  day.  But  then 
the  cavalry  forced  at  last  an  entrance  by  the  Colline 
gate,  and  rode  round  to  take  the  enemy  in  the  rear. 
The  Vitellians  broke  and  fled,  and  the  hill  was 
carried.  Meanwhile  their  comrades  of  both  the 
other  columns  had  also  forced  an  entrance,  and 
pushed  through  to  the  Campus  Martius,  fighting 
their  way  forward  inch  by  inch.  At  last  the 
Flavians  were  inside  the  very  walls  of  Rome. 

But  still  the  soldiers  of  Vitellius  fought  with 
fury,  as  they  fell  slowly  back  along  the  narrow 
streets.  The  unarmed  citizens  of  Rome  crowded 
to  look  on  as  at  some  gigantic  gladiatorial  con- 
test waged  for  their  marvelling  and  applauding. 
Wounded  and  fleeing  soldiers  who  sought  refuge 
from  the  pursuer  in  shops  and  houses  by  the  way 
were  hounded  out  again  to  meet  their  doom,  and 
the  base  civilian  reaped  the  harvest  of  the  plunder 
of  the  dead  while  the  legionary  sped  forward, 
always  bent  on  slaughter.  Among  the  heaps  of 
the  slain,  which  cumbered  every  way,  roysterers 
and  harlots  made  merry  in  riotous-glee.  One  last 
stand  was  made  by  the  defenders  at  the  Praetorians' 
Camp  hard  by  the  Porta  Pia.  And  there,  in  one 
last  splendid  sally  out  upon  the  swarming  foe, 
Vitellius'  soldiers  perished  to  a  man,  all  their 
wounds  in  front,  and  their  faces  to  the  enemy. 
Night  fell,  and  Rome  was  taken.1 

1  Tac.  iii.  59-84. 


228    FLAVIAN   INVASION  OF  ITALY   ch.h 

§  8.    The  Death  of  Vitellius 

Thus  the  Emperor  Vitellius  lost  the  greater 
cantle  of  the  world  with  very  sloth  and  gluttony  ; 
he  had  slept  away  kingdoms  and  provinces.  The 
end  of  this  his  pitiable  life  may  be  told  in  the 
words  of  Suetonius,  the  biographer  of  the  Caesars. 
Suetonius'  father  had  fought  in  Otho's  army  at 
Bedriacum  in  the  spring  of  the  year  as  tribune 
of  the  Thirteenth  legion  from  Pannonia,1  and  may 
himself  have  seen  and  told  his  son  the  scene  which 
the  latter  tells  as  follows  :2 

Word  was  brought  unto  him  by  his  espiall  that  the  enemie 
approched.  Immediatly  therfore  shutting  himself  close  within 
a  bearing  chaire,  accompanied  with  two  persons  onely,  his 
baker  and  his  Cooke,3  secretly  hee  tooke  his  way  to  the  Aven- 
tine  hill  and  his  fathers  house :  minding  from  thence  to  make 
an  escape  into  Campania.  Soone  after,  uppon  a  flying  and  head- 
lesse  rumour,  That  peace  was  obtained,  he  suffred  him  selfe 
to  be  brought  backe  to  the  Palace.  Where,  finding  all  places 
solitary  and  abandoned :  seeing  those  also  to  slinke  from  him 
and  slip  away  who  were  with  him,  he  did  about  him  a  girdle 
full  of  golden  peeces  of  coine,4  and  fled  into  the  Porters  lodge, 
having  first  tied  a  ban-dog  at  the  doore  and  set  against  it  the 
bedsteed  and  bedding  thereto.  By  this  time  had  the  Avant 
curriers  of  the  maine  armie  broken  into  the  Palace  :  and  meeting 
noe  bodie  searched  as  the  manner  is,  everie  blind  corner.  By 
them  was  hee  plucked  out  of  his  lurking  hole :  and  when  they 
asked  who  he  was  (for  they  knewe  him  not),  and  where  upon 
his  knowledge  Vitellius  was,  he  shifted  them  of  with  a  lie :  after 

1  Suetonius,  Otho,  io. 

2  Suetonius,  Vitellius,  16,  17.     I  use  the  translation  of  Philemon  Holland, 
A.D.  1606.     The  notes  to  the  translation  are  his,  not  mine. 

3  "That  made  his  deinty  pastry  works  and  sweet  meates  :  meete  grooms 
to  accompanie  such  a  glutton." 

4  "15  shilling  peeces  and  better.'' 


sEcvm  FLAVIAN  INVASION  OF  ITALY  229 

this,  beeing  once  knowen,  hee  intreated  hard  (as  if  he  had 
somewhat  to  deliver  concerning  the  life  and  safetie  of  Vespasian) 
to  be  kept  sure  in  the  mean  season,  though  it  were  in  some 
prison :  and  desisted  not  untill  such  time  as  having  his  hands 
pinnioned  fast  at  his  backe,  an  halter  cast  about  his  necke,  and 
his  apparell  torne  from  his  bodie,  he  was  haled  halfe  naked  into 
the  Forum.  Among  many  skornefull  indignities  offered  unto 
him  both  in  deede  and  word  throughout  the  spatious  street 
sacra  via  from  one  end  to  the  other,  whiles  they  drew  his  head 
backward  by  the  bush  of  his  haire  (as  condemned  malefactours 
are  wont  to  be  served)  and  set  a  swordes  point  under  his  chinne, 
and  all  to  the  end  he  might  shew  his  face  and  not  holde  it 
down :  whiles  some  pelted  him  with  dung  and  durtie  mire,  others 
called  him  with  open  mouth  Incendiarie1  and  Patinarium:2  and 
some  of  the  common  sort  twitted  him  also  with  faults  and 
deformities  of  his  bodie  (for,  of  stature  he  was  beyond  measure 
tall :  a  red  face  he  had,  occasioned  for  the  most  part  by  swilling 
in  wine,  and  a  grand  fat  paunch  besides :  hee  limped  somewhat 
also  by  reason  that  one  of  his  thighes  was  enfeebled  withe  the 
rush  of  a  chariot  against  it  what  time  he  served  Caius3  as  his 
henxman  at  a  Chariot  running) :  and  at  the  last  upon  the  staires 
Gemoniae  with  many  a  small  stroke  all  to  mangled  he  was  and 
killed  in  the  end :  and  so  from  thence  drawne  with  a  drag  into 
the  River  Tiberis. 

One  saying  only  by  him,  as  he  was  led  along 
amid  mockery  and  torment,  and  that  indeed  worthy 
of  a  man,  was  recorded,  when  to  a  tribune  who 
stood  insulting  him  he  answered,  "  Yet  once  I  was 
your  Emperor."4  And  presently,  on  the  very  spot 
where  two  days  before  the  body  of  the  murdered 
Sabinus  had  lain,  Vitellius,  too,  lay  dead. 

Rome    was    taken    on    the    21st    of    December 

1  "Or  firebrand,  because  he  burnt  the  Capitoll." 

2  "Or   Platter  Knight,  for  his  gormandize  and  huge  platter  aforesaid." 
See  above,  p.  132. 

3  Caligula.  *  Tac.  iii.  85. 


23o    FLAVIAN   INVASION  OF  ITALY   ch.h 

a.d.  69.  Within  a  very  short  while  the  Vitellian 
army  in  Campania  laid  down  its  arms,  and  Lucius 
Vitellius  was  put  to  the  sword.  The  ten  months' 
fighting  was  ended.  It  remained  for  the  wise  and 
thrifty  Vespasian  to  heal  the  wounds  which  that 
bitter  civil  strife  had  cut  so  deep  into  the  body  of 
the  Roman  State. 


CHAPTER    III 

THE    REBELLION    ON    THE    RHINE 

Strong  heart  with  triple  armour  bound, 
Beat  strongly,  for  thy  life-blood  runs, 
Age  after  Age,  the  Empire  round — 
In  us  thy  Sons, 

Who,  distant  from  the  Seven  Hills, 
Loving  and  serving  much,  require 
Thee, — thee  to  guard  'gainst  home-born  ills, 
The  Imperial  Fire  ! 

Rudyard  Kipling. 

§  i.    The  Tribes  of  the  "Low  Countries  " 

The  Rhine  is  the  only  great  river  of  Europe  which, 
although  not  absorbed  into  a  larger  stream,  yet 
fails  to  keep  its  name  as  far  as  the  sea  for  at  least 
the  greater  bulk  of  its  waters,  and  thus  all  but 
loses,  as  it  were,  its  own  identity.  After  a  course 
of  some  five  hundred  miles  from  the  Lake  of  Con- 
stance, of  which  more  than  four  hundred,  from 
Basle  northwards,  have  lain  entirely  in  German 
territory,  the  river  a  mile  or  two  below  Emmerich 
crosses  the  Dutch  frontier  and,  almost  immediately 
dividing  into  two  channels,  surrenders  its  name. 
The  northern  channel,  called  the  Lek,  flows  by 
Arnhem    to    Rotterdam ;    the    southern    channel, 

231 


232      REBELLION  ON  THE  RHINE     cH.ra 

called  the  Waal,  by  Nymwegen *  to  Dordrecht. 
Fifteen  miles  above  Dordrecht  the  river  Maas 
enters  the  Waal  from  the  south,  and  the  combined 
stream  is  called  the  Merwede  as  far  as  that  town. 
Here  at  Dordrecht  the  southern  stream  is  again 
divided.  A  broad  northerly  channel,  the  De  Noord, 
flows  to  join  the  Lek  a  few  miles  above  Rotterdam, 
and  this  channel  from  that  point  to  the  German 
Ocean  takes  the  name  of  the  Maas  again.  The 
southerly  stream  from  Dordrecht  to  the  sea  by 
Briel  is  called  the  Oude  Maas.  The  actual  name 
of  the  Rhine  clings  only  to  a  small  channel  leaving 
the  Lek  by  Wyk,  and  called  the  "  Crooked  Rhine  " 
— Kromme  Rhyn.  This  at  Utrecht  again  divides : 
one  branch,  now  called  the  Vecht,  flowing  north  to 
the  Zuider  Zee ;  the  other,  under  the  name  of  the 
"  Old  Rhine  " — Oude  Rhyn — passing  by  Leiden  to 
the  North  Sea  at  Katwyk.  A  sketch  plan  may 
serve  to  illustrate  these  divisions  of  the  stream  for 
the  last  hundred  miles  of  its  course  from  Emmerich 
to  the  sea.2 

The  land  enclosed  by  the  two  arms  of  the  Rhine, 
the  Lek  on  the  north  and  the  Waal  on  the  south, 
measuring  some  sixty  miles  in  length  and  about 
twelve  at  its  greatest  breadth,  was  known  to  the 
Romans  as  the  "  Island  of  the  Batavians  " — Insula 
Batavorum.  These  folk  were  a  German  tribe  who 
originally  counted  as  part  of  the  larger  tribe  of  the 
Chatti,    who   dwelt    chiefly    north    of   the    Taunus 


1  In  Dutch  "  Nijmegen." 

2  No  attempt  is  made  to  show  the  actual  course  of  the  streams,  whose 
windings  are  innumerable,  but  distances  are  roughly  to  scale. 


9QM 

81  ^ 

r      *•• 

H  I 

O 
JS    < 

<■"  z    , 
'c  %  A 

s 

3 

< 

S0S 

cc 

o 

o 

f         "**\ 

#tt     UB 

s 

D 

r^       ' 

a 

- 1 

[CO       t 

0. 

< 

1    «Dd 

1 

ITfnuJ 

'd] 

1  bJao 

—  1 

< 

»4 

*<> 

co  "Zl-O 
Z         6}>-» 

Uj  \ 

ffiC# 

z       ** 

Uj      * 

k/<5 

iV 

N 

Jl 

A  — 

43 

■a 

*  A 

<        ^ 

Q 

wl 

3          ^ 

*<y 

*1 

**      i 

<n 

5r 

0  U2 

5a    r 

*Jr\ 

z 

f 

JU- 

N   / 

<& 

s 

1) 

3 

o 

if 

0 

In 

On 

#8 
1 °° 

111 

o 
Q 

\  /   '5 

m  Wm 

§K°M 

, 

:>>    ^v 

•E          ^v 

a: 

*.                 ^v 

VI 

*                       X 

i» 

^C 

•a 

a 

v  /°<a 

<*: 

Mj 

X 

•   fc: 

© 

CO 

o 

- 

V      S3 

3S 

ac 

I  5 

II  li 

<DQ 

234      REBELLION  ON  THE  RHINE     ca.ni 

mountains  by  the  upper  waters  of  the  Lahn.1  But 
quarrels  at  home  had  driven  the  "  Batavians"  to 
take  up  their  goods  and  chattels  and  wander  off  to 
the  north-west,  until  they  settled  in  the  "  Island  " 
and  westwards  of  it  as  far  as  the  sea.  There  they 
were  when  Julius  Caesar  heard  of  them,  and  there 
they  have  remained  ever  since,  still  as  in  Tacitus' 
day  "  famous  for  valour."  The  Dutch  name 
"  Betuwe  "  for  the  land  in  the  eastern  part  of  the 
"  Island "  preserves  their  name,  as  their  de- 
scendants continue  to  preserve  their  independence 
of  their  neighbours  and  remote  kinsmen,  the 
Germans.2 

Side  by  side  with  them  in  their  "  Island," 
probably  in  the  western  part  of  it,  there  dwelt 
another  tribe  of  close  kinship  with  them,  speaking 
the  same  language,  and  not  inferior  to  them  in 
courage,  though  fewer  in  numbers.  These,  by 
name  the  Cannenefates,3  seem  to  have  spread  also 
northwards  along  the  narrow  strip  of  land  between 
the  Zuider  Zee  and  the  ocean,  from  Amsterdam  to 
Helder  in  "  North  Holland,"  if  the  Dutch  name  for 
the  coast  here,  viz.  Kennemerland,  keeps  their 
memory.  North-east  of  the  Zuider  Zee  and  along 
the  marshes,  dunes,  and  islands  of  the  coast  dwelt 
the  Frisii,  a  hardy  race  of  cattle-breeders  and  fisher- 
folk,  as  are  their  descendants,  the  men  of  Fries- 
land,  to-day.      In   Germany   and   Gaul   tribe  after 

1  Now  Hessen-Nassau. 

2  Caesar,  E.G.  iv.  io;  Tacitus,  Germania,  29;  Histories^  iv.  12. 

3  Cannenefates,  Heraeus  and  Mommsen  ;  Canninefates,  Halm  ;  but  the 
e  occurs  several  times  in  the  Medicean  MS.  and  ap.  Pliny,  N.H.  iv.  101. 
In  inscriptions  the  name  is  spelt  in  at  least  five  different  ways,  but  Cannene- 
fates seems  the  most  common.     Cf.  Ruggiero,  Diz.  Epig.  ii.  p.  80,  sub  voc. 


sec.  i      REBELLION  ON  THE  RHINE       235 

tribe  have  wandered  over  the  country,  and  the 
history  of  these  lands  is  a  veritable  kaleidoscope  of 
races.  But  the  great  gift  of  the  German  rivers  to 
their  children  in  the  Low  Countries  has  been  so 
great  a  security  from  enemies,  owing  to  the 
difficulty  of  attack  and  the  poverty  in  earlier  days 
of  the  plunder  to  be  won,  that  their  sturdy  valour, 
already  famous  in  Roman  days,  has  known  how  to 
maintain  them  in  unconquerable  possession  of  their 
still  quiet,  land  of  slow -moving  streams.  And 
the  Romans,  perhaps  better  than  any  race  since 
their  day,  knew  how  to  make  of  these  peoples 
faithful  and  useful  allies  rather  than  ever  bitter 
foes.  The  Batavians  were  not  called  upon  to  pay 
tribute,  but  supplied  as  many  as  one  thousand 
cavalry  and  nine  thousand  infantry  to  the  Roman 
army.  The  eight  cohorts  of  Batavians  attached  to 
the  Fourteenth  legion  were  stalwart  if  quarrelsome 
troops,  proud  of  their  nation  and  of  indomitable 
courage.  The  Imperial  bodyguard  itself,  which 
protected  the  person  of  the  Emperor  at  Rome,  was 
formed  of  men  of  this  tribe.  They  were  com- 
manded by  their  own  nobles  and  not  by  Roman 
officers,  and  by  virtue  of  this  privilege  also  the 
Batavian  regiments  ranked  high  among  the  auxiliary 
troops  of  the  Roman  army.  Their  kinsmen,  the 
Cannenefates,  in  like  manner  paid  no  money  to 
the  Roman  treasury  but  gave  men  to  the  army. 
The  Frisii  sent  hides  yearly  by  way  of  tribute,  and 
men  as  well.  These  also  had  been  faithful  allies 
of  Drusus  and  Germanicus  in  their  wars  over  the 
Rhine  under  the  Emperors  Augustus  and  Tiberius. 


236      REBELLION  ON  THE  RHINE     ch.ih 

But  they  were  farther  away  from  the  Roman 
influence,  and  rose  in  revolt  in  the  year  a.d.  28. 
Since  that  time  they  had  given  no  small  trouble  to 
the  Romans  ;  and  though,  later,  Corbulo,  the  first 
general  of  his  time,  had  punished  them  severely, 
the  greater  part  at  least  of  the  Frisii  remained 
independent  of  Rome  after  the  Emperor  Claudius 
had  withdrawn  all  Roman  troops  to  the  west  bank 
of  the  Rhine.  Their  neighbours  to  the  east,  the 
Chauci,  a  tribe  of  very  great  size,  resident  between 
the  Ems  and  the  Elbe,  mariners  and  fishermen, 
owned  no  allegiance  to  Rome.1  But  at  least  the 
"Island"  counted  as  being  within  the  frontiers  of 
the  Roman  Empire.  For  Drusus'  great  engineer- 
ing works,  made  in  9  B.C.,  had  done  much  to  secure 
this.  Under  his  direction  the  Roman  army  of  the 
Lower  Rhine  had  in  that  year  constructed  both  a 
Fosse  and  a  Mole.  The  mole  or  "Agger"  was 
thrown  out  into  the  Rhine  from  the  left  bank  just 
above  the  parting  of  the  channels  below  Emmerich, 
not  far  from  Cleve.2  By  this  the  greater  bulk  of 
the  river's  water  was  directed  into  the  northern 
arm,  the  Lek,  and  thus  the  "Island"  was  easily 
reached  from  the  Roman  shore,  while  it  was 
separated  from  the  tribes  over  the  Lek  by  a  great 
mass  of  water.  The  use  of  this  Agger  was 
great,  and  it  was  strengthened  again  by  Pompeius 
Paulinus  in  a.d.  55.  The  fosse  was  dug  from  the 
Lek,  a  mile  or  two  above  Arnhem,  to  the   river 


1  Cf.  Tacitus,  Germania,  35. 

2  Cleve,  probably  the  "  Oppidum  Batavorum  "  of  Hist.  iv.  19,  was  then 
<>n  the  Rhine  bank  ;  it  is  now  some  distance  away. 


sec.  ii     REBELLION  ON  THE  RHINE       237 

Yssel,  upon  the  course  of  the  "  New  "  or  "  Guelders  " 
Yssel,  and  thus  gave  the  Roman  flotilla  upon  the 
Rhine  access  to  the  Zuider  Zee.  This  was  equally 
useful  for  the  purpose  of  any  hostile  operations 
against  the  Frisii,  and  for  interrupting  in  case  of 
need  the  communications  of  the  Batavians  and 
Cannenefates  with  the  Frisii  and  Chauci.1  By  such 
means,  and  by  requiring  of  them  service  in  the 
Imperial  army,  the  Romans  for  long  years  kept  a 
grip  over  the  two  tribes  of  the  Island.  And  their 
Gallic  neighbours  on  the  south,  chief  of  whom  were 
the  Nervii  in  modern  Belgium,  had  for  long  years 
past  been  fully  subject  to  Rome. 

Yet  once,  in  the  years  a.d.  69-70,  a  great  storm 
of  mutiny  of  the  native  troops  in  the  army,  and 
of  rebellion  among  the  tribes  in  these  the  Roman 
11  Low  Countries,"  broke  upon  the  Roman  dominion 
and  all  but  overthrew  it.  It  was  truly  the  "  Indian 
Mutiny  "  of  Roman  history. 


§  2.   The  Cause  of  the  Rebellion 

The  cause  of  this  rising  was  the  natural  love 
of  independence  and  of  liberty  which  was  felt  by 
the  German  tribes  who  came,  however  remotely, 
under  Roman  influence.  In  its  beginnings, 
indeed,  the  real  meaning  of  the  war  was  hidden 
under  the  disguise  of  a  movement  in  favour  of 
Vespasian  against  the  Emperor  Vitellius,  and 
the    rising    was    not    only    encouraged,    but    even 

1  The  Fosse,  Tac.  Ann.  ii.  8.  I.     The  Agger,  Tac.   Ann.   xiii.    53.  3  ; 
Hist.  v.  19.  3. 


238      REBELLION  ON  THE  RHINE     ch.hi 

directly  promoted,  by  the  Flavian  leaders.  None 
the  less  this  opportunity  so  recklessly  given  to  the 
tribes  was  from  the  first  but  the  mere  pretext  or 
occasion  of  their  fighting.  From  the  beginning  of 
the  revolt  among  the  Batavians  to  their  final  proud 
submission,  the  war  was  an  armed  plea  for  liberty. 
The  use  made  of  this  liberty,  had  the  tribes  won  it, 
would  doubtless  have  been  ferocious  and  barbaric. 
Circumstances  made  of  the  war  "  one  of  the  most 
singular  and  most  dreadful  in  all  ages."1  The 
motives  of  the  leaders  of  the  revolt  were  in  large 
measure  those  of  private  revenge  and  selfish 
ambition.  To  the  annals  of  the  Roman  army  the 
war  contributed  little  but  a  record  of  lamentable 
cowardice  and  dishonour.  And  yet,  in  spite  of  all 
these  undoubted  facts,  the  cause  of  the  war  suc- 
ceeds in  ennobling  its  history  beyond  that  of  the 
other  fighting  of  these  two  wild  years  of  strife. 
Apart  from  the  audacity  or  calmness  of  a  few,  the 
splendid  courage  of  the  Roman  troops  engaged 
upon  both  sides,  or  the  misdirected  loyalty  of 
soldiers  to  their  generals,  there  is  no  sunshine  to 
light  up  the  thick  gloom  which  enwraps  the  civil 
wars  of  Otho  and  Vitellius,  Vitellius  and  Vespasian. 
There  was  no  great  principle  at  stake  in  either 
war,  and  only  such  can  justify  the  appeal  to  arms. 
The  tribes  were  fighting  for  a  principle.  They 
fought  savagely,  ignorantly,  treacherously.  They 
were  happy  even  in  their  ultimate  defeat.  Yet  the 
cause  of  their  rising  did  them  honour. 

That    this    was    the    cause   of    the    rebellion    is 

1  Mommsen. 


sEc.n     REBELLION  ON  THE  RHINE       239 

shown  alike  in  the  peoples  and  in  the  leaders  who 
shared  in  the  enterprise. 

A.  The  Peoples  of  the  Revolt. — The  peoples 
rose  for  liberty.  No  other  battle-cry  would  have 
gathered  to  the  Batavians'  standards  the  other 
tribes  of  the  Rhine,  some  already  subject  to  Rome, 
some  threatened  by  her  in  past  years,  and  eager  for 
revenge  as  well  as  plunder.  The  obvious  weakness 
of  the  Roman  army  on  the  German  frontier,  when 
Vitellius  had  drained  it  of  all  its  best  troops  for  his 
march  to  Italy,  and  the  fierce  internecine  struggle 
raging  in  Italy  itself,  seemed  to  give  the  restless 
German  tribes  a  unique  opportunity  for  rebellion 
and  defiance.  Thus  the  Batavians  and  Cannene- 
fates  were  not  left  long  alone  in  their  endeavour. 
They  presently  found  allies  in  the  Marsaci  at  the 
mouth  of  the  Scheldt.1  The  Frisii  joined  them  at 
once.2  The  tribes  beyond  the  Rhine  seized  the 
chance  offered  them.  The  Tencteri  opposite 
Cologne,  the  Bructeri  on  the  Ems,  shared  in  the 
first  attack  on  Castra  Vetera.3  Higher  up  the 
river,  the  Chatti  from  the  north  of  Taunus,  the 
Mattiaci  from  its  southern  slopes,  the  Usipi  from 
the  lands  opposite  Coblenz,  made  an  early  onslaught 
on  the  great  Roman  fort  at  Mainz.4  The  Chauci, 
the  tribe  lying  to  the  east  of  the  Frisians,  sent  the 
insurgent  leader  aid  not  only  at  the  height  of  the 
struggle,  but  even  again  when  he  seemed  in  his 
last  most  desperate  straits.5  Of  all  the  famous 
German  tribes   beyond  the   Rhine   known  at    this 

1  Tac.  iv.  56.         2  Tac.  iv.  15.  3  Tac.  iv.  23  ;  see  below,  pp.  264-268. 

4  Tac.  iv.  37  ;  see  below,  p.  284.  5  Tac.  iv.  79  ;  v.  19. 


24o      REBELLION  ON  THE  RHINE     ch.ih 

time  to  the  Romans,  not  one  gave  the  latter  any 
help,  and  two  only  took  no  interest  in  the  war. 
The  Cimbri  were  but  a  shadow  of  their  former 
selves ;  the  Cherusci  allowed  their  native  indolence 
free  play.1 

On  the  Roman  side  of  the  Rhine  the  Cugerni 
to  the  west  of  Castra  Vetera  joined  the  insurgents 
at  once,  and  remained  true  to  them  to  the  end.2 
Their  neighbours  the  Ubii,  inhabiting  Cologne  and 
the  surrounding  district,  had  for  some  years  past 
been  a  centre  of  Roman  influence  among  the  wilder 
tribes  over  the  river,  and  were  hated  and  distrusted 
by  these  in  consequence.  But  the  tide  of  German 
successes  swept  even  these  into  the  movement  at 
last,  though  their  motive  was  always  self-preserva- 
tion rather  than  any  active  dislike  of  Rome,  and 
they  returned  to  their  early  allegiance  at  the 
earliest  possible  moment.3 

Finally,  there  dwelt  in  Gaul  a  German  tribe,  the 
Tungri,  far  removed  from  the  fatherland.  These 
had  expelled  the  Gauls  from  the  district  round  the 
present  town  of  Liege,  in  Belgium,  in  the  valley  of 
the  Meuse  (the  Maas  of  Holland) ;  and  the  city  of 
Tongres,  fifteen  miles  to  the  north  of  Liege,  pre- 
serves their  name.  This  folk  supplied  at  least  two 
auxiliary  cohorts  to  the  Roman  army,  and  these 
had  fought  for  the  Vitellians  against  Otho.  But 
the  German  rising  excited  at  once  their  national 
feeling.  One  of  the  cohorts  went  over  to  the 
enemy  in  the  first  engagement ;  the  other  quickly 

1  Cf.  Tac.  Germ.  36,  37.  2  Tac.  iv.  26;  v.  16,  18. 

3  Tac.  iv.  28,  55,  63,  77  ;  v.  24  ;  cf.  Germ.  28. 


sEc.n     REBELLION  ON  THE  RHINE       241 

followed  suit ;  and  the  whole  tribe  threw  in  their 
lot  with  the  rebels.1  Practically  all  "  Germany  "  on 
both  banks  of  the  Rhine  was  at  one  time  in  arms 
against  the  Romans  in  this  war.  Only  a  truly 
"  national  war "  could  have  produced  such  a  unity 
among  so  many  widely  scattered  tribes.  The  call 
to  liberty  found  then  an  instant  answer  in  the 
German's  heart.  The  struggle  between  Roman 
and  Teuton  never  ceased.  "  Tarn  diu  Germania 
vincitur." 2  And  this  most  dangerous  plague  of 
"  nationalism "  spread  even  to  Gaul,  though  here 
the  tribes,  accustomed  for  a  century  to  the  Roman 
rule,  resisted  the  infection  longer,  and  many  suffered 
no  taint  of  it  at  all.  But  the  Gallic  districts  on  the 
north  and  north-east,  which  came  into  close  contact 
with  the  Germans,  could  not  but  be  affected  by  the 
movement. 

The  Belgae,  chief  of  whom  were  the  Nervii, 
famous  foes  of  Julius  Caesar  in  old  days,  were 
neighbours  of  the  Batavians  and  Tungri.  The 
small  tribe  of  the  Baetasii  dwelt  between  the  Nervii 
and  Tungri  in  Brabant,  where  the  village  of  Betz, 
near  Brussels,  recalls  their  name.  The  Moselle 
valley  gave  easy  access  from  the  Rhine  at  Coblenz 
to  the  spread  of  the  disorder  to  its  tribes,  the 
Treveri,  whose  name  is  kept  in  that  of  the  capital 
city  of  the  valley,  Treves,  and,  higher  up  the  river, 
the  Mediomatrici,  in  whose  territory  Metz  lies. 
The  news  of  the  rising  penetrated  the  Ardennes, 
travelling    up   the    long    valley   of  the   Maas,  and 

1  Tac.  ii.  14;  iv.  16,  55,  66,  79  ;  cf.  Germ.  2. 
2  Tac.  Germ.  37. 


242      REBELLION  ON  THE  RHINE     ch.ih 

reached  the  Lingones  at  the  source  of  the  river. 
From  this  tribe  it  could  be  carried  over  the  water- 
shed to  the  valleys  of  the  Saone  and  Rhone,  and 
be  told  to  the  powerful  tribe  of  the  Sequani,  who 
lay  between  these  rivers  and  the  Jura.  And  a  little 
to  the  west  of  the  middle  course  of  the  Maas, 
separated  from  the  Treveri  by  this  river,  the  Remi, 
"  the  leading  canton  in  Belgica,"  x  quickly  received 
tidings  of  the  rebellion  on  the  Rhine. 

But  from  the  days  of  Julius  Caesar  the  tribes  of 
Gaul  had  ever  been  jealous  of  one  another,  and, 
when  even  the  national  hero  Vercingetorix  failed 
to  unite  them  all  in  his  magnificent  struggle  against 
the  Roman  invader,  no  lesser  man  coming  after 
him,  when  the  Romans  held  the  land  in  their 
masterful  grip,  could  achieve  even  the  like  amount 
of  success.  Gallic  nationalism  was  an  ever-present 
peril  to  the  Romans  for  more  than  a  century  after 
the  death  of  its  greatest  champion.  There  was  no 
single  Emperor  from  Augustus  to  Vespasian  who 
was  not  made  aware  of  its  existence.  Yet  the 
attempt  to  win  liberty  was  always  spasmodic,  and 
a  tribe  which  took  up  arms  for  this  cause  at  one 
time  would  be  found  a  few  years  later  resisting  the 
similar  efforts  of  a  neighbouring  canton.  When 
Julius  Vindex  in  the  last  year  of  Nero's  principate 
gathered  round  his  standard  a  hundred  thousand 
Gauls  to  fight  for  Gallic  independence,2  the  Sequani 
were  among  those  who  took  up  arms  for  him  ;  but 


1  Mommsen. 
2  Tac.  i.  1 6,  51,  53,  64  ;  Plutarch,  Calba,  4  and  6  ;  see  my  Life  of  Nero, 
pp.  496-497- 


sec.  ii     REBELLION  ON  THE  RHINE       243 

the  Treveri  and  Lingones  not  only  held  aloof  from 
his  cause,  but  gleefully  assisted  the  Roman  Army  of 
Germany  under  Verginius  Rufus  to  crush  the  rising. 
The  two  tribes  steeped  their  hands  in  the  blood  of 
the  twenty  thousand  slain  at  Vesontio,1  capital  city  of 
their  Gallic  kinsmen,  the  Sequani.  Yet  the  Treveri 
had  fought  for  Florus  and  Sacrovir  when  these 
rose  against  the  Romans  in  the  principate  of 
Tiberius.2  And  little  more  than  a  year  had  passed 
since  the  death  of  Vindex,  when  the  policy  of  these 
three  tribes  was  to  be  completely,  almost  ludi- 
crously, reversed  ;  when  the  Treveri  and  Lingones 
were  to  be  found  among  the  foes,  the  Sequani 
among  the  friends,  of  Rome.  The  German  leader 
had  every  reason  to  declare  with  emphasis  that 
"  Gaul  had  fallen  by  its  own  strength."  3  But  all 
his  urging  of  this  home-truth  was  unable  to  get  the 
mastery  of  this  ineradicable  tendency  of  the  Celt. 
The  great  German  historian  of  our  own  day  cannot 
refrain  from  gibing  at  this  characteristic  weakness 
of  the  Gauls  ;  for  when  the  attempt  was  made  by 
them  in  the  year  70  to  follow  the  example  of  their 
German  allies  and  to  erect  an  Imperium  Galliarum 
independent  of  Rome,  he  labels  it  "  a  tragedy  and 
at  the  same  time  a  farce."  4 

When,  therefore,  the  Gallic  tribes,  their  neigh- 
bours, were  invited  by  the  rebel  Germans  to  take 
part  in  their  enterprise,  their  answer  was  far  from 
unanimous.     At  first  indeed  they  gave  ready,  even 


1  Besancon.  2  Tac.  Ann.  iii.  40,  42. 

3  "  Gallias  suismet  viribus  concidisse,"  Tac.  iv.  17. 

4  Mommsen. 


244      REBELLION  ON  THE  RHINE     ch.ih 

strenuous,  aid  to  the  small  Roman  army  which 
sought  to  make  headway  against  the  rebels.  The 
Treveri  stood  stoutly  on  the  defence  against  the 
Germans,  even  running  a  palisade  and  trench  along 
the  whole  line  of  their  threatened  frontier,  and  con- 
tending vigorously  with  the  assailants.  A  hundred 
years  before,  the  Remi  had  been  Caesar's  most 
faithful  allies,  and  now  no  Roman  disaster  shook 
their  loyalty.  The  Sequani  also  refused  always  to 
join  the  enemies  of  Rome.  But  the  early  triumphs 
won  by  the  insurgents,  and  especially  the  siege  of 
Castra  Vetera,  the  Roman  stronghold  of  the  Lower 
Rhine,  made  the  loyalty  of  some  others  grow  cold. 
They  remembered  the  money  which  they  gave 
each  year  to  the  Treasury,  the  men  whom  they 
were  compelled  to  supply  to  the  armies,  of  Rome. 
The  Treveri,  Lingones,  Baetasii,  and  Nervii,  joined 
the  insurgents.1  The  "hope  of  liberty"  beguiled 
them. 

And  when  once  the  iron  hand  of  the  conqueror 
was  removed,  the  old  struggling  for  primacy 
among  the  tribes  could  be  renewed,  and  more  than 
one  folk  dreamt  of  glory  to  be  gained  at  the 
expense  of  their  kinsmen.'2  Surely  Rome's  end 
seemed  hard  at  hand.  Once  more,  as  in  Caesar's 
day,  the  Druids  preached  rebellion,  proclaiming 
that  the  burning  of  the  Capitol  was  a  sign  from 
Heaven    that    the  Empire   was    departed  for  ever 

1  Tac.  iv.  25,  26,  37  (Treveri)  ;  56  and  77  (Lingones) ;  56  and  66 
(Baetasii) ;  66  (Nervii)  ;  67  (Sequani) ;  67-69  (Remi). 

2  "  Mox  valescentibus  Germanis  pleraeque  civitates  [Gallorum]  adversum 
nos  arma  sumpsere  spe  libertatis  et,  si  exuissent  servitium,  cupidine  imperi- 
tandi,"  Tac.  iv.  25. 


sec. n     REBELLION  ON  THE  RHINE       245 

from  Rome.  And  the  people  were  but  too  ready 
to  be  credulous  when  the  priest  preached  a  holy 
war.1  Not  for  a  moment  was  any  appeal  made  to 
any  Flavian  sympathies  which  perhaps  some  Gauls 
might  have.  But  they  were  bidden  think  of  the 
ills  endured  for  so  many  weary  years,  of  the  "  hap- 
less servitude  which  falsely  they  called  peace." 2 
First  war,  then  liberty,  then  Empire,  but  an 
Imperium  Galliarum,  not  an  Empire  of  Rome 
—  such  was  the  bait  which  ensnared  them.3 
The  miserable  pretext  of  Vespasian's  name  was 
quickly  flung  aside.  The  secret  meeting  of  con- 
spirators held  in  the  private  house  at  Cologne 
agreed  gaily  enough  together :  Liberty  must  first 
be  firmly  rooted  in  the  land ;  then  the  tribes  need 
but  "discuss"  together  the  question  of  the  limit 
they  might  choose  to  set  to  their  exercise  of  power.4 
These  rebels  had  not  even  the  wretched  mockery 
of  a  senile  Emperor  for  whom  to  fight,  as  had  our 
own  sepoys  in  times  past.  "  Far  be  it  from  us," 
cried  Civilis  to  the  Tungri,  "  to  seek  rule  over 
others.  It  is  not  for  this  that  we  have  taken  up 
arms,  that  Batavians  and  Treveri  may  govern  the 
nations.  Far  be  such  arrogance  from  us."  5  When 
the  company  of  urchins  set  out  to  rob  the  pastry 
cook  of  his  richest  cake,  they  fared  out  together 
in  all  amity;  it  was  afterwards  that  they  fell  to 
quarrelling  about  the  largest  slice. 

B.    The  Leaders  of  the  Revolt. — This  cause,  the 

1  Tac.  iv.  54.  2  Tac.  iv.  16,  17.  3  Tac.  iv.  59. 

4   "  Coalita  libertate  disceptaturas  Gallias  quern  virium  suarum  terminum 
velinty'  Tac.  iv.  55.  6  Tac.  iv.  66. 


246      REBELLION  ON  THE  RHINE     ch.hi 

desire  to  be  freed  for  ever  from  the  Romans,  was  pro- 
fessed by  the  leaders  of  the  revolt,  at  first  secretly, 
but  later  on  with  open  frankness.  Their  names 
seem  those  of  Romans  ;  their  military  experience 
had  been  gained  in  Roman  armies  and  Roman 
camps  ;  but  they  were  no  more  Roman  in  sympathy 
than  they  were  in  race.  Long  since,  the  history  of 
Arminius,  somewhat  grotesquely  celebrated  as  the 
"liberator  of  Germany,"  had  shown  that  a  Roman 
education  was  not  the  slightest  guarantee  for  loyalty, 
but  rather  an  inducement  and  encouragement  to 
disaffection  and  secret  treason.  Other  instances 
since  had  confirmed  this  gloomy  fact.  The  native 
chiefs  would  come  to  Italy  for  education,  would 
learn  Roman  manners  and  the  Roman  language, 
and  return  home  to  kindle  rebellion  against  Rome. 
As  early  as  the  first  century  of  our  era  educa- 
tion bred  sedition.  Omne  ignotum  pro  magnifico. 
It  had  been  better  if  the  ruling  nation  had 
been  less  well  known  to  the  princes  of  Germany. 
It  was  easy  for  them  to  see  signs  of  her  weak- 
ness and  vacillation.  It  was  hard  for  them  to 
realise  the  grim  strength  of  courage  and  deter- 
mination which  was  still,  even  after  many  years 
of  Empire,  the  bedrock  of  her  people  and  even  of 
some  among  her  rulers.  So  both  Rome  and  the 
Germans  paid  dearly  for  their  common  mistake. 
But  while  the  common  folk  who  rose  against  Rome 
may  win  men's  admiration,  little  of  this  can  be 
reserved  for  those  who  of  selfish  ambition  played 
upon  their  people's  ignorance,  turned  the  arts  which 
they  had  learnt   from    Rome,  the  skill  which  they 


sec. ii     REBELLION  ON  THE  RHINE       247 

had  there  acquired,  to  the  hurt  of  their  teachers, 
and  therefore  inevitably  added  ingratitude  and 
treachery  to  the  more  venial  charge  of  self-seeking 
which  can  be  brought  against  them. 

The  three  chief  leaders  of  the  Gauls  in  the  revolt 
were  Julius  Classicus  and  Julius  Tutor  of  the 
Treveri,  and  Julius  Sabinus  of  the  Lingones.  The 
one  great  German  leader  was  Julius  Civilis. 

Classicus  was  prefect  of  the  ala  Treverorum, 
which  served  as  an  auxiliary  squadron  in  the  Roman 
army.  Tutor  had  been  appointed  by  Vitellius 
"  prefect  of  the  bank  of  the  Rhine." l  Sabinus 
boasted  that  he  was  great-grandson  of  Julius  Caesar 
himself.2  Yet  there  was  little  Roman  about  these 
men  save  the  name  Julius,  and  this  was  the  com- 
monest of  names  for  the  chieftains  of  Gaul.3  Classicus 
was  the  noblest  by  birth  as  well  as  the  wealthiest 
in  his  tribe.  He  was  a  descendant  of  its  early  line 
of  kings,  and  his  royal  ancestors  had  ever  been  the 
adversaries  rather  than  the  allies  of  Rome.4  Tutor 
fought  the  Romans  to  the  last.  Sabinus  used  his 
supposed  descent  only  to  bid  his  followers  call  him 
Caesar.''     Julius  Civilis,''  above  all,  was  all  the  more 

1  I.e.  to  guard  the  strip  between  the  Nava  and  Moselle  (Heraeus).  The 
general  "cura  ripae  "  belonged  to  Hordeonius  Flaccus,  Governor  of  Upper 
Germany  (Tac.  ii.  57).  2  Tac.  iv.  55. 

3  Cf.  Julius  Florus  of  the  Treveri ;  Julius  Sacrovir  of  the  Aedui  (Tacitus, 
Annals,  iii.  40)  ;  Julius  Vindex,  the  Aquitanian  noble  ;  Julius  Valentinus  of 
the  Treveri  ;  Julius  Auspex  of  the  Remi  {Hist.  iv.  69) ;  Julius  Biiganticus, 
the  Batavian  {Hist.  ii.  22)  ;  Julius  Calenus  of  the  Aedui  {Hist.  iii.  35)  ; 
etc.      Claudius  was  also  a  common  praenomen. 

4  Tac.  iv.  55.  6  Tac.  iv.  67. 

6  According  to  the  Medicean  MS.  of  Hist.  iv.  13,  "Julius  Paulus  et 
Claudius  Civilis."  But  we  have  Julius  Civilis,  ap.  Hist.  i.  59,  and  Frontinus, 
Strateg.  iv.  3.  14.  Hence  the  MS.  reading  in  iv.  13  is  altered  to  read 
"Julius  Paulus  et  Julius  Civilis"  (Halm),  or  "Julius  Civilis  et  Claudius  Paulus" 
(Heraeus). 


248      REBELLION  ON  THE  RHINE     ch.  m 

a  barbarian  at  heart  for  the   Roman  veneer  upon 
him. 

This  man  was  of  royal  Batavian  stock.  He  had 
served  twenty-five  years  in  the  Roman  army,  and 
was,  according  to  his  own  story,  an  early  friend  of 
Vespasian  himself.1  He  was  brave,  eloquent,  and 
a  ready  speaker.  Twice  he  had  run  no  small  risk  of 
his  life  :  once  under  Nero,  when  the  then  Governor 
of  Lower  Germany,  Fonteius  Agrippa,  sent  him  in 
chains  to  Rome  to  answer  a  charge  of  planning 
rebellion  ;  once,  still  more  recently,  when  in  January 
a.d.  69  the  Army  of  the  Rhine  declared  for  Vitellius. 
From  the  former  peril  Galba's  clemency  had  released 
him  ;  from  the  latter  his  own  influence  with  his  tribe. 
For  the  Vitellian  leaders  feared  lest  his  punishment 
should  anger  the  Batavian  cohorts  who  were  at 
that  time  quartered  among  the  Lingones.2  Civilis, 
however,  was  not  a  man  to  forgive  such  insults. 
With  the  craft  of  a  savage  he  hid  his  resentment 
for  a  time,  but  waited  his  opportunity.  This  came 
when  Antonius  Primus,  the  Flavian  leader,  just 
before  his  invasion  of  Italy,  wrote  to  him  inciting 
him  to  cause  the  Batavians  to  revolt,  wishing  to 
hinder  thereby  the  sending  of  any  reinforcements 
to  Vitellius  from  Germany.  Hordeonius  Flaccus, 
the  Governor  of  Upper  Germany,  himself  specially 
appointed  by  Vitellius  to  guard  the  bank  of  the 
Rhine,  urged  the  same  upon  him  in  a  private  inter- 
view.3 Encouraged  by  these  two  Romans,  Civilis 
declared    for    Vespasian,    even    as    Vindex    in    the 

1  Tac.  iv.  13,  32;  v.  26.  2  Tac.  i.  59;  iv.  13. 

3  Tac.  iv.  13  ;  v.  26. 


sec, i     REBELLION  ON  THE  RHINE       249 

preceding  year  had  risen  nominally  on  Galba's 
behalf.  But  just  as  Vindex  had  harboured  other 
thoughts  in  his  heart  when  he  revolted,  so  Civilis 
nourished  a  "  deeper  plan."  1  When  in  November 
Antonius,  now  that  the  battle  of  Cremona  had  been 
won,  sent  again  bidding  him  cease  from  further 
warfare,  Civilis  threw  off  the  mask.  By  refusing 
to  obey,  he  changed  the  outward  aspect  of  the 
Batavians'  revolt  from  participation  in  the  civil  war 
to  that  of  flat  rebellion  against  Rome.  He  had 
never  intended  anything  else  from  the  beginning. 
His  very  personal  deformity — for  he  had  lost  the 
sight  of  one  eye — he  was  wont  to  quote  as  sign  of 
his  enmity  against  the  Romans,  comparing  himself 
in  this  also  to  Hannibal  and  Sertorius,  the  foes  of 
Rome.2  But  neither  Sertorius  nor  Hannibal  was 
guilty  of  the  cruel  tricks  which  delighted  the  German 
savage.  The  massacre  of  the  heroic  little  garrison 
of  Castra  Vetera,  when  they  surrendered  at  last  on 
promise  of  their  lives  and,  disarmed  and  marched 
out  five  miles  along  the  road,  were  there  butchered 
in  cold  blood,  Civilis  professed  to  deplore.  Even 
the  Nana  at  Cawnpore  was  saved  from  this  hypocrisy 
by  drunkenness.  It  was  after  this  deed  that,  to 
celebrate  the  work  of  destruction  which  was,  he 
gleefully  thought,  now  fully  accomplished,  the 
German  barbarian  fulfilled  the  vow  which  he  had 
made   when   he    took    up    arms,   and    for    the    first 

1  The  "  altius  consilium  "  to  which  he  "  studium  partium  praetendit  "  (Tac. 

iv-  x3>  x4)- 

2  Heraeus'  note  is  comic  enough  to  deserve  repetition  :  "  Er  war  einaugig, 
wie  Hannibal  und  Sertorius  (gleichfalls  Todfeinde  Roms),  Ziska  und  Nelson  ; 
vgl.  die  Kemerkung  Plutarch's  Serf.  I." 


250      REBELLION  ON  THE  RHINE 

time  cropped  short  the  hair  which,  stained  red 
with  dye,  flowed  about  his  shoulders.  And  others 
also  should  have  joy  of  his  gallantry  and  triumph. 
To  rejoice  the  heart  of  the  fierce  maiden  pro- 
phetess Veleda,  who  dwelt  amid  the  black  forest 
on  the  river  Lippe's  banks,  one  Roman  legionary 
legate  was  reserved  as  booty  from  the  butchery. 
Happily  his  escort  murdered  him  while  yet  upon 
the  journey  towards  her.  And  for  his  little  son's 
delight,  so  ran  the  tale,  this  Civilis,  this  eloquent 
Roman  soldier  of  twenty-five  years'  service,  set  up 
captives  tied  to  stakes,  to  be  the  mark  for  the 
child's  darts  and  javelins.1  This  was  the  leader  who 
plunged  his  tribesmen  into  desperate  war,  and  above 
all  others  shook  the  structure  of  Roman  dominion 
on  the  Rhine  to  its  lowest  foundations. 


§  3.    The  Roman  Army  on  the  Rhine 

The  usual  garrison  of  Germany,  or  "  Army  of 
the  Rhine,"  consisted  of  eight  legions  and  a  corre- 
sponding number  of  auxiliaries.2  Of  the  legions 
four  were  in  Upper,  and  four  in  Lower,  Germany, 
these  two  provincial  districts  meeting  at  a  point 
about  half-way  between  Bonn  and  Coblenz,  near  the 
village  of  Brohl.3  The  troops  of  each  district  were 
under  the  command  of  a  legate  of  at  least  praetorian 
rank,4  although  the  civil  administration  in  them 
remained  in   the  hands  of  the   Governor  of  Gallia 


1  Tac.  iv.  60,  61.  2  Cf.  Tac.  Ann.  iv.  5. 

3  Cf.  Mommsen,  Provinces,  Eng.  Trans,  vol.  i.  p.   119. 

4  I.e.  who  had  at  least  been  praetor  at  Rome. 


SEC.   Ill 


REBELLION  ON  THE  RHINE      251 


Belgica  until  the  principate  of  Domitian,1  and  the 
collection  of  taxes  was  controlled  by  the  procurator 
of  that  same  province  until  at  least  the  middle  of 
the  second  century  of  our  era.'2  The  Rhine  was 
now  the  limit  of  Roman  military  occupation  and 
administration.  The  Emperor  Claudius  had  with- 
drawn all  Roman  troops  to  the  left  bank  of  the  river 
in  the  lower  province  after  Corbulo's  campaign  in 
a.d.  47,3  and  Dubius  Avitus,  after  a  punitive  expedi- 
tion against  the  Frisii  ten  years  later,  had  again 
retired  behind  the  river.4  In  the  upper  province 
since  the  days  of  Augustus  there  had  twice  been 
trouble  with  the  tribe  of  the  Chatti  in  the  valley  of 
the  Main,  which  enters  the  Rhine  at  Mainz,  and 
two  Roman  armies  had  penetrated  up  the  valley  of 
the  tributary  to  punish  the  natives — the  first  in 
a.d.  41  under  Galba,  who  later  became  Emperor  of 
Rome ; 5  the  second,  nine  years  later,  under  the  poet 
Publius  Pomponius  Secundus.6  But  though  the 
small  tribe  of  the  Mattiaci  under  the  Taunus  hills 
remained  under  Roman  control,  and  Romans  enjoyed 
the  hot  springs  of  Wiesbaden  and  worked  the  silver 
mines  in  the  neighbourhood,7  no  Roman  troops 
seem  to  have  been  quartered  in  the  Main  valley  or 
on  the  right  bank  of  the  Rhine,  except  that  the 
bridge-head    opposite    Mainz    was    occupied    by  a 

1  The  first  known  "legatus  Germaniae,"  as  distinct  from  the  "legate  of 
the  army  in  Germany,"'  is  Javolenus  Priscus,  the  jurist,  in  A.D.  90  (C./.L. 
iii.  2864). 

2  Procurator  of  Belgica  and  of  both  Germanics,  Dessau,  Inscr.  Lat.  selectae, 
No.  1340  (?  Hadrian),  1362-1  (c.  a.d.  160). 

3  Tac.  Ann.  xi.   18,  19.  4  Ibid.  xiii.  56. 

5  Dio,  lx.  8.  6  Tac.  Ann.  xii.  27,  28. 

'   Cf.  Pliny,  N.H.  xxxi.  2  ;  Tac.  Germ.  29. 


252       REBELLION  ON  THE  RHINE     ch.  hi 

castellum,  and  is  therefore  still  called  Castel  to-day. 
It  was  not  until  the  annexation  of  the  Neckar  valley 
by  Vespasian  in  a.d.  74-75  that  the  Romans  began 
to  push  their  military  frontier  forward  beyond  the 
Rhine  in  the  upper  province,  and  the  chapter  opens 
concerning  the  great  defensive  works,  walls,  forts, 
ramparts,  and  palisades  which  were  made  to  link 
the  Rhine  and  Danube  together.1 

Claudius'  invasion  of  Britain  had  caused  a  con- 
siderable displacement  of  troops  on  the  Rhine 
frontier,  since  in  the  first  century  a.d.  the  "  German 
Army  "  was  the  most  conveniently  placed  of  all  the 
legionary  forces  in  Europe  for  the  purpose  of  invading 
the  island  and  strengthening  the  army  of  occupation 
when  it  had  landed  and  gained  its  first  victories. 
Hence  in  a.d.  68  only  seven  legions  lay  on  the 
Rhine.2  And  the  civil  wars,  above  all  the  Vitellian 
invasion  of  Italy  in  the  spring  of  the  next  year,  had 
very  greatly  drained  away  the  strength  of  the  Rhine 
army  to  the  south.  By  the  summer  of  a.d.  69  nearly 
one  hundred  thousand  men  of  all  arms  must  have 
been  withdrawn  by  Vitellius  from  the  German 
provinces.3  To  a  certain  small  extent  new  levies 
had  partially  replenished  the  legions,  or  the  portions 
of  the  legions,  which  had  been  left  behind.  But  the 
new  recruits  hastily  enlisted  on  the  spot  were  drawn 
largely,  if  not  entirely,  from  the  local  auxilia,  and 
were  Roman   in   nothing   but  in  name  and  in  the 

1  By  far  the  best  succinct  account  of  the  Roman  "  Limes  :'  here,  and  of  its 
history  from  Vespasian  to  Marcus  Aurelius,  has  been  given  us  recently  in  the 
last  published  work  of  our  Master  in  Roman  History,  Professor  Pelham,  in 
his  paper,  "A  Chapter  in  Roman  Frontier  History "  {Transactions  of  the 
Royal  Historical  Society,  N.S.  vol.  xx.   1 906). 

2  See  above,  p.  26,  for  these.  3  Cf.  Tac.  i.  61. 


sec.  .1.    REBELLION  ON  THE  RHINE       253 

citizenship  which  their  enrolment  in  the  legions 
gave  to  them.1  There  was  therefore  a  great  de- 
terioration of  the  troops,  not  only  in  numbers,  but 
also  in  quality,  discipline,  fidelity,  as  the  event  was 
but  too  quickly  to  prove. 

In  the  autumn  of  a.d.  69  the  Batavian  chief 
Civilis  declared  for  Vespasian.  At  that  moment 
the  troops  on  the  Rhine  seem  to  have  been  distri- 
buted as  follows  : — 

(1)  The  Legionaries.  —  The  legionaries  nearest 
the  sea  were  those  in  camp  at  Castra  Vetera," 
nearly  one  hundred  and  fifteen  miles  from  the 
coast,  and  twenty-five  from  the  parting  of  the 
channels  at  the  east  end  of  the  "  Island  of  the 
Batavians.a  This  place,  situated  at  Birten  by 
Xanten,  on  the  left  bank  of  the  Rhine  just 
below  Wesel,  was  then  the  chief  fortress  of 
Lower  Germany,  and  was  garrisoned  always,  until 
the  days  of  Domitian,4  by  two  legions  encamped 
together.  The  two  legions  in  a.d.  69  were  the 
Fifth  Alaudae  and  the  Fifteenth  Primigenia,  the 
latter  under  Munius  Lupercus  as  legate.  But  the 
main  bulk  of  the  Fifth  with  its  eagle  had  marched 
for  Italy,  and  only  a  detachment  (vexillum)  was 
left  in  camp.  The  Fifteenth,  on  the  other  hand, 
had  sent  only  a  detachment  to  the  south,  and  its 
eagle  stayed  behind  with  the  greater  part  of  the 
regiment.  But,  taken  together,  the  men  of  the 
two    legions    at    Vetera    mustered    scarcely    five 

1  Cf.  Tac.  ii.  15. 

-  This  I  call   "Vetera"  simply  henceforward,  following  Tacitus,  iv.   35, 
57,  etc. 

3  See  above,  p.  232.  4  Cf.  Suetonius,  Domitian,  7. 


254      REBELLION  ON  THE  RHINE     ch.hi 

thousand  men,  and  did  not  amount  to  the  strength 
of  a  single  legion  when  this  was  fully  up  to 
strength.1 

Thirty-five  miles  up  stream  from  Vetera  lay 
Novaesium,  the  modern  town  of  Neuss  on  the  left 
bank,  twenty-two  miles  down-stream  from  Cologne, 
and  nearly  opposite  Dusseldorf.2  This  was  the 
camp  of  the  Sixteenth  legion,  the  most  "  stay-at- 
home  "  of  all  the  legions  on  the  Rhine,3  and  its 
eagle  had  never  left  Germany,  though  up  to  a.d.  40 
it  had  been  quartered  at  Mainz.  Its  legate  was 
probably  Numisius  Rufus,  but  he  was  at  this  time, 
for  some  unexplained  reason,  not  with  his  legion, 
but  at  Vetera.4  This  legion  had  sent  a  vexillum  to 
the  war  in  Italy.  Its  strength  in  camp  therefore 
at  Novaesium  cannot  be  put  above  four  thousand 
men  at  most.5 

No  legionary  troops  were  at  this  time  stationed 
at  Cologne.  But  twenty  miles  up-stream  from 
Cologne,  at  Bonn,  there  lay  the  First  legion  under 
its  legate  Herennius  Gallus.  Most  of  the  men  of 
this,  the  "  premier  regiment "  in  the  Roman  army, 
had  been  summoned  to  Italy.  In  spite  of  new 
levies  it  could  place  only  some  three  thousand 
men  in  the  field.6 

In  Upper  Germany,  the  only  legionary  troops  at 

1  Tac.  iv.  22. 

2  Neuss  is  now  a  mile  and  a  half  from  the  Rhine.  The  actual  site  of  the 
camp  is  said  to  have  been  at  Grimlinghausen,  a  mile  away  from  the  modern 
Neuss  up-stream.  Neuss  itself  was  on  the  river  as  late  as  A.D.  1310.  See 
Heraeus'  note  on  Tac.  iv.  26. 

3  "Sesshaft" — Pfitzner,  Gesch.  d.  Kaiser legionen. 

4  By  a  comparison  of  Tac.  iv.  22  and  59. 

5  Tac.  iv.  26  ;  ii.  100. 

6  Tac.  ii.  57  ;  iv.  19,  20. 


sec.  in    REBELLION  ON  THE  RHINE       255 

this  time  on  the  Rhine  were  those  in  camp  at 
Moguntiacum,  the  modern  Mainz,  then  as  always 
the  chief  Roman  fortress  on  the  upper  course  of  the 
river,  and  about  a  hundred  miles  south-east  of 
Bonn.  This  also  was  a  "double  camp,"  and  con- 
tained the  two  legions,  Fourth  Macedonica  and 
Twenty-second  Primigenia.  The  latter  was  under 
the  command  of  the  legate  Dillius  Vocula,  who  was 
the  only  general  of  the  smallest  merit  upon  the 
whole  course  of  the  river  when  the  revolt  broke 
out.  The  Fourth  had  sent  a  detachment  to  Italy, 
and  its  eagle  stayed  behind.  It  therefore  may 
have  numbered  some  four  thousand  men  at  most. 
Its  regimental  reputation  was  a  poor  one,  and  it 
was  not  to  increase  this  in  the  coming  war.  The 
bulk  of  the  Twenty-second  legion  had  marched 
south  with  the  Vitellians.  The  whole  legionary 
garrison  of  Mainz  can  scarcely  have  numbered  more 
than  six  thousand  men.1 

Finally,  at  Vindonissa,  the  modern  Windisch, 
near  Basle,  one  hundred  and  seventy-five  miles 
south  of  Mainz,  there  should  have  been  encamped 
the  Twenty-first  legion  Rapax,  a  regiment  notorious 
for  savage  courage  marred  by  a  tendency  to  insub- 
ordination. But  the  entire  legion  had  by  this  time 
vanished  with  Caecina  over  the  Alps,  there  to  do 
desperate  deeds.2 

The  entire  legionary  Army  of  the  Rhine,  there- 
fore, at  this  time  numbered  barely  some  18,000  men, 
distributed  as  follows  : — 

1  Tac.  ii.  100  ;  iii.  22  ;  iv.  37. 

2  Tac.  i.  61,  67  ;  ii.  43,  etc. 


256      REBELLION  ON  THE  RHINE     ch.  m 

At  Vetera  :   Leg.  V.  and  XV.  =  5000  men 

„  Novaesium  :   Leg.  XVI.  =  ?4ooo     „ 

,,  Bonn :  Leg.  I.  =     3000     „ 

,,  Mainz :  Leg.  IV.  and  XXII.  =  ?6ooo     „ 


Total  18,000  men 

(2)  The  Auxiliaries. — It  is  quite  impossible  to 
form  any  estimate  at  all  of  the  total  number  of 
auxiliary  troops  at  this  time  forming  part  of  the 
garrison  of  the  Rhine.  A  little  is  known  of  their 
nationality,  their  position,  and  their  value  ;  nothing 
at  all  of  their  numbers. 

In  nationality  the  auxiliary  cohorts  found  upon 
the  Rhine  before  the  year  a.d.  70  were  mostly 
natives  of  Germany,  Gaul,  or  the  Upper  Danube. 
Thus  the  cohorts  of  Batavians,  Cannenefates, 
Tungri,  and  Ubii  were  Germans ;  those  of  Belgae, 
Nervii,  and  Nemetes x  were  Gauls ;  those  of  Raeti 
and  Vindelici  were  natives  of  the  districts  on  the 
Upper  Danube.  Cohorts  of  other  races  also  served 
on  this  frontier,  such  as  those  of  the  Asturians  and 
Vascones  from  Northern  Spain,  the  Breuci  from 
Pannonia,  and  Silaunenses  perhaps  from  the  east. 
But  it  seems  certain  that  the  majority  of  the 
auxiliary  cohorts  was  composed  of  native  troops 
levied  near  the  frontier  itself.  Cavalry  squadrons 
were  in  like  manner  furnished  by  the  Cannenefates, 
Batavians,  and  Treveri.2 

The  places  of  encampment  for  some  of  these  are 

1  The  Nemetes  dwelt  upon  the  eastern  slopes  of  the  Vosges  on  the  left 
bank  of  the  Rhine  by  Speier,  the  Roman  Noviomagus,  and  south  of  Speier 
towards  Selz. 

3  Cf.  Alex.  Riese,  ap.  Dizionario  Epigrafico,  sub  voc.  "Germania." 


sec. m    REBELLION  ON  THE  RHINE       257 

known.  Nearest  the  sea  of  all  troops  of  the  Roman 
army  upon  this  frontier  were  two  cohorts  of  Gauls 
who  occupied  some  petty  Roman  forts  in  the 
Island  of  the  Batavians,  and  probably  other  corps 
besides  of  Nervii  and  Tungri  very  recently  enlisted.1 
In  garrison  with  the  legionaries  at  Vetera  there 
were  Ubii,  cavalry  of  the  Treveri,  and  one  squadron 
of  Batavian  horse  under  Claudius  Labeo.2  Labeo 
was  himself  kinsman  to  the  insurgent  leader  Julius 
Civilis,  but  hated  him  so  intensely  that  his  loyalty 
to  the  Romans  was  unshaken  alike  by  imprisonment 
and  constant  pursuit  which  he  endured  at  the  hands 
of  the  rebels.  He  was  throughout  a  constant  thorn 
in  their  side.3  Higher  up  the  river  there  was  one 
squadron  of  horse  at  Asciburgium  (Asberg  by 
Mors4);  at  Novaesium  there  were  some  auxil- 
iaries ; 5  at  Bonn  cohorts  of  the  Belgae,  with  the 
Italian  ala  Picentina ; 6  and  at  Mainz  were  the 
cohorts  of  the  Batavians  and  Cannenefates.7 

The  best  troops  among  these  auxiliary  forces 
were  certainly  the  eight  cohorts  of  Batavians  who 
had  been  attached  to  the  Fourteenth  legion.  They 
had  seen  much  service  in  Britain,  and  more  recently 
during  the  Vitellian  invasion  of  Italy.  But  their 
temper  gave  great  ground  for  distrust.  They  had 
quarrelled  fiercely  with  their  comrades  of  the 
legion  during  Valens'  march  for  Italy,  and  broken 
out  into  open  mutiny,  which  it  had  been  difficult  to 

1  Tac.  iv.  15,  reading  "  Tungrorum  "  for  "  Germanorum." 
-  Tac.  iv.  18.  3  Tac.  iv.  i8,  56,  66,  70. 

4  On  the  road  between  Vetera    and  Novaesium,   twenty  miles  from  the 
former  (Tac.  iv.  33). 

6  Tac.  iv.  62.  G  Tac.  iv.  20,  62.  7  Tac.  iv.  19. 

S 


258      REBELLION  ON  THE  RHINE     ch.hi 

quell.1  They  had  repeated  the  offence  later  in  the 
year  while  quartered  at  Turin  with  the  legion,  and 
the  Emperor  had  been  compelled  to  separate  them 
once  and  for  all  from  the  legionaries.2  They  had 
therefore  been  sent  back  to  the  Rhine,  and  were, 
at  the  moment  when  Civilis  raised  the  standard  of 
revolt,  in  the  lines  at  Mainz,  together  with  their 
kinsmen  the  Cannenefates.  Before  news  of  Civilis' 
rising  reached  them,  however,  Vitellius,  alarmed 
at  last  at  the  threatened  invasion  of  Italy  by  the 
Flavian  troops,  sent  recalling  the  cohorts  of  both 
tribes  to  Rome.  They  had  started  on  the  march 
when  a  messenger  reached  them  from  Civilis  im- 
ploring their  help.  They  hesitated  not  a  moment, 
but  abruptly  turned  and  marched  for  the  north.3 

The  hatred  of  Rome  thus  displayed  by  these 
cohorts  was  commonly  shared  by  all  the  German 
and  Gallic  auxiliaries  a*t  this  time  on  the  Rhine. 
The  sudden  flood  of  the  mutiny  swept  the  whole 
native  army  away,  so  that  no  single  regiment  could 
be  trusted.  Old  and  new  grievances,  the  remem- 
brance of  ancient  liberties,  suspicion  and  long- 
nurtured  ill-will  felt  towards  the  alien,  jealousy  of 
the  regular  army,  dislike  of  the  officers,  all  com- 
bined to  excite  the  native  troops  against  the 
Romans  on  the  Rhine  as  they  roused  the  sepoys 
against  us  in  India  fifty  years  ago.  Their  own 
native  officers  for  the  most  part  stimulated  or 
acquiesced  in  their  mutiny.  Julius  Classicus,  the 
insurgent  leader  himself,  was  prefect  of  a  squadron 

1  Tac.  i.  59  ;  ii.  27-29.  -  Tac.  ii.  66,  69. 

3  Tac.  iv.  15,  19. 


sec.  ni    REBELLION  ON  THE  RHINE       259 

of  horse  of  the  Treveri.1  When  an  officer  like 
Claudius  Labeo  was  found  to  lead  his  men  against 
the  rebel  army,  they  deserted  him  and  went  over  to 
their  brethren,  and  he  was  happy  indeed  to  escape 
with  his  life.2  It  was,  in  fact,  now  for  the  first  time 
that  the  Roman  army  system  in  its  method  of 
recruiting  native  auxiliaries  was  seriously  tested, 
and  it  broke  down  hopelessly.  Recruiting,  of 
course,  was  far  more  easy  when  the  natives  were 
enlisted  to  serve  in  clan-corps  in  or  near  to  their 
own  homeland.  But  that  such  a  saving  of  trouble 
was  very  dearly  purchased  was  a  lesson  first  taught 
the  Roman  military  authorities  by  the  mutiny  of 
the  native  army  on  the  Rhine.  That  Vespasian 
had  duly  learnt  this  lesson  will  be  shown  at  the  end 
of  this  chapter.3 

In  the  autumn  of  the  year  a.d.  69  the  Roman 
Army  of  the  Rhine  was  therefore  weak  in  numbers, 
and  weaker  still  in  discipline  and  loyalty.  It  was 
strewn  in  widely  separated  fragments  along  three 
hundred  miles  of  frontier  from  Mainz  to  the 
German  Ocean.  Its  commander-in-chief,  the 
governor  Hordeonius  Flaccus,  had  but  recently 
been  appointed.  He  was  old,  slothful,  timorous,  a 
martyr  to  gout,  secretly  treacherous  to  the  Emperor 
Vitellius,  distrusted  and  despised  by  the  troops.4 
The  army  had  lost  its  ablest  leaders  and  many  of 
its  best  troops.  It  was  soaked  through  and  through 
with    disaffection,    mistrust,    inefficiency.       It    was 

1  Tac.  iv.  55.  2  xac.  iv.  18,  66. 

3  See  below,  pp.  329-331.  4  Tac.  i.  9,  56,  88;  iv.  13,  24. 


260      REBELLION  ON  THE  RHINE     ch.ih 

ever  more  and  more  bewildered  and  distracted  by 
the  doubtful  issue  of  the  struggle  for  Empire  raging 
furiously  in  Italy,  and,  when  at  last  Vespasian's 
cause  won  the  day,  it  was  hostile  to  that  new  and 
unknown  Emperor.  It  saw  the  strength  of  the 
Roman  Army  and  Roman  State  rent,  as  it  seemed, 
in  pieces  before  its  eyes.  It  was  upon  this  army, 
in  a  state  so  sorrowful  and  hazardous,  that  there 
suddenly  burst  the  tempest  of  a  great  national 
insurrection,  with  the  objects  of  which  a  great  part 
of  that  army  sympathised  heartily.  That  disaster 
followed  disaster,  that  this  Army  of  the  Rhine 
ceased  to  exist,  that  the  flame  of  revolt  ran  along 
the  entire  length  of  the  frontier  like  fire  along  a 
gunpowder  train, — these  were  but  natural  results. 
Never  had  Rome  known  or  endured  the  like.  Her 
army  had  played  her  false.  Yet  in  such  an  army 
how  could  she  have  confidence  ? 

In  the  course  of  a  few  months  soldiers  successively  of  Nero, 
of  the  Senate,  of  Galba,  of  Vitellius,  and  of  Vespasian  ;  the  only 
support  to  the  dominion  of  Italy  over  the  two  mighty  nations  of 
the  Gauls  and  the  Germans,  while  the  soldiers  of  the  auxiliaries 
were  taken  almost  entirely,  and  those  of  the  legions  in  great  part, 
from  those  very  nations  ;  deprived  of  their  best  men ;  mostly 
without  pay  ;  often  starving  ;  and  beyond  all  measure  wretchedly 
led — they  were  certainly  expected  to  perform  feats  inwardly  and 
outwardly  superhuman.     They  ill  sustained  the  severe  trial.1 

But  though  her  army  on  the  Rhine  disappeared, 
though  her  forts  and  frontier  defences  were  shat- 
tered, though  mutiny  seemed  victorious  and 
treachery  triumphant,  Rome  was  still  invincible. 

1  Mommsen,  Provinces,  Eng.  Trans,  vol.  i.  p.  146. 


Coin  of  Galba. 


Coins  of  Otho. 


Coin  of  Vitellius. 


Coin  of  Vespasian. 


sec.  iv    REBELLION  ON  THE  RHINE       261 

§  4.    The  War,  up  to  the  Relief  of  Vetera 

A.  The  Clearing  of  the  "Island" — Julius  Civilis, 
the  Batavian  chieftain,  had  been  encouraged  to  raise 
a  revolt  on  the  Lower  Rhine  in  the  interests  of 
Vespasian,  that  he  might  keep  Vitellius'  troops 
on  that  river  busily  employed.  In  his  heart  he 
cherished  the  deeper  design  of  striking  for  the 
liberty  of  his  native  land.  The  opportunity  for 
action  was  quickly  given  him.  The  Emperor 
Vitellius,  on  news  of  the  Flavian  invasion  of  Italy, 
sent  commands  for  a  general  conscription  through 
the  Batavian  lands.1  The  officers  whom  the  governor 
appointed  to  carry  out  the  orders  acted  harshly  at 
least,  and  perhaps  the  graver  charges  of  injustice 
and  lust  brought  against  them  were  not  entirely 
lacking  in  truth.  Civilis  called  his  tribesmen  into 
the  secret  recesses  of  one  of  their  sacred  groves 
and  made  to  them  a  fiery  speech.2  They  at  once 
decided  to  rebel,  refused  to  obey  the  Roman  demand 
for  men,  and  sent  messengers  to  the  Cannenefates 
their  neighbours,  and  to  the  cohorts  of  their  people 
stationed  at  Mainz,  urging  them  to  join.3 

The  first  blow  was  struck  by  the  Cannenefates, 
who  joyfully  lent  their  aid.  There  was  one  of 
their  tribe,  Brinno  by  name,  whose  father  had 
defied  and  fought  the  Romans  in  the  principate  of 
Caligula,  and  the  son  loved  them  no  better.  Placed 
upon  one  of  the  large  German  shields,  and  raised 

1  Tac.  ii.  97  ;  iv.  13. 

•  Tacitus  was  not  in  the  grove.  Whether  some  one  there  later  told  him 
the  speech,  or  he  invented  it,  cannot  be  shown ;  the  latter  is  the  more 
probable.  3  Tac.  iv.  14. 


262      REBELLION  ON  THE  RHINE     ch.ih 

on  high  above  the  crowd  by  his  eager  followers,  he 
was  elected  leader  after  the  fashion  of  his  tribe,  and 
forthwith  called  them  to  arms.  He  sent  also  to  the 
Frisii,  calling  them  to  the  field,  and  with  his  tribes- 
men fell  suddenly  upon  two  Gallic  cohorts  of 
auxiliaries,  quartered  in  the  "  Island."  These  fled, 
and,  after  burning  some  small  Roman  forts  in  the 
neighbourhood,  fell  back  to  the  upper  end  of  the 
island  where  with  the  rest  of  the  Roman  auxiliaries 
in  the  district  they  stood  at  bay. 

Then  Civilis  himself  took  the  field,  at  the  head 
of  a  mixed  company  of  Batavians,  Cannenefates, 
and  Frisii.  Near  Nymwegen  or  Arnhem,1  he  and 
his  army  attacked  the  auxiliaries.  These  now  had 
the  Roman  Rhine  flotilla  on  their  flank  to  help 
them.  But  treachery  quickly  got  the  better  of  both 
army  and  fleet.  A  Tungrian  cohort  in  the  former, 
and  the  Batavian  oarsmen  employed  upon  the  latter, 
played  them  false,  and  the  Romans  were  speedily 
defeated  and  expelled  one  and  all  from  the  "  Island," 
while  the  whole  fleet  of  twenty-four  vessels  was 
either  captured  or  destroyed. 

It  was  time  for  the  Roman  legionary  to  come  to 
the  rescue.  Hordeonius  Flaccus  sent  ordering  the 
garrison  of  Vetera  to  advance  at  once  against  the 
rebels,  under  command  of  Munius  Lupercus,  legate 
of  the  Fifteenth  legion.  He  marched  the  twenty- 
five  miles  down-stream  to  the  "  Island,"  and  threw 
his  force  into  it  over  the  Waal.  His  army  was  com- 
posed of  the  legionaries  of  the  Fifth  and  Fifteenth 
legions,  cavalry  of  the  Treveri  and   Batavians,  and 

1  Tacitus'  geographical  knowledge  is  vague  and  incredibly  unsatisfactory. 


sec.  iv     REBELLION  ON  THE  RHINE       263 

also  auxiliaries  of  the  Ubii.  He  found  Civilis  quite 
ready  to  fight  him,  probably  again  in  the  neighbour- 
hood of  Arnhem.  In  the  battle  the  native  regiments 
with  one  consent  betrayed  him.  The  Batavians 
promptly  went  over  to  the  foe :  the  Ubii  and 
Treveri  ran.  His  legionaries  indeed  stood  firm, 
and,  when  the  battle  was  lost,  drew  off  in  good  order 
to  Vetera  again.  But  the  outlook  was  ominous 
enough.  A  mere  handful  of  legionaries  could  do 
little  to  withstand  or  suppress  a  national  rising. 
And  soon  the  rebels  received  reinforcements  more 
valuable  than  any  which  Civilis  had  as  yet  gathered 
to  his  cause.1 

This  befell  when  the  cohorts  of  Batavians  and 
Cannenefates  from  Mainz  marched  into  Civilis' 
camp,  proud  and  exultant  after  forcing  their  way 
through  the  Roman  legionary  line  itself.  The 
men  of  the  First  legion  had  sallied  out  from  their 
camp  at  Bonn  to  bar  the  way  north  to  the  rebel 
cohorts.  They  had  trusted  that  the  governor 
Flaccus  was  hard  upon  the  rebels'  heels  with  the 
garrison  of  Mainz,  and  they  had  hoped  therefore 
to  catch  the  cohorts  between  two  fires.  Their 
own  legate,  Herennius  Gallus,  had  been  reluctant 
to  run  the  risk.  His  hesitation  showed  prudence 
if  scanty  pluck.  For  no  Flaccus  ever  appeared 
pursuing  after  the  enemy.  The  cohorts,  experi- 
enced and  bold  veterans,  feared  not  a  whit  the 
legionaries  and  their  auxiliary  allies  whom  they  saw 
thrown  across  their  road  outside  the  gates  of  Bonn. 
They  formed  square  and  thrust  hard  at  the  Romans, 

1  Tac.  iv.  16-18. 


264       REBELLION  ON  THE  RHINE     ch.hi 

whose  auxiliaries  fled  as  usual,  and  the  angry 
legionaries  found  their  line  pierced,  themselves 
flung  aside,  and  the  cohorts  disappearing  gaily  down 
the  road  towards  Cologne.  Pursuit  was  out  of  the 
question,  and  the  legion  was  left  to  digest  its  dis- 
grace as  best  it  could,  while  the  cohorts  marched 
calmly  north,  leaving  Cologne  untouched,  past  the 
ramparts  of  Vetera,  and  reached  Civilis  with  their 
new  honours  thick  upon  them.  The  bitter  re- 
crimination which  now  occupied  the  leisure  of  the 
legionaries  at  Bonn  did  not  greatly  avail  to  sweeten 
the  cup  of  a  very  notable  defeat.1 

Civilis  was  justly  encouraged.  He  was  now 
leader  of  a  very  respectable  force,2  and  could  carry 
the  war  outside  the  limits  of  the  "  Island "  on 
his  own  account.  Having  made  his  entire  army 
take  the  oath  of  fidelity  to  Vespasian,  he  sent 
a  message  to  the  garrison  at  Vetera,  twenty- 
five  miles  away,  bidding  this  do  the  like.  It 
defiantly  refused.  Thereupon  with  his  entire  avail- 
able force  marching  by  both  banks  of  the  Rhine 
and  with  his  fleet  moving  up-stream  in  company, 
Civilis  led  the  rebel  force  to  the  siege  of  Vetera. 
This  stronghold  of  the  Romans  on  the  Lower  Rhine 
must  be  taken. 

B.  The  Siege  of  Vetera. — "  The  tide  of  warfare," 
it  has  been  said,  "  ebbs  and  flows  on  an  ocean  which 
is  studded  with  strategical  objectives."  But  this  tide 
of  the  German  mutiny  rolled  fast  in  a  narrow  bed, 
pushing  its  surging  flood  steadily  up  the  valley  of 
the  Rhine,  and  presently  casting  a  secondary  wave 

1  Tac.  iv.  18-20,  25.  2  "Justi  exercitus  ductor,"  Tac.  iv.  21. 


sec.  iv    REBELLION  ON  THE  RHINE       265 

along  the  course  of  the  lateral  valley  of  its  tributary 
the  Moselle,  while  the  main  flood  went  steadily 
sweeping  up  the  greater  river  to  the  south.  Then 
in  due  course  it  receded  heavily  back  down  both 
the  streams,  and  its  last  murmurs  died  away  in  the 
peaceful  slow-moving  waters  of  the  Northern  Sea. 
In  its  rising  it  lapped  greedily  round  the  walls  of 
many  a  Roman  fortress ;  in  its  ebb  it  left  behind 
it  white  staring  ruins,  and  rotting  heaps  of  slain. 
Nowhere  did  it  beat  more  furiously  in  its  onset  than 
upon  the  ramparts  behind  which  the  scanty  little 
garrison  of  Vetera  stood  staunchly  on  the  defensive. 
Hurled  back  ever  and  again,  it  still  foamed  up  upon 
the  defiant  barricades,  or  worked  stealthily  to  under- 
mine the  foundations  of  them.  So  in  the  end  the 
cruel  flood  worked  its  will,  and  Vetera  fell.  But 
before  this  came  to  pass  many  weeks  went  by,  of 
siege  and  relief,  of  hope  and  despair,  of  famine  and 
surprise.  The  siege  of  Vetera  is  the  one  heroic 
episode  in  the  first  chapter  of  this  war. 

As  was  the  case  with  the  Residency  of  Lucknow 
and  its  small  company  of  defenders,  so  Vetera  and 
its  garrison  were  but  ill-prepared  to  stand  a  siege. 
Its  troops,  like  our  own,  had  sallied  out  to  find  the 
foe,  and  been  beaten  back  to  find  shelter  in  its  walls. 
They,  like  our  own,  felt  the  pinch  of  hunger  as  the 
days  went  by,  and  the  assailants  pressed  them  close, 
now  by  fierce  onslaught,  now  by  sullen  blockade. 
They,  like  our  own,  had  scarcely  men  enough  to 
defend  the  walls,  "  only  5000  men  to  defend  a 
camp  built  to  hold  two  legions."  They,  like  our 
own,  were  hampered  and  embarrassed  by  a  crowd 


266      REBELLION  ON  THE  RHINE     ch.ih 

of  civilians,  traders,  women,  children,  who  took 
refuge  behind  the  walls  defended  by  the  troops. 
For  in  the  long  years  of  peace  upon  the  Lower 
Rhine  many  buildings  had  grown  up  round  about 
outside  the  military  camp  "  in  manner  of  a  town- 
ship,"1 where  the  Roman  merchants  and  traders 
had  their  dwellings  and  stored  their  goods,  and 
where  the  women  and  the  children  of  these  and  of 
the  troops  had  their  homes.  All  these  buildings 
had  for  purposes  of  the  defence  to  be  destroyed, 
and  the  non-combatants  given  shelter  behind  the 
fortifications.  This  made  the  demand  upon  the 
stock  of  provisions  all  the  heavier,  and  these,  largely 
through  the  garrison's  own  fault,  were  already  too 
scanty.  The  very  defences  themselves  had  been 
built  in  the  days  of  Augustus  rather  to  serve  as  a 
good  base  of  operations  directed  against  the  German 
tribes  over  the  river,  and  as  a  post  from  which  to 
observe  their  movements,  than  to  protect  Roman 
legionaries  against  attack.  The  greatest  and  most 
prudent  of  all  the  emperors  had  not  foreseen  so 
desperate  a  reversal  of  fortune.  Hence  the  very 
fortifications  were  inadequate.  If  the  Roman  did 
not  entirely  accept  the  modern  maxim  that  the 
history  of  entrenched  camps  is  almost  always  the 
history  of  capitulations,  yet  at  least  he  too  relied  on 
arms  in  the  field  rather  than  on  stone  walls  for 
victory  and  for  safety.2 

Yet   for   all    this    there   was   never   a   moment's 


1   "  In  niodum  municipii  "  :  an  excellent  example  of  the  growth  of  towns 
as  due  to  the  system  of  permanent  cantonments. 
-  Tac.  iv.  21-23. 


sec.  iv     REBELLION  ON  THE  RHINE       267 

thought  of  capitulation,  of  even  parleying  with  the 
rebels.  "  They  would  not,"  they  sent  answer  to 
Civilis,  "  listen  to  a  traitor's  advice  nor  to  that 
of  enemies.  Vitellius  was  their  Emperor :  they 
would  keep  their  faith  to  him  and  their  weapons 
until  their  last  breath.  Let  not  a  runaway  Batavian 
think  to  control  the  destinies  of  Romans.  Let 
him  rather  expect  the  due  penalty  of  his  desertion." 
In  the  old  spirit  of  the  Romans  in  Gaul  a  century 
earlier,  they  provoked  the  enemy  to  wrath  and  bade 
him  do  his  worst.  Walls  and  entrenchments  were 
hastily  strengthened,  and  the  garrison  waited  for 
the  coming  of  the  rebels. 

They  had  not  long  to  wait.  Furious  with 
wrath  at  the  answer  given  to  his  challenge,  Civilis 
passionately  hurled  his  motley  army  upon  the  fort. 
Here  veteran  troops  advanced  under  the  worn 
colours  of  the  Roman  army,  and  plied  all  Roman 
arts  of  siege-craft.  There  wild  barbarians  rushed  to 
swarm  over  the  defences,  brandishing  on  high  the 
rudely  carved  images  of  wild  beasts,  which  signified 
each  its  special  tribe  and  nation,  and  had  been 
brought  from  the  gloomy  recesses  of  sacred  forests 
to  urge  their  savages  forward  to  the  work  of  plunder 
and  of  butchery.  With  rocks  and  stones  hurled  by 
the  catapults  mounted  on  the  walls,  and  with  fiery 
spears,  the  garrison,  grimly  fighting,  drove  back 
all  assaults,  and  when  night  fell  the  fort  was  still 
inviolate.     Civilis'  first  attempt  had  failed. 

The  insurgent  leader  knew  that  but  a  few  days' 
provisions  were  all  that  remained  to  the  besieged. 
He  ceased  from  further  direct  attack,  and  his  army 


268      REBELLION  ON  THE  RHINE     ca.ni 

was  spread  round  the  walls,  waiting  for  hunger  or 
for  treachery  to  do  the  work  where  force  had 
failed,  like  some  crafty  beast  of  prey  couching 
long  before  the  final  spring.  Now,  if  ever,  the 
Romans  higher  up  the  river  must  march  to  relieve 
the  garrison  and  raise  the  blockade. 

C.  The  Advance  of  the  Relieving  Army. — Even 
the  old  infirm  governor,  Hordeonius  Flaccus  him- 
self, saw  that  Vetera  must  if  possible  be  saved. 
Not  only  did  he  issue  orders  at  Mainz  for  the 
instant  sending  of  a  relieving  army  from  that  camp, 
but  he  himself  accompanied  it.  Fearful,  however, 
of  the  toils  of  a  march,  he  journeyed  by  water  down- 
stream, which  failed  to  increase  the  small  respect  in 
which  his  troops  already  held  him.  The  officer 
appointed  to  lead  the  force  by  land  was  Dillius 
Vocula,  legate  of  the  Twenty-second  legion.  His 
army  consisted  of  picked  soldiers  from  the  two 
legions  encamped  at  Mainz,  the  Fourth  and  Twenty- 
second,  and  was  constantly  increasing  in  numbers 
during  the  march  to  the  north  by  the  addition  of 
Gallic  auxiliaries  who,  at  Flaccus'  orders,  came  to 
join  the  army.  The  distance  from  Mainz  to  Vetera 
was  one  hundred  and  seventy-five  miles.  For  the 
first  hundred  miles,  as  far  as  Bonn,  Vocula  met  with 
no  opposition,  and  advanced  by  forced  marches. 
At  Bonn  the  men  of  the  First  legion  were  added 
to  the  relieving  column.  And  at  Cologne,  twenty 
miles  on,  Hordeonius  Flaccus,  finding  himself  hope- 
lessly unpopular  with  the  troops,  finally  handed 
over  the  entire  control  of  all  the  operations  to  the 
legionary  legate.     Vocula  had  put  down  disaffection 


sec.  iv     REBELLION  ON  THE  RHINE      269 

and   lack   of  discipline   with  a  firm  hand,  and  the 
soldiers  admired  him  the  more  they  feared  him. 

But  from  Cologne  onwards  the  difficulties  of  the 
relieving  army  multiplied.  Vocula  was  hampered 
not  only  by  lack  of  supplies  and  an  inefficient  com- 
missariat staff,  but  also  by  an  exceptionally  low 
Rhine.  The  state  of  the  river  not  only  made 
all  navigation  very  difficult,  and  delayed  the  corn 
ships,  but  it  also  made  it  necessary  to  post  patrols 
at  intervals  all  along  the  left  bank  to  prevent  parties 
of  Germans  crossing  and  falling  upon  the  flank  or 
rear  of  the  Roman  column  as  it  advanced.  Pro- 
gress was,  therefore,  very  slow,  the  more  so  as  the 
men  were  dispirited,  and  troops  in  this  state  of 
mind  march  very  badly.  "  The  old  defences  of  the 
Empire,"  they  said,  "  were  deserting  them  :  the  gods 
were  angry."  At  Novaesium,  however,  twenty-two 
miles  below  Cologne,  fresh  reinforcements  were 
picked  up  in  the  men  of  the  Sixteenth  legion  ;  and 
the  legate  of  that  legion,  Herennius  Gallus,  now 
shared  with  Vocula  the  responsibilities  of  command. 
But  here  part  of  the  army  was  left  in  camp  with 
Flaccus,  the  governor.  The  foe  were  now  close  at 
hand,  and  a  fortified  base  camp  was  necessary. 
The  rest  of  the  column  then  marched  slowly  for- 
wards as  far  as  Gelduba,  where  a  small  fort  on 
rising  ground  overlooked  the  Rhine.1  This  is  now 
the  village  of  Gellep,  between  Kaiserwerth  and 
Urdingen.  Here  the  two  generals,  Vocula  and 
Gallus,  halted  their  men.  Vetera  was  only  twenty- 
five  miles  away,  but  the  country  was  swarming  with 

1  Cf.  Pliny,  Nat.  Hist.  xix.  5,  90. 


270      REBELLION  ON  THE  RHINE 

foes  in  front  and  on  both  flanks.  The  fort  was  hold- 
ing out  stubbornly,  and  the  generals  of  the  reliev- 
ing army  would  take  no  risks  with  their  own  force. 
The  position  at  Gelduba  was  strongly  entrenched, 
and  some  time  actually  was  spent  in  drilling  and 
exercising  the  army.  More  valuable  training,  how- 
ever, was  afforded  by  an  expedition  of  part  of  the 
force  under  Vocula  against  a  hostile  tribe,  which 
threatened  the  left  flank  of  any  farther  advance. 
These,  the  Cugerni,  were  Germans,  and  possibly  the 
remnants  of  the  ancient  Sugambri,  whom  Tiberius 
sixty  years  before  had  settled  on  the  left  bank  of 
the  Rhine.  They  now  dwelt  in  the  district  round 
the  modern  small  town  of  Goch,  thirteen  miles 
west  of  Xanten  (Vetera).  They  had  joined  Civilis, 
and  a  raid  upon  their  territory,  it  was  hoped,  would 
teach  them  to  keep  quiet  for  the  present.  During 
Vocula's  absence  a  stranded  corn  ship  led  to  a  fight 
between  the  Romans  left  in  camp  at  Gelduba  and 
the  marauding  Germans,  who  came  down  upon  the 
vessel  to  plunder  her.  In  this  the  Romans  got 
the  worse,  and  lost  the  ship.  The  troops,  as 
usual,  were  furious  with  their  general  Gallus,  and 
only  Vocula's  timely  return  and  stern  treatment  of 
the  ringleaders  stopped  another  incipient  mutiny. 
The  patrols  on  the  river  bank  had  proved  quite 
unable  to  prevent  German  roving  bands  from 
crossing  the  stream.  Vocula,  therefore,  found  the 
enemy  active  behind  him,  threatening  his  supply 
trains,  and  ravaging  the  lands  of  his  still  faithful 
allies,  the  Ubii  round  Cologne.  Even  Marcodurum, 
the  present  Diiren  on  the  Roer,  twenty-four  miles 


nc.iv    REBELLION  ON  THE  RHINE       271 

west-south-west  of  Cologne,  garrisoned  carelessly  by 
a  cohort  of  this  loyal  tribe,  who  trusted  for  safety 
rather  to  their  distance  from  the  Rhine  than  to 
vigilance,  was  surprised  and  sacked  by  the  enemy, 
contemptuous  alike  of  the  Roman  camps  at  Nova- 
esium  and  Gelduba.  That  the  relieving  army  was 
itself  in  a  state  of  semi-blockade  at  Gelduba  is 
shown  by  its  long-continued  inactivity.1 

Meanwhile  the  Germans  grew  weary  of  the 
blockade  of  Vetera.  The  garrison  might  be 
starving,  but  it  gave  no  signs  of  any  thought  of 
surrender.  The  approach  of  the  relieving  army 
also  seemed  to  call  for  greater  efforts  by  the 
besiegers,  and  while  Vocula  and  his  men  were 
miserably  wasting  time  at  Gelduba  instead  of 
pushing  right  through  to  the  beleagured  fortress, 
a  fierce  assault  by  the  enemy  again  tested  the 
endurance  and  valour  of  its  garrison  to  the  utmost. 
All  messages  to  them  from  the  relieving  column, 
all  news  of  its  despatch  or  approach,  had  been  care- 
fully intercepted  by  Civilis.  For  all  they  knew 
they  were  left  to  their  fate,  to  starve  or  to  perish 
fighting,  unless  they  would  betray  their  honour. 
And  now  a  still  greater  mass  of  German  savages 
from  over  the  Rhine  flung  themselves  upon  the 
entrenchments.  The  night  did  but  add  to  the 
perils  and  horrors  of  the  day.  Huge  fires  were 
seen  blazing  in  the  enemy's  lines,  just  outside  their 
ramparts,  and  the  figures  of  the  barbarians  were  lit 
up  by  the  flames  as  they  sat  drinking  and  carousing. 
Then,  hot  with  wine,  the  savages  swarmed  again  to 

1  Tac.  iv.  24-28. 


272       REBELLION  ON  THE  RHINE     ch.hi 

the  assault.  But  the  light  of  the  fires  made  their 
bodies  an  easy  mark  for  the  steady,  deliberate,  and 
unerring  aim  of  the  legionaries.  Many  of  their 
bravest  chiefs  had  been  picked  off,  while  the  walls 
of  the  camp,  dark  and  frowning,  defied  the  blind 
shooting  of  the  enemy,  before  Civilis  noticed  the 
error  of  his  men.  Then  the  fires  were  stamped 
out,  and  the  blackness  of  night  covered  the  move- 
ments of  besiegers  and  besieged  alike.  But  the 
legionaries  fought  grimly  on  by  ear,  now  that  sight 
failed  them.  Heavy  stakes  and  stones  were  hurled 
down  into  the  darkness  where  the  noise  was  loudest. 
The  sound  of  scaling  ladders  planted  against  the 
walls  called  them  instantly  to  the  spot.  They 
thrust  the  stormers  back  with  their  shields,  and 
followed  their  flight  with  a  rain  of  javelins.  Yet 
many  of  the  foe  made  good  their  footing  on  the 
ramparts,  there  to  meet  death  from  the  short  stab- 
bing sword  of  the  legionary.  The  grey  dawn 
brought  new  methods,  but  no  relaxation  of  attack. 
The  Batavians  wheeled  a  great  two-storied  tower 
crammed  with  men  up  to  the  Praetorian  gate,  where 
the  level  ground  gave  them  easy  access.  It  was 
battered  to  pieces  by  the  Romans  with  poles  and 
beams,  and  a  sudden  sally  drove  the  enemy  off. 
If  one  more  rash  than  his  comrades  approached 
within  reach  of  the  walls,  he  was  suddenly 
gripped  by  the  iron  hand  of  a  crane,  whirled  up 
into  the  air  before  his  fellows'  very  eyes,  and  flung 
a  mangled  body  over  the  rampart  wall  into  the 
heart  of  the  camp.  Once  more  the  garrison  had 
repelled  this  most  savage  of  onslaughts.     The  dis- 


sec.  iv    REBELLION  ON  THE  RHINE       273 

comfited  Germans  drew  off,  and  sate  down  sullenly 
again  to  beleaguer  the  fort  which  no  courage  of 
theirs  could  take.1 

D.  The  Relief  of  Vetera. — At  this  point  there 
arrived  in  the  Roman  camp  at  Gelduba  the  news  of 
the  battle  of  Cremona.  It  made  no  difference  to 
the  military  operations.  The  tidings,  it  is  true, 
were  at  once  sent  to  Civilis.  He  was  informed 
that  the  Roman  army  on  the  Rhine  had  renounced 
its  allegiance  to  Vitellius  and  declared  for  Vespasian. 
His  object,  therefore,  had  been  won,  and  it  was 
time  for  him  to  cease  from  all  further  hostilities. 
Civilis  took  no  notice,  but  with  still  greater  energy- 
urged  on  the  war.  Thereby  he  at  last  threw  aside 
the  mask,  and  the  struggle,  which  up  to  this  point 
could  have  been  in  theory  viewed  merely  as  a 
chapter  in  the  history  of  the  Civil  War,  hence- 
forward took  on  its.  true  colours  of  a  national  rising 
of  the  Germans  against  Rome.  Politically  and 
historically  this  was  most  significant.  But  its 
only  influence  upon  the  course  of  the  war  was 
that  it  lent  all  the  greater  reason  for  hate,  if  not 
the  greater  hatred,  to  the  combatants  on  both 
sides.2 

The  relieving  army  was  not  left  undisturbed  in 
camp  at  Gelduba.  Civilis,  keeping  enough  men 
to  carry  on  the  blockade  of  Vetera,  sent  the  rest  to 
rush  the  Roman  camp.  Sacking  a  small  fort  on 
the  way  at  Asciburgium,  near  Mors,  twenty  miles 
south  of  Vetera,  the  Germans  completely  surprised 
the  Roman  army,  and,   forcing  their  way  into  the 

1  Tac.  iv.  28-30.  2  Tac.  iv.  31,  32. 

T 


274      REBELLION  ON  THE  RHINE     ch.™ 

lines,  began  a  ready  massacre  in  the  midst  of  the 
general  panic.  The  German  and  Gallic  auxiliaries 
of  Vocula  were  utterly  terrified,  and  their  cowardice 
all  but  led  to  a  complete  disaster.  From  this  the 
Roman  general  was  saved  only  by  the  timely  arrival 
of  some  cohorts  of  Spanish  auxiliaries,  the  Vascones. 
These,  who  had  recently  been  levied  by  Galba, 
none  the  less  understood  the  primary  duty  of  the 
soldier,  and,  hearing  the  noise  of  the  fighting  when 
they  were  still  some  distance  from  the  camp, 
marched  straight  to  the  sound.  They  came  un- 
expectedly upon  the  enemy  in  the  rear,  and  the 
consternation  which  they  caused  gave  time  to  the 
legionaries  to  rally  from  their  first  shock  of  sur- 
prise, and  drive  the  Germans  with  heavy  loss  from 
the  camp.  But  Vocula  or  his  sentries  had  done 
little  to  earn  the  esteem  of  the  army  by  this 
engagement. 

Neither  did  he  follow  up  at  once  the  success 
which,  somewhat  in  his  own  despite,  he  had  gained. 
Had  he  done  this,  in  the  opinion  of  the  Roman 
historian,  the  German  army  would  have  melted 
away,  and  Vetera  would  have  been  relieved  without 
further  trouble.  But  still  Vocula  tarried  at  Gel- 
duba,  and  thereby  gave  Civilis  the  chance  to  play 
one  last  card.  Direct  assault  had  twice  failed. 
Hunger  had  been  defied.  The  attack  upon  the 
relieving  army  had  been  repulsed.  Perhaps  strata- 
gem would  give  him  the  prize  which  seemed  so 
nearly  in  his  grasp.  The  garrison  had  heard 
nothing  of  the  approach  of  a  relieving  force,  until 
one  morning  they  saw  paraded  outside  the  ramparts 


sec.  iv    REBELLION  ON  THE  RHINE       275 

the  captive  standards  of  a  Roman  army  and  a  string 
of  Roman  prisoners.     They  were  colours  and  men 
taken    in    the   attack  on   Gelduba,   and   were    con- 
temptuously displayed  as  sole  survivors  of  the  one 
army  which  could  have  come  to  save  the  fort.      But 
Civilis'  trick  recoiled   on   his  own   head.      One   of 
the  prisoners — and  history  would  have  done  well  to 
record   his   name — shouted  to   the  besieged  a  few 
words,  enough  to  tell  them  the  truth,  that  the  Army 
of  the  Rhine  was  not  destroyed,  but  was  even  then 
hard    at    hand.     His    captors    struck   him   dead   to 
earth,  but  his  heroism  had  saved  the  garrison  from 
despair.      And   soon,   as  they   strained   their    eyes 
over  the  illimitable   expanse  of  plain   towards  the 
south,   they   saw   smoke   rising  upon    the   horizon. 
Vocula  and  his  army  had  struck  camp  at  last,  and 
the  burning  houses,  which  on   their  advance  they 
put  to  the  flames,  gave  the  signal  of  their  approach. 
The  whole  German  beleaguering  force  drew  out  to 
meet  the  coming  foe.    The  anxious  garrison  saw  their 
comrades  rapidly  advance,  and  halt.      Next  came  a 
moment's  pause,  when  it  seemed  as  if  they  would 
entrench  themselves  against  the  German  onslaught. 
But  then  the  whole  weary  army,  animated  as  by 
one  wild  longing  for  battle,  in  little  order  but  with 
fierce  shouts,  dashed  upon  the  waiting   Germans. 
Every  gate  of  Vetera  was  at  once  flung  open,  and 
the   besieged    garrison   sallied   out    to  join   in    the 
last  desperate  onset  on  the  enemy.     The  fighting 
was  bitter.       But  presently  down   went   the  rebel 
leader's    horse,   and  the    rumour    spread    fast    that 
Civilis  was  slain.     The  Germans  fled  in  wild  panic, 


276      REBELLION  ON  THE  RHINE     gk.hi 

and   the   Romans   at   last   marched   into   the   camp 
which  had  been  so  stoutly  defended.1 


§  5.  Flood  Tide :   The  Success  of  the  Mutiny 

A.  The  Retention  of  Vetera. — Vetera  was  success- 
fully relieved.  There  then  followed  on  Vocula's  part 
an  action  which  remains  the  one  military  puzzle  of 
this  war.  The  general  strengthened  the  defences 
of  the  camp,  sent  away  all  non-combatants  under 
escort  up-stream  to  Novaesium,  took  one  thousand 
of  the  best  men  from  the  old  garrison  and  added 
them  to  his  own  force,  and  with  this  small  addition 
to  his  strength  retired  again  up  the  river  to  Gel- 
duba.  In  actual  fact,  against  his  orders  more  than 
one  thousand  followed  the  return  march,  and  when 
these  were  commanded  to  return  to  Vetera  they 
refused,  saying  that  they  had  endured  the  hard- 
ships of  a  siege  long  enough  already.  There  was, 
therefore,  a  Roman  garrison  still  left  behind  in  the 
fortress,  of  numbers  smaller  by  more  than  a  fifth 
than  before.  It  is  true  that  a  certain  amount  of 
provisions  had  been  supplied  to  them  ;  they  were 
relieved  from  the  presence  of  a  hungry  unwarlike 
crowd  of  civilians,  and  their  defences  had  been  im- 
proved. None  the  less  they  sent  after  Vocula, 
complaining  bitterly  that  he  had  abandoned  them 
to  their  fate,  and  imploring  him  to  return.  Certainly 
they  were  left  to  themselves  in  the  middle  of  a 
savagely  hostile  country,  and  at  any  minute  the 
fate  from  which  Vocula  had  marched  to  save  them 

1   Tac.  iv.  33,  34. 


secv      REBELLION  ON  THE  RHINE 


77 


might  seem  to  threaten  them  again.  Had  they 
been  once  relieved  only  to  endure  again  the  agony 
of  suspense,  the  perils  of  assault,  the  miseries  of 
blockade  ?  The  Germans  had  vanished  ;  but  when 
the  Roman  army  withdrew  southwards,  how  could 
it  be  but  that  they  would  reappear  in  greater 
numbers  and  lively  exultation  ?  Then,  would 
Vocula  again  come  to  save  them  ?  Who  knew  but 
that  the  uncertainties  of  war  would  compel  his 
presence  elsewhere?  If  this  were  the  case  they, 
a  weakened  garrison,  must  endure  the  extremest 
penalties  at  the  savages'  hands. 

When  once  the  relief  of  Vetera  had  been  accom- 
plished at  the  cost  of  such  toil  and  fighting,  one  of 
two  courses  might  have  seemed  open  to  the  Roman 
general.  He  might  have  held  the  fort  with  his 
whole  force  and  made  it  the  base  of  operations 
for  an  advance  upon  the  foe  whom  he  had  just 
heavily  defeated.  Or,  if  he  judged  this  too  rash, 
he  might,  now  that  the  primary  object  of  his  march, 
the  rescue  of  the  garrison,  had  been  gained,  have 
evacuated  and  destroyed  Vetera,  falling  back  with 
the  whole  army  as  far  towards  Mainz  as  he  thought 
it  expedient,  taking  the  rescued  with  him.  A  very 
few  days  proved  clearly  that  the  first  of  these  plans 
was  beyond  his  strength,  owing  to  the  difficulty 
of  supplies.  The  enemy  were  by  now  complete 
masters  of  the  river.  All  supplies,  therefore,  had 
to  come  by  road  from  Novaesium.  A  first  convoy 
got  through  safely,  while  Civilis  was  recovering 
from  his  recent  defeat.  But  a  second  attempt  met 
with  woeful  results.     Vocula  had  again  despatched 


278      REBELLION  ON  THE  RHINE     ch.ih 

the  corn  collectors  with  a  strong  escort  and  waggons 
north  to  Novaesium.  The  escort  was  guilty  of 
scandalous  carelessness.  They  never  gave  a  thought 
to  the  possible  appearance  of  the  enemy,  but  stowed 
their  heavy  weapons  gaily  in  the  empty  waggons, 
and  strolled  blithely  along  the  road  beside  them  or 
wandered  over  the  countryside.  Presently  the 
convoy  halted.  There  was  an  obstacle  on  the 
narrow  road  in  front.  Up  from  ambush  sprang  the 
Germans  and  fell  upon  them.  Sheer  desperate 
fighting,  lasting  all  the  day,  did  at  the  end  carry  the 
Romans  through  to  Gelduba,  where  they  found 
protection  at  the  camp  there.  But  to  escort  a 
laden  convoy  thence  back  along  the  road  to  Vetera 
seemed  beyond  their  powers.1  It  was  clear  that 
Vocula,  if  he  stayed  at  Vetera,  had  not  men  enough 
to  guard  his  line  of  communications  and  to  carry  on 
offensive  operations  as  well. 

The  Roman  historian  himself  at  this  point  passes 
judgment  upon  Vocula  in  a  manner  which  does  little 
credit  to  the  general's  military  intelligence,  and 
none  at  all  to  his  own.  It  is  a  veritable  master- 
piece of  improbability,  almost  of  folly.  Tacitus 
states  his  belief  that,  after  his  victory  over  Civilis 
outside  Vetera,  Vocula  ought  again  immediately 
to  have  pursued  the  flying  enemy.  If,  instead  of 
this,  he  busied  himself  in  strengthening  the  fortifica- 
tions, "  as  if  another  siege  were  threatening,"  "  he 

1  I  do  not  know  why  Mommsen  {Provinces,  Eng.  Trans,  vol.  i.  p.  135) 
supposes  the  attack  on  the  convoy  to  have  been  when  it  was  proceeding  "  with 
provisions  *'  in  the  reverse  direction  from  Novaesium  to  Gelduba.  It  is  clear 
from  iv.  35  that  the  waggons  were  empty,  and  that  the  "quantum  in  regressu 
discriminis  adeundum  foret "  is  a  thought  for  the  future,  not  an  experience  of 
the  past. 


sec.v     REBELLION  ON  THE  RHINE      279 

had  misused  victory  so  often  that  he  was  rightly 
suspected  of  preferring  war." l  This  preference,  it 
is  to  be  supposed,  was  for  war  rather  than  peace, 
and  not  for  war  rather  than  victory.  Presumably 
the  general  felt  that  his  talents  were  best  displayed 
in  war,  and  therefore  desired  this  to  continue !  He 
therefore  fortified  Vetera  instead  of  pursuing  the 
enemy. 

It  is  hard  to  speak  calmly  of  such  a  judgment,  of 
him  who  passed  it,  of  those  who  seem  to  accept  it." 
Vocula  had  already  had  every  reason  to  distrust  the 
temper  of  his  own  troops,  to  appreciate  the  great- 
ness of  the  danger  threatened  by  the  Germans' 
bravery  and  cunning.  He  had  just  extricated  a 
beleaguered  garrison  with  very  serious  difficulty. 
He  was  involved  in  a  country  swarming  with  savage 
foes.  His  own  life  was  every  moment  at  stake — not 
only  imperilled  by  the  enemy,  but  also  by  mutiny 
and  treachery  among  his  own  troops.  Only  inflexible 
severity  and  success  in  war  had  kept  his  own  regi- 
ments in  hand.  Had  another  immediate  success 
been  possible  for  him,  as  Tacitus  supposes,  not 
only  military  fame  and  honour,  but  self-preservation 
itself,  must  have  compelled  him  to  do  his  utmost 
to  secure  it. 

If   Vocula    did    not    pursue    the    enemy,    either 


1  Tac.  iv.  34  :  "  Sed  Vocula,  omissis  fugientium  tergis,  vallum  tui  risque  ca- 
strorum  augebat,  tamquam  rursus  obsidium  immineret,  corrupta  totiens  victoria 
non  falso  suspectus  bellum  made." 

2  As  e.g.  in  Church  and  Brodribb  :  "  The  line  of  conduct  which  he  actually 
pursued  was  so  inexplicable  as  to  suggest  suspicions  of  treachery,  which  the 
historian  himself  seems  to  have  thought  justified  by  the  facts"  (The  History 
of  Tacitus,  p.  243).  Cm  bono  the  "  treachery  "  ?  Vocula  at  least  lost  his 
life. 


280      REBELLION  ON  THE  RHINE     ch.ih 

at  once  or  after  some  days  had  passed,  the  simple 
explanation  is  that  he  did  not  feel  himself  strong 
enough  to  do  so.  His  object  had  been  the  rescue 
of  the  garrison.  This  was  at  last  effected.  But 
pursuit  was  a  very  different  matter.  Even  after  a 
battle  fought  and  won  for  the  sake  of  victory  upon 
the  field,  immediate  pursuit  does  not  follow  as  a 
necessary  consequence. 

The  fear  of  a  return  blow  provoked  by  premature  pursuit  and 
of  losing  the  fruits  of  victory  in  the  endeavour  to  make  it  more 
complete  will  always  restrain  him  [the  commander-in-chief].  .  .  . 
Every  battle  entails  extreme  excitement  and  the  utmost  strain 
of  all  the  intellectual  and  physical  forces.  A  state  of  exhaustion 
accordingly  follows  as  a  natural  consequence.  After  a  victory, 
moreover,  there  is  a  feeling  that  further  sacrifices  are  purpose- 
less, or  that  they  would  not  be  sufficiently  recompensed  by  the 
probable  additional  results.1 

Certainly  for  immediate  pursuit  after  a  desperate 
battle,  and  with  troops  utterly  worn  out  by  marching 
and  fighting,  Vocula  could  have  neither  inclination 
nor  the  means.     For  an  advance  northwards  against 

1  Von  der  Goltz,  The  Nation  in  Anns,  pp.  362-363.  Of  course  this 
writer  insists  that  this  reluctance  to  pursue  is  due  chiefly  to  modern  conditions 
of  war:  "This  immediate  pursuit,"  he  goes  so  far  as  to  say,  "has  not  only 
nearly  always  not  taken  place  in  late  wars,  but  it  lies  in  the  nature  of  the 
modern  battle  that  it  will,  as  a  rule,  be  absent."  The  Russo-Japanese  War 
confirms  his  statement.  Cf.  too  the  very  striking  sentence  of  our  own  British 
general :  "It  is  perhaps  necessary  to  have  been  a  responsible  commander 
during  an  attack  to  realise  the  immense  reaction  of  relief  when  success  is 
attained,  a  reaction  coincident  with  an  intense  longing  to  tempt  fate  no 
further.  '  You  have  won  the  battle,'  a  voice  seems  to  whisper  in  your  ear  ; 
*  the  enemy  are  going :  for  God's  sake  let  them  go  ;  what  right  have  you  to 
order  still  more  men  to  lose  their  lives  this  day  ? ' "  (Sir  Ian  Hamilton,  A 
Staff-Officer's  Scrap-Book,  vol.  i.  p.  117).  Part  of  this  feeling  would  hardly 
be  applicable  to  Vocula,  or  indeed  to  any  other  general  in  savage  warfare. 
But  the  "  reaction"  felt  after  victory  would  be  all  the  stronger  when  the  fruits 
of  victory  were  the  very  tangible  ones  of  a  rescued  garrison,  and  these  had 
been  fully  secured  by  it. 


sEc.v     REBELLION  ON  THE  RHINE       281 

the  Germans  after  allowing  his  men  a  few  days' 
rest,  the  general  had  neither  men  nor,  as  has  been 
seen,  food  enough.  Truly  it  needed  "  the  most 
unmilitary  of  historians  "  to  suggest  that  the  suspi- 
cion, product  of  ignorance  and  malignity  combined, 
was  true,  that  Vocula  failed  to  pursue  because  he 
desired  to  protract  the  war.  This,  too,  must  be 
added  to  the  large  rubbish  -  heap  of  Tacitus' 
41  military  "  judgments. 

The  Roman  general  was  therefore  unable  either 
to  advance  against  the  Germans  from  Vetera  or  to 
remain  there  with  his  force.  But — the  real  problem 
— why  did  he  not  then  evacuate  it  altogether  ? 1 

It  cannot  be  supposed  with  probability  that 
Vocula  deliberately  intended  to  sacrifice  to  his  own 
safety  the  remnants  of  the  garrison  whom  he  left 
behind  ;  that  to  cover  a  dangerous  retreat  it  was 
necessary  to  leave  a  force  behind  him  in  Vetera, 
although  he  knew  that  this  force  would,  as  a  conse- 
quence, be  destroyed.  It  is  true  that,  if  this  situa- 
tion had  then  actually  existed,  a  Roman  general 
might  have  been  willing  to  demand  this  self-sacrifice 
of  his  rear-guard.  This  had  actually  happened  but 
a  short  while  before  in  Judaea,  where  at  Bethoron 
a   gallant    little   rear-guard    of  four   hundred    men 

1  Mommsen  supposes  that  when  the  convoy  was  cut  up  Vocula  went 
temporarily  to  Gelduba  to  its  support,  but  always  intended  to  return  to 
Vetera.  His  men,  however,  refused  to  return  "and  to  take  upon  themselves 
the  further  sufferings  of  the  siege  in  prospect ;  instead  of  this  they  marched  to 
Novaesium,  and  Vocula,  who  knew  that  the  remnant  of  the  old  garrison  of 
Vetera  was  in  some  measure  provisioned,  had  for  good  or  evil  to  follow " 
{Provinces,  Eng.  Trans,  vol.  i.  p.  135).  There  is  not  a  hint  of  anything  of 
this  in  Tacitus,  and,  though  we  may  criticise  the  motive  ascribed  him  by  the 
historian,  yet  in  the  fact  as  stated  that  he  chose  to  do  what  he  did  we  must 
needs  believe.  Surely  Mommsen's  view  is  based  on  a  wrong  interpretation 
of  the  words  "  aliis  redire  in  castra  abnuentibus  "  ? 


282       REBELLION  ON  THE  RHINE     ch.hi 

had  willingly  laid  down  their  lives  in  order  to 
ensure  the  safe  retreat  of  the  main  army.1  When 
Havelock  relieved  Lucknow,  he  found  himself  un- 
able to  extricate  the  garrison  and  non-combatants. 
He  therefore,  being  also  unable  to  keep  open  his 
line  of  communications,  allowed  himself  and  his 
force  to  be  besieged  anew  until  Colin  Campbell's 
second  army  of  relief  advanced  to  save  the  whole. 
There  were  Roman  commanders  who  would,  under 
these  conditions,  have  done  their  utmost  to  extricate 
the  troops  and  have  left  the  non-combatants  to 
their  pitiable  fate.2  But,  as  a  matter  of  fact,  things 
were  not  yet  so  desperate  with  Vocula  and  his 
army.  The  non-combatants  had  been  sent  away 
to  safety  without  difficulty.  The  whole  force  might 
have  been  withdrawn  with  still  greater  ease,  since 
the  Germans  took  some  time  to  recover  from  their 
defeat.3 

That  the  Roman  general  had  some  strategical 
object  in  view  when  he  left  a  garrison  in  Vetera  is 
certain.  That  in  this  he  made  a  bad  miscalculation 
events  quickly  proved.  That  he  himself  recognised 
this  and  made  heroic,  if  unavailing,  efforts  to  repair 


1  Cf.  my  history  of  the  principate  of  Nero,  p.  371. 

2  E.g.  Suetonius  Paulinus,  on  his  retreat  from  London  towards  Chester 
in  a.d.  60.  Cf.  my  Nero,  p.  213.  But  it  must  be  remembered  in  his  excuse 
that  his  army  was  then  the  one  and  only  hope  of  every  man,  woman,  or  child 
of  the  Romans  at  that  time  in  the  island  of  Britain. 

•!  It  was  once  suggested  to  me  by  an  Undergraduate,  in  answer  to  an 
invitation  to  a  class  in  lecture  for  suggestions,  that  Vocula  only  went  to  Vetera 
to  get  his  thousand  men,  and  did  not  care  what  happened  to  the  rest.  I  fear 
this  does  not  seem  to  me  very  probable.  In  view  of  the  difficulties  which  the 
relieving  column  had  to  face,  this  would  seem  to  be  a  case  of  plenty  to  do  and 
little  to  get,  without  Sam  Weller's  comment.  Neither  would  Vocula  have 
strengthened  the  fortifications  in  this  case.  This  suggestion,  like  Mommsen's, 
I  therefore  banish  to  a  note. 


secv     REBELLION  ON  THE  RHINE       283 

his  mistake  was  also  speedily  to  be  shown.  His 
intention  was  probably  to  keep  the  Germans  in 
check  by  a  fort  threatening  their  rear  if  they  ad- 
vanced south,  while  he  himself  was  busy  collecting 
all  available  forces  with  which  to  return  to  Vetera 
and,  using  it  as  his  base  of  operations,  penetrate 
the  enemy's  country  and  finish  the  war.  Now  that 
the  news  of  the  battle  of  Cremona  had  come,  he  may 
have  looked  for  the  speedy  arrival  on  the  Rhine  ot 
reinforcements  from  over  the  Alps,  and  have  marched 
south  to  meet  them  and  move  them  forward.1 

That  some  weeks  must  pass  before  the  struggle 
in  Italy  was  ended  Vocula  perhaps  did  not 
foresee.  And  certainly  he  did  not  anticipate  the 
series  of  disasters  which  immediately  after  his 
return  up  the  Rhine  befell  the  Roman  arms  on  the 
river.  In  holding  Vetera,  even  though  this  fort 
had  now  been  strengthened  beyond  the  fear  ot 
capture  by  assault,  he  committed  an  error  of  over- 
confidence,  somewhat  akin  to  that  of  which  the 
Federal  Government  at  Washington  was  guilty 
when,  in  defiance  of  the  advice  of  its  military 
commander,  it  ordered  the  garrison  at  Harper's 
Ferry  to  stay  at  its  post  in  September  1862.  In 
both  cases  the  fort  was  meant  to  check  the  depreda- 
tions of  a  vigorous  foe.  In  both  cases  the  fort  was 
sacrificed  and  the  garrison  lost.2  It  is  possible  that 
Vocula  had  this  reason  as  well  for  seeking  to  retain 
Vetera  in  Roman  hands,  that  he  saw  that  the  Gallic 
tribes  in  the  neighbourhood  of  the    Lower  Rhine 

1  As  suggested  to  me  by  another  Undergraduate  on  the  same  occasion. 
2  See  Note  G,  "Vetera  and  Harper's  Ferry. " 


2S4      REBELLION  ON  THE  RHINE     ch.hi 

were  becoming  restless  and  that  conspiracy  was 
hatching  among  them.  This  was  a  new  and  a 
terrible  danger.  If  the  Romans  evacuated  the  one 
great  Roman  fortress  on  the  lower  river,  this 
evacuation  might  well  be  the  spark  which  exploded 
the  mine. 

B.  The  Death  of  Vocula. — No  sooner  had  Vocula 
and  his  main  army  left  Vetera  and  marched  back 
to  Novaesium,  passing  Gelduba  on  the  way,  than 
Civilis  and  his  Germans  appeared,  following  hard 
upon  his  heels.  They  took  Gelduba,  and  their 
cavalry  pressed  forward  to  Novaesium,  outside  of 
which  place  they  met  and  routed  the  Roman  horse. 
Inside  there  raged  mutiny  and  bloodshed.  Flaccus, 
the  governor,  was  dragged  one  night  from  his  bed 
and  murdered  by  a  mob  of  soldiers.  Vocula  himself 
barely  escaped  the  same  fate  by  disguising  himself 
in  the  garb  of  a  slave.  But  the  approach  of  Civilis 
frightened  the  legionaries  back  to  their  obedience. 
Then,  however,  there  came  the  news  of  peril  on  the 
Upper  Rhine,  even  at  Mainz  itself,  which  was  being 
threatened  by  a  mixed  force  of  Germans,  belonging 
to  the  three  tribes  of  the  Chatti,  Usipi,  and  Mattiaci. 
Vocula  was  compelled  to  hasten  to  its  relief,  lest  he 
and  his  army  should  be  cut  off  completely  from  his 
communications  with  Italy  and  the  hope  of  rein- 
forcements. The  tribesmen  were  caught  unawares 
and  routed  with  loss.  Mainz  was  saved  for  the 
time.  But  meanwhile  the  whole  of  the  lower  course 
of  the  Rhine  was  left  to  itself,  and  the  rebellion 
spread  unchecked.1 

1   Tac.  iv.  36,  37. 


secv     REBELLION  ON  THE  RHINE      285 

Then  indeed,  at  the  beginning  of  the  year  a.d. 
70,  the  Roman  cause  seemed  at  the  lowest  ebb, 
and  the  Gauls  first  wavered,  then  renounced  their 
loyalty.  The  victories  of  their  German  neighbours 
excited  them ;  the  news  of  the  burning  of  the 
Capitol  became  the  text  for  the  Druids'  eloquence 
concerning  the  coming  doom  of  Rome  ;  and  rumours 
also  reached  them  of  successes  gained  upon  the 
Danube  frontier  by  Dacian  and  Sarmatian  tribes. 
It  was  surely  time  for  them  to  show  their  national 
patriotism,  when  care  for  their  own  safety  seemed 
to  suggest  this  course.  The  three  Gallic  chieftains, 
Classicus,  Tutor,  and  Sabinus,1  met  secretly  in  a 
private  house  at  Cologne,  and  their  council  was 
attended  also  by  representatives  of  the  tribes  of 
the  Ubii,  Tungri,  Treveri,  and  Lingones.  The 
conspirators  decided  to  call  the  Gauls  to  arms,  and 
to  block  the  Alpine  passes  against  the  coming  of 
fresh  troops  from  Italy.  The  infection  of  mutiny 
had  gripped  them  at  last. 

The  hatching  of  the  plot  was  at  once  betrayed 
to  Vocula  at  Mainz.  His  troops  were  fractious 
and  insubordinate.  They  still  resented  Vespasian's 
triumph,  and  grudged  to  own  him  as  their  Emperor. 
The  general  was  sorely  straitened  on  every  side. 
Yet  he  never  hesitated  or  flinched.  Certainly  he  had 
made  mistakes,  but  he  was  a  true  Roman — the  only 
one  left  upon  the  Rhine.  He  marched  at  once 
down-stream  for  Cologne,  and  thence  for  Vetera. 
With  so  deadly  a  new  danger  threatening,  the 
garrison    must    at    all    costs    be    relieved   and    the 

1  For  these  leaders  see  above,  p.  247. 


286      REBELLION  ON  THE  RHINE     ch.«i 

fort  evacuated.  He  had  left  them,  as  events  now- 
showed,  in  dire  peril.  He  would  not  abandon  them 
without  an  effort  to  save  them.  Already  he  was  well- 
nigh  within  sight  of  the  camp  when  his  auxiliary7 
leaders,  Classicus  and  Tutor,  deserted  with  their 
tribesmen  to  the  Germans.  The  traitors  had  been 
waiting  their  best  opportunity.  They  allowed  the 
general  to  surround  himself  with  foes  and  to 
see  the  object  of  his  determination  all  but  won. 
Then  they  played  him  false.  There  is  little 
that  is  sweet  about  Gallic  falseness.  Freedom 
is  better  won  by  sacrifice  than  by  black  treachery. 
Vocula  had  no  choice  left  him  but  to  retreat. 
He  withdrew  his  legionaries,  all  that  remained 
to  him,  back  to  Novaesium.  He  knew  them 
to  be  desperate,  and  not  for  one  moment  to 
be  trusted.  Emissaries  from  the  mutineers  were 
almost  openly  busy  in  their  ranks.  There  was  no 
succour,  no  refuge  for  them  nearer  than  Italy.  The 
Germans  were  up  in  their  front,  the  Gauls  of  the 
Moselle  valley  on  their  flank  and  rear ;  the  savages 
across  the  Rhine  were  separated  but  by  the  river, 
on  which  was  only  a  German  fleet.  Many  of  his 
men  preferred  a  Civilis  to  a  Vespasian.  They 
were  cowed  and  angry.  To  such  a  recreant  band 
of  men,  once  Roman  soldiers,  their  general,  Vocula, 
made  at  Novaesium  his  last  appeal.  The  purport 
of  it  was  long  remembered  in  after  years.  The 
historian,  however  great  a  master  of  the  sham 
rhetoric  of  the  schools,  could  hardly  have  invented 
a  speech  which  breathes  so  passionate  a  scorn — 
the  scorn  of  a  Roman  whose  troops  threatened  to 


sKcv     REBELLION  ON  THE  RHINE       287 

join  Germans  and  Gauls  against  the  Imperial  city. 
Their  old  comrades  of  the  auxilia  were  urging 
them  to  murder  their  officers  and  come  over  to 
them.  Vocula  knew  the  whole.  Many  implored 
him  to  escape  secretly  while  yet  there  was  time. 
But  he  despised  safety  if  so  be  that  he  could  save 
his  honour  and,  if  there  should  yet  be  shame  in 
their  hearts,  the  honour  of  his  troops  as  well.  He 
faced  them,  noisy  and  turbulent,  with  treachery  and 
murder  in  their  thoughts,  boldly  and  alone.  "  As 
to  his  own  fate,"  he  told  them,  "  he  cared  not  a 
whit.  But  the  honour  of  the  Roman  army  was 
at  stake.  What  though  fortune  seemed  to  fail 
them,  though  their  courage  seemed  for  the  moment 
shaken  ?  Could  they  forget  the  examples  of  old 
days,  those  many  times  when  Roman  legions 
perished  at  their  posts  rather  than  yield  ground 
to  the  foe  ?  Such  memorials  did  not  fail  them. 
Would  they  march  humbly  in  the  train  of  Germans 
and  of  Gauls  against  the  walls  of  Rome  ?  mount 
guard  for  a  Trevir?  ask  a  Batavian  for  the  battle 
signal  ?  For  eight  hundred  and  twenty  years  the 
army  of  the  Roman  people  had  done  homage  to 
Jupiter,  their  great  and  glorious  god,  by  offering 
the  spoils  of  countless  triumphs  won.  To  Jupiter, 
and  to  Ouirinus,  parent  of  their  city  Rome,  he 
turned  to  pray  that  they  might  never  suffer  a  Tutor 
and  a  Classicus  to  defile  the  camp  of  a  Roman 
army." 

He  ended  his  appeal,  and  a  confused  clamour  was 
heard  in  the  ranks.  But  it  was  not  the  clamour  of 
repentance    and   applause.      The    men    of  the   last 


288      REBELLION  ON  THE  RHINE     ch.™ 

Roman  army  on  the  Rhine  had  made  their  choice. 
Vocula  had  failed.  So  let  death  come  to  him  when 
it  willed.  His  very  slaves  baffled  him  when,  like  a 
Roman,  he  would  have  turned  his  sword  against 
himself.  It  mattered  very  little.  The  murderer, 
a  deserter,  was  sent  by  Classicus,  and  passed 
openly  on  his  business  through  the  ranks  of  the 
men  to  their  general's  tent.  So  Vocula  found  rest 
from  soldiering  at  last. 

A  fouler  page  of  history  was  never  written  in 
the  military  annals  of  Rome.1 

C.  The  Loss  of  Germany. — Vocula  was  dead ; 
the  other  legates  of  the  traitor  legions  were  in 
chains.  The  men  joined  the  rebels,  part  of  whom 
under  Tutor  fell  upon  Cologne  and  Mainz,  and  took 
both  without  trouble  ;  part  under  Classicus  hurried 
to  make  an  end  at  last  of  the  heroic  little  garrison 
at  Vetera,  which  Civilis  still  besieged  in  vain.  Now 
all  the  tossing  waves  of  mutiny  surged  round  the 
last  stronghold  of  the  Romans  on  the  Rhine.  It 
stood  alone,  as  a  grim  dark  rock  amid  the  foaming 
of  the  raging  western  sea.  Still  the  scanty,  hungry 
garrison  held  out  desperately.  There  was  no  one 
now  to  shout  the  news  of  relief  to  come,  nor  any 
need  for  the  barbarian  to  parade  prisoners  before  their 
eyes.  The  rebel  army,  mutineers  and  Germans, 
lay  passive  round  about  the  walls,  waiting  the  end. 
Every  living  animal  within  the  camp  was  consumed 
for  food.    The  besieged  devoured  roots  and  shrubs, 


1  Tac.  iv.  54-59.  Cf.  Mommsen  :  "  In  Roman  military  history  Cannae 
and  Carrhae  and  the  Teutoburg  Forest  are  glorious  pages  compared  with 
the  double  disgrace  of  Novaesium." 


secv     REBELLION  ON  THE  RHINE       289 

the  very  grass  in  the  streets  and  on  the  ramparts. 
Even  then  the  rebels  dared  not  storm  the  fort, 
but  won  the  hungry  men  at  last  to  surrender  by 
the  solemn  promise  of  their  lives.1  Then  all  who 
were  left  of  the  Four  Thousand  laid  down  their 
arms  and  marched  out  defenceless  through  the  gate, 
trusting  to  the  word  of  a  savage.  They  marched 
five  miles  along  the  road.  Then  the  barbarians  fell 
upon  them.  Those  who  escaped  fled  back  towards 
Vetera.  They  found  their  fort  in  flames,  and 
perished  with  it.  So  the  Four  Thousand  of  Vetera 
died,  as  died  the  garrison  of  Cawnpore. 

The  legionary  legate  Lupercus  was  saved  from 
the  massacre  to  be  sent  to  the  prophetess  Veleda2 
of  the  Bructeri.  As  he  was  being  taken  up  the 
river  Lippe  his  escort  slew  him.  A  few  of  the 
under-officers  were  kept  as  prisoners.  This  was 
the  end.  The  Roman  legions  on  the  Rhine  were 
traitors  or  destroyed.  Some  of  the  former,  men  of 
the  First  and  Sixteenth,  regiments,  were  ordered 
by  the  mutineers  to  Treves,  under  command  of  a 
certain  Claudius  the  Holy,  a  man,  says  the  historian, 
with  one  eye  lost,  repulsive  of  appearance,  and  even 
more  weak  in  intellect.  He  was  a  worthy  leader 
of  such  troops.  But  one  auxiliary  squadron  of 
Italian    horse,  the  ala   Picentina,   could    brook   the 


1  And  Tacitus  calmly  writes  :  "Donee  egregiam  Iaudem  fine  turpi  macu- 
larent,"  iv.  60.  Whose  is  the  cold  "disgrace"  if  not  his  who  cannot  realise 
the  sufferings  and  the  heroism  of  these  men  ?  No  doubt  he  felt,  as  he 
penned  the  lines,  that  he  was  the  truest  Roman  of  them  all— he,  a  stilted 
pleader  at  a  decadent  Bar. 

2  Veleda,  ap.  Statius,  Silvae,  i.  4,  90.  BeXfjSa,  a/>.  Dio  lxvii.  5.  The 
MSS.  readings  of  the  Histories  and  Germania,  c.  8,  vary  between  Veleda  and 
Velaeda. 

U 


29o      REBELLION  ON  THE  RHINE     ch.ih 

misery  and  disgrace  no  longer.  They  defiantly  left 
the  line  of  march  and  rode  bravely  back  to  Mainz, 
there  to  wait  for  better  days.  On  the  road  they 
met  by  chance  with  Vocula's  assassin  and  slew  him. 
His  name  is  given  to  us ;  but  why  should  the  scroll 
of  infamy  be  lengthened  needlessly  ? 

Only  Claudius  Labeo  now  was  left,  and  he  strove 
to  hold  out  with  his  auxiliary  corps  of  Baetasii, 
Nervii,  and  Tungri  behind  the  line  of  the  Maas. 
It  was  a  vain  hope.  His  native  troops  promptly 
joined  the  German  rebels,  and  Labeo  was  happy 
to  escape,  a  fugitive.1 

The  tide  had  reached  its  height.  The  Roman 
Army  of  the  Rhine  was  no  more.  All  Roman  forts 
were  burnt,  save  Mainz,  and  Vindonissa,  a  lonely 
fort  far  in  the  south,  then  without  a  garrison. 
Civilis  and  his  Germans  were  triumphant.  Ger- 
many was  free.  The  "  Empire  of  the  Gauls,"  the 
Imperium  Galliarum,  had,  as  it  seemed,  dethroned 
Rome  from  her  supremacy  in  northern  lands.  "  The 
whole  proud  Army  of  the  Rhine,  the  first  army  of 
the  Empire,  had  surrendered  to  its  own  auxiliaries. 
Rome  had  surrendered  to  Gaul."2 


§  6.   The  Ebb  :  Reduction  of  the  Gallic  Revolt 

A.  The  Gathering  of  the  Romans. — Then  at  last 
Rome  bestirred  herself.  It  was  now  the  spring 
of  the  new  year  a.d.  70,  and  the  Civil  War  was 
ended.  The  Flavian  cause  had  triumphed.  Ves- 
pasian   was    on    his    way    to     Italy.       Mucianus, 

1  Tac.  Lv.  60-66.      For  Labeo  see  above,  p.  257.  2  Mommsen. 


sec.  vi     REBELLION  ON  THE  RHINE       291 

until  he  came,  was  regent  at  Rome.  In  quick 
succession  the  latter  despatched  legions  from 
Italy  northwards  to  the  scene  of  war.  No  longer 
were  treacherous  auxiliaries,  half-hearted  and 
mutinous  legionaries,  captains  inert,  unskilled,  or 
betrayed,  to  contend  with  Germans  and  Gauls, 
flushed  with  victory  over  so  contemptible  a  foe. 
But  a  veritable  Roman  army  of  eight  veteran 
legions  under  brilliant  and  tried  generals  was  now 
to  strive  with  tribes  who  could  gain  freedom,  but 
who  used  it  in  heart-breaking  quarrels  among  them- 
selves. Men  who  fight  for  freedom  are  not  seldom 
apt  to  translate  it  in  terms  of  mastery  over  others. 
For  the  woes  of  the  uncivilised  at  least,  liberty  from 
Rome  was  not  a  panacea. 

Five  legions  were  sent  from  Italy  to  the  Rhine. 
Three  of  these  belonged  to  the  victorious  Flavian 
army.  These  were  the  Eighth  Augusta,  the 
Eleventh  Claudia,  and  the  Seventh  Claudia.1  Two 
others  had  been  part  of  Vitellius'  army,  the  Second 
Adjutrix2  and  the  famous  Twenty-first  Rapax. 
But  for  war  against  the  German  these  soldiers 
were  of  equal  service.  These  five  legions  crossed 
,the  Alps  by  the  three  passes  of  the  Great  St. 
Bernard  (Pennine  Alps),  Little  St.  Bernard  (Graian 
Alps),  and  Mont  Genevre  (Cottian  Alps).  Summons 
also  were  sent  to  Britain  for  the  Fourteenth  legion, 

1  A  MS.  imperfection  has  led  some  editors  to  substitute  XIII.  Gemina 
for  VII.  Claudia,  ap.  Tac.  Hist.  iv.  68.  Both  Halm  and  Heraeus  have 
omitted  the  Third  legion  altogether.  Since  E.  Ritterling's  paper  (in  the 
Westdeutsche  Zeitschrift  fiir  Geschichte  und  Kunst,  Jahrgang  xii.  (1893), 
Heft  2,  pp.  105  sqq.,  "  Zur  romischen  Legionsgeschichte  am  Rhein,"  i., 
VII.  Claudia  must  be  read. 

2  For  this  legion  see  above,  p.  214. 


292      REBELLION  ON  THE   RHINE     CH.  m 

the  old  and  deadly  foes  of  the  Batavian  cohorts,1 
and  to  Spain  for  two  legions,  the  First  Adjutrix2 
and  Sixth  Victrix.  These  speedily  arrived.  The 
army  was  divided  into  two,  and  two  commanders 
were  appointed.  Petilius  Cerialis,  the  cavalry  leader 
of  the  last  part  of  the  Flavian  advance  to  Rome,3 
was  selected  to  conduct  the  war  on  the  Lower  Rhine  ; 
Annius  Gallus,  Otho's  old  general,4  was  bidden  clear 
the  Upper  Rhine  of  rebels,  and  bring  the  hostile 
tribes  again  to  subjection.  The  larger  part  of  the 
army,  consisting  of  the  Legions  I.  Adjutrix,  VII. 
VIII.  and  XL,  was  given  to  Gallus.  Cerialis  had 
at  first  only  the  Twenty-first  legion  under  his  com- 
mand. But  it  was  not  long  before  the  remaining  three 
legions,  II.  VI.  and  XIV.,  joined  him.     And,  so  far 

1  See  above,  p.  59. 

2  There  is  no  doubt  of  the  MS.  reading  "  sexta  ac  prima  ex  Hispania 
accitae,"  iv.  68,  i.e.  I.  Adjutrix.  Halm,  however,  substitutes  "decuma"  for 
"prima,"  i.e.  Leg.  X.  Gemina,  following  a  suggestion  first  made,  so  far  as  I 
know,  by  Sir  Henry  Savile,  in  his  translation  of  the  Histories  three  centuries 
ago.  The  reason  for  the  change  seems  to  be  that  X.  Gemina,  undoubtedly  a 
Spanish  legion,  is  found  later  engaged  in  this  war  upon  the  Rhine  (Hist. 
v.  19),  while  the  presence  of  I.  Adjutrix  is  not  elsewhere  mentioned.  Also, 
it  is  urged,  the  order  VI.  and  X.  is  more  natural  than  VI.  and  I.  Ritterling, 
however  (op.  cit.  pp.  112-114),  argues  convincingly  for  the  MS.  reading  and 
destroys  the  objections.  That  I.  is  not  again  mentioned  is  due  to  the  fact 
that  it  belonged  to  the  Upper  army,  the  records  of  whose  war  are  lost,  thanks 
to  the  sudden  break  in  the  Tacitus  MS.  after  Book  v.  c.  26.  The  legion 
has,  however,  left  records  of  itself  in  Germany,  dating  to  the  years  a.d.  73 
and  following.  Cf.,  too,  Hardy,  Studies  in  Roman  History,  p.  209.  The 
cause  of  the  arrival  of  X.  Gemina,  not  given  in  Tacitus,  is  well  explained  by 
Ritterling  as  follows  : — In  the  summer  of  the  year  there  was  a  great  inroad 
of  Sarmatians  into  Moesia,  which  resulted  in  the  defeat  and  death  of  the 
governor,  Fonteius  Agrippa  (Josephus,  Beit.  Jud.  vii.  4  fin.).  Vespasian, 
therefore,  sent  Rubrius  Gallus  as  governor  to  Moesia,  and  ordered  Annius 
Gallus  to  send  Leg.  VII.  Claudia  from  Upper  Germany  to  his  help.  In  place 
of  VII.  Claudia  he  received  Leg.  XIV.  from  Cerialis  in  Lower  Germany,  and 
X.  Gemina  was  called  up  from  Spain  to  be  sent  to  Cerialis  in  place  of  Leg. 
XIV.  In  the  autumn  of  a.d.  70,  therefore,  VII.  Claudia  is  in  Moesia,  XIV. 
in  Upper  Germany,  X.  Gemina  in  Lower  Germany,  where  it  is  found,  ap. 
Hist.  v.  19. 

3  See  above,  pp.  223-225.  *  Hist.  i.  87.     See  above,  p.  70. 


sec.  vi     REBELLION  ON  THE  RHINE       293 

as  the  incomplete  records  of  the  war  are  concerned, 
the  great  brunt  of  the  fighting  fell  on  him.  Practically 
nothing  is  known  of  Gallus'  equally  successful  opera- 
tions in  the  Upper  German  province.1 

The  Twenty-first  legion,  marching  by  the  most 
direct  route  of  the  Great  St.  Bernard,  arrived  first 
at  its  old  headquarters  Vindonissa.  There  it  was 
joined  by  the  auxiliaries  of  Noricum  under  Sextilius 
Felix,  who  marched  through  Raetia  over  the 
Arlberg  Pass,  and  so  by  Feldkirch  to  the  Lake 
of  Constance  and  the  Rhine.  A  special  picked 
squadron  of  cavalry  of  mixed  nationality,  called 
the  ala  Singularium,  also  joined  the  army  here. 
Significantly  enough,  it  was  commanded  by  Civilis' 
own  nephew,  Julius  Briganticus,  whose  hatred  of 
his  uncle  was  cordially  felt  in  return  by  him.  This 
army  under  Cerialis  was  to  march  at  once  down- 
stream on  Mainz  and  Lower  Germany.  Meanwhile 
the  greater  part  of  the  troops,  diverted  over  the 
other  two  passes  to  the  valley  of  the  Rhone,  was  to 
march  up  that  river  upon  the  hostile  Gallic  tribe  of 
the  Lingones.  These  subdued,  this  army  under 
Gallus  could  either  threaten  the  Treveri  on  their 
rear  and  thus  secure  Cerialis  from  their  attack,  or, 
if  Cerialis,  operating  from  the  Rhine,  had  already 
received  their  submission,  could  march  by  Besancon 
for  the  Upper  Province  to  complete  the  work  in 
that  district  which  Cerialis,  pressing  ever  north- 
wards, had  left  unfinished.2 

1  Tacitus'  unfinished  MS.  tells  us  nothing  of  this.      See  below,  p.  315. 

2  This  is  the  strategy,  I  think,  to  be  deduced  from  the  very  fragmentary 
hints  in  Tacitus'  narrative.  It  is  clear  that  the  Lingones  had  to  be  subdued, 
and,  from   Frontinus,   Strat.   iv.  3.   14,  that  they  were  actually  so  subdued. 


294      REBELLION  ON  THE  RHINE     ch.  m 

B.  The  Struggle  with  the  Treveri. — While  this 
Roman  army  was  gathering  to  reap  the  harvest  of 
vengeance,  all  was  confusion  and  dissension  in  the 
rebel  ranks.  The  insurgents  had  not  even  to 
wait  for  the  coming  of  the  Romans  to  suffer 
their  first  reverse.  The  Lingones  under  Sabinus 
attacked  their  neighbours  the  Sequani  and  were 
rudely  repulsed.  This  first  blow  to  Gallic  Unity,1 
added  to  the  rumours  of  the  approach  of  the 
new  Roman  army,  caused  the  feeling  in  Gaul 
to  change.  The  new  movement  was  speedily 
voiced  in  a  great  Gallic  Council,  which  itself  was 
called  together  by  the  Remi,  the  tribe  inhabiting 
the  region  between  the  rivers  Marne  and  Aisne, 
a  folk  long  since  notorious  for  its  loyalty  to  the 
Romans.      The  Council  voted  for  submission  and 

When,  however,  this  happened  is  uncertain,  but  that  it  befell  early,  and  at 
Gallus'  hands  in  co-operation  with  Cerialis'  advance  north,  seems  to  me 
probable.  For  the  army  destined  for  the  Lower  Rhine  must  hasten  forward 
as  speedily  as  possible,  but  must  not  have  its  advance  endangered  on  its  left 
rear.  If  Cerialis  and  not  Gallus  subdued  the  Treveri  in  the  Moselle  valley, 
this  was  due  to  the  facts,  probably,  that  the  Lingones  gave  Gallus  some 
trouble,  that  the  Treveri  were  cut  off  from  the  Lingones  by  the  Mediomatrici 
higher  up  the  valley  (cf.  below),  and  turned  their  whole  attention,  therefore, 
to  the  Rhine,  and  that,  therefore,  Cerialis  could  not  afford  to  advance  upon 
the  Germans  leaving  his  rear  endangered  by  them.  It  is  a  great  pity  that 
even  the  Tacitean  account  of  Gallus'  operations  is  not  preserved  to  us. 

At  Mirebeau-sur-Beze,  thirteen  miles  north-east  of  Dijon,  and  so  in  the 
Lingones'  land,  were  recently  found  building  tiles  stamped  Vexilla  legionum, 
with  marks  of  the  legions  I.  VIII.  XL  XIV.  XXI.  Mommsen-a/. 
Hermes,  xix.  pp.  437-441 — regarded  this  as  evidence  of  a  reserve  depot  built 
by  the  detachments  of  these  legions  during  their  advance  to  the  north  in  this 
year  a.d.  70.  But  there  are  difficulties  in  the  way  of  this  view,  e.g.  the 
presence  of  vexilla  of  XIV.  and  XXI. — for  XIV.  as  a  whole  has  not  yet 
arrived  from  Britain  (cf.  iv.  79),  and  XXI.  is  only  heard  of  as  being, 
apparently  as  a  whole,  at  Vindonissa.  Ritterling  (op.  cit.  pp.  1 16-120), 
followed  by  Hardy  (op.  cit.  p.  215),  therefore  refers  these  tegulae  to  the 
muster  of  the  Upper  German  army  for  Domitian's  war  against  the  Chatti  in 
A.D.  83,  or  against  the  rebel  Antonius  Saturninus  five  years  later.  Probably, 
therefore,  they  are  not  to  be  connected  with  the  strategy  of  a.d.  70  or  the 
reduction  of  the  Lingones,  though  the  idea  is  a  tempting  one. 

1   "  Sequanorum  prospera  acie  belli  impetus  stetit,"  Tac.  iv.  67. 


src.  vi    REBELLION  ON  THE  RHINE      295 

peace.  Old  inter-tribal  animosities  determined  the 
vote.  Only  Treveri  and  Lingones  refused  com- 
pliance, and  their  warriors  still  remained  in  the 
field.  But,  even  so,  no  concerted  action  was  taken 
by  the  three  chief  rebel  leaders — Civilis  of  the 
Germans,  Classicus  and  Tutor  of  the  Gauls.  Their 
preparations  to  meet  the  coming  attack  were  scanty 
and  inadequate.  To  Tutor  the  task  of  blocking  the 
Alpine  passes  does  seem  to  have  been  entrusted ; 
but  he  left  them  serenely  alone,  and  the  Romans 
had  no  difficulty  in  crossing  any  one  of  them. 
Civilis  went  gaily  hunting  after  the  slippery  fugitive 
Labeo,  who  lightly  baffled  all  his  efforts  to  catch 
him.  Classicus  peacefully  rested  upon  his  uncertain 
laurels.1 

Such  efforts  at  defence,  however,  as  Tutor  made 
succeeded  in  collecting  a  considerable  army  com- 
posed not  only  of  the  Treveri  with  infantry  and 
cavalry  of  the  sometime  Roman  army,  but  also  of 
new  levies  furnished  by  three  small  tribes — the 
Triboci  in  Lower  Alsace,  the  Vangiones  in  the 
district  of  Worms,  and  the  Caeracates.2  With  this 
force  Tutor  at  first  showed  some  activity.  Sextilius 
Felix  was  in  command  of  the  advance  guard  of 
Cerialis'  army,  and  sent  forward  one  auxiliary  cohort 
to  reconnoitre  on  the  march  from  Vindonissa.  This 
cohort  came  into  touch  with  Tutor's  men  and  was 
destroyed.  But  on  the  advance  of  the  Roman  army 
in  strength,  the  Gaul's  force  melted  away  rapidly. 


1  Tac.  iv.  67-70. 

2  A  tribe  not  elsewhere  mentioned,  but  perhaps  situated  on  the  left  bank 
of  the  Rhine  behind  Mainz. 


296      REBELLION  ON  THE  RHINE 

The  former  veterans  of  the  Roman  army  returned 
promptly  to  their  old  allegiance,  and  the  native 
soldiers  of  the  three  tribes  followed  them  over  into 
the  Roman  camp.  Tutor  was  left  with  none  but 
the  Treveri  to  obey  him.  He  was  therefore  forced 
to  fall  rapidly  back  before  the  Roman  advance, 
and,  avoiding  Mainz,  now  garrisoned  by  the  ala 
Picentina  which  wished  him  no  good,1  he  retreated 
to  the  northern  bank  of  the  river  Nava,  where  he 
hoped  to  be  able  to  make  a  stand.  This  small 
stream,  the  modern  Nahe,  flows  into  the  Rhine 
between  the  townships  to-day  of  Bingen  on  its 
right  and  Bingerbrtick  on  its  left  bank.  At  the 
latter  was  the  Roman  town  of  Bingium,  and  the 
road  from  Mainz  northwards  crossed  the  Nahe  by 
a  bridge  to  the  town.  This  bridge  was  destroyed 
by  the  Gauls  who  lined  the  farther  bank  to  frustrate 
the  Romans'  passage  of  the  stream.2  But  Felix, 
on  arriving  opposite  the  enemy's  position,  was  not 
long  baffled.  A  deserter  showed  him  a  ford,  and 
Tutor's  men  were  driven  from  their  position. 
Tutor  fled,  and  the  tribesmen  were  scattered  and 
sorely  dismayed. 

By  this  time  Cerialis  himself  had  arrived  with 
the  legionary  army  at  Mainz.  Evidently  distrusting 
his  Gallic  auxiliaries,  he  dismissed  them  to  their 
homes,  saying  to  them  briefly  that  a  war  undertaken 
by  Roman  troops  needed  no  help  from  them,  but 
was  as  good  as  ended  already.     The  Gauls  retired 


1  See  above,  pp.  2S9-290. 

2  This,  as  Heraeus  says,  must  have  contained  more  water  in  Roman  days 
than  it  does  to-day. 


secvi     REBELLION  ON  THE  RHINE       297 

both  thankfully  and  humbly — "  proniores  ad  officia 
quod  spernebantur."  1  Cerialis'  action  was  not  the 
result  of  disdainful  self-confidence,  but  rather  of 
great  wisdom  and  insight  into  the  native  character. 
After  the  recent  disasters  which  had  befallen  Roman 
troops  on  the  Rhine,  it  was  good  policy  for  the 
new  general  to  openly  assure  the  Gauls,  by  word 
and  deed,  that  even  they  were  not  indispensable. 
British  officers  have  before  now  used  similar  methods 
with  native  troops,  and  with  good  results. 

At  Mainz,  Cerialis  quickly  decided  that  his  next 
step  must  be  the  reduction  of  the  Treveri  in  the 
Moselle  valley.  Eager  as  he  was  to  penetrate  to 
the  heart  of  the  mutiny  in  Lower  Germany,  he 
could  not  advance  beyond  Coblenz,  where  the 
Moselle  enters  the  Rhine,  unless  he  had  secured 
himself  from  attack  on  the  rear  by  these  most 
troublesome  Gauls.  .  Gallus  was  engaged  to  the 
south  of  them  with  their  allies  the  Lingones.  The 
Treveri  could  not  at  once  be  left  to  him.  The 
time  had  come  to  make  an  end  of  their  resistance. 
Already  they  were  isolated  from  help.  The  traitor 
legionaries  of  the  First  and  Sixteenth  legions  who 
had  been  sent  to  their  capital  city,  now  Treves, 
had  felt  the  prick  of  repentance  as  soon  as  the 
Roman  army  had  forced  the  passage  of  the  Nahe. 
They  solemnly  administered  to  themselves  the  oath 
of  loyalty  to  Vespasian,  and,  though  at  once  the 
boy  rebel-leader  Valentinus  hurried  to  the  town, 
they  remained  defiant  of  him,  and  marched  away 
up-stream  to  the  friendly  folk  of  the  Mediomatrici, 

1  Tac.  iv.  71. 


293       REBELLION  ON  THE  RHINE     ch.hi 

centred  round  the  modern  Metz.  Here  they  halted 
and  waited  cautiously  on  the  development  of  events. 
To  Valentinus  and  to  Tutor,  who,  after  his  defeat, 
had  found  his  way  also  to  Treves,  was  left  only  the 
melancholy  pleasure  of  butchering  in  cold  blood  the 
two  captive  legates  of  the  legions.  So  Freedom 
was  justified  of  her  barbarian  children. 

The  Treveri  now  prepared  for  resistance  to  the 
last,  being  greatly  encouraged  by  Valentinus'  youth- 
ful energy  and  raging.  The  road  to  their  capital 
city  left  the  Rhine  north  of  the  inflow  of  the  Nahe, 
and  crossed  undulating  country  to  the  Moselle 
below  Neumagen.  Thence  it  ran  to  Treves,  keep- 
ing on  the  right  bank  of  the  river.  From  Bingen 
to  Treves  the  distance  is  some  seventy  miles. 
Cerialis  would  doubtless  advance  by  this  road. 
His  intention  to  attack  them  was  soon  discovered 
by  the  tribe.  Valentinus,  therefore,  and  his  army 
moved  out  of  the  capital  six  miles  down-streamf 
where  they  took  up  and  fortified  a  position  at 
Rigodulum,  the  modern  hamlet  of  Riol,  on  rising 
ground  overlooking  the  Moselle.  Encamped  here, 
they  covered  the  approach  to  their  city.  Here, 
therefore,  Valentinus  waited  for  Cerialis'  coming. 

The  other  rebel  leaders,  Civilis  and  Classicus, 
on  hearing  of  Tutor's  defeat  at  Bingium,  had 
joined  forces,  and  now  sent  to  Treves  bidding 
Valentinus  not  to  fight.  It  is  possible  that  their 
plan  was  to  evacuate  the  valley  of  the  Moselle  and 
to  draw  the  soldiers  of  the  Treveri  north  to  join 
their  own  main  army.  If  this  was  their  intention, 
it  failed, — partly  owing  to  the  natural  reluctance  of 


skc.v,     REBELLION  ON  THE  RHINE       299 

the  tribesmen  to  surrender  their  homes  to  the 
enemy  ;  partly,  perhaps,  because  the  boy  leader 
scorned  the  counsel  of  the  older  men  ;  partly  by 
reason  of  Cerialis'  rapidity  of  movement.  This 
general  having  once  decided  upon  a  short  campaign 
against  the  Treveri,  wasted  no  time.  In  three  days 
he  marched  his  men  sixty  miles  up  the  Moselle, 
and  was  upon  the  native  army. 

There  was  but  little  spirit  left  in  the  tribesmen, 
and  they  made  but  a  feeble  defence  of  the  position 
at  Riol.  The  hill  had  the  Moselle  on  its  left,  and 
the  brook  of  the  Fellerbach  circling  round  its  rear. 
Its  crest  was  lined  by  the  defenders.  The  position 
mav  be  sketched  as  follows  : — 


To  Coblenz 


'cselle  R  1  r  e~    _ 

— ;^^«»™">^  ^-^     Lines  of  attack  by 

%■*       1i   ^^^Roman  infantry 
^-Crqman^  caualrgjidvapji8.'-' 

While  the  Roman  cavalry  were  sent  round  the 
hill  by  the  slopes  between  its  crest  and  the  brook, 
the  infantry  were  launched  in  a  frontal  attack  up 
hill  against  the  foe.  They  stormed  it  with  vigour 
and  success,  and  Valentinus'  army  was  hurled,  a 
routed  mass  of  fugitives,  down  the  further  slope 
upon  the  cavalry  waiting  to  receive  them  and  hew 
them  down.  They  went  down  the  hill,  says  the 
Roman  historian,  like  a  house  falling.1     Valentinus 

1   "  Ruinae  modo. " 


3oo      REBELLION  ON  THE  RHINE     ch.  m 

himself  was  taken  prisoner,  but  Tutor,  if  he  ever 
took  a  part  in  the  defence,  escaped.  Next  day 
Cerialis  entered  Treves  unopposed.  There  he  was 
presently  met  by  the  repentant  legionaries  from 
Metz.  He  had  sent  for  these  to  co-operate  with 
him  in  his  attack  on  the  Treveri  by  advancing 
upon  the  tribesmen's  rear.  They  arrived,  however, 
too  late  to  take  any  part  in  the  engagement  at  Riol. 
They  were  pardoned  by  the  general,  and  received 
into  his  army.  The  Treveri  also,  and  such  of  the 
Lingones  as  were  with  them,1  made  their  formal 
submission,  which  was  accepted,  and  no  further 
penalty  was  imposed  on  them.  The  Roman 
general  prepared  to  stay  for  some  days  in  the 
town,  until  reinforcements  should  reach  him  and 
enable  him  to  essay  the  last  and  most  perilous  part 
of  the  campaign  by  moving  against  the  Germans. 
Meanwhile  he  busied  himself  in  receiving  the 
tribes'  submission,  and  in  speech -making.  All 
seemed  safe  on  the  Moselle,  but  in  fact  was  far 
from  being  so.2 

For  while  the  Roman  army  lay  resting  in  the 
town  by  the  river,  on  the  hills  to  the  north  of  it 
the  tribesmen  were  gathering  in  great  numbers. 
Civilis  and  Classicus  themselves  had  hastened 
towards  the  town,  and  Tutor  joined  them.  Lin- 
gones and  Batavians,  Ubii  and  Tencteri  and  Bruc- 
teri,  all  were  massing  together  under  shelter  of  the 

1  This  and  iv.  77  are  the  only  mentions  of  the  Lingones  in  connection 
with  the  whole  campaign  against  the  Treveri.  It  does  not  seem  at  all 
probable  that  the  whole  tribe  was  engaged  in  the  defence  of  their  allies' 
capital.  But  again  we  cannot  be  certain  of  anything  about  them  in  the 
absence  of  information  about  Gallus'  movements. 

2  Tac.  iv.  70-74. 


secv.     REBELLION  ON  THE  RHINE      301 

friendly  hills.  A  great  storm  was  preparing  to 
sweep  down  upon  the  valley  which  reposed  at  last 
so  peacefully  beneath. 

This  news  of  the  gathering  of  the  tribes  reached 
Cerialis  in  due  course,  and  roused  him  from  his  sense 

of  security.      Ov  yap  \6yots  irvpyovrat  tt6\i<;.      At  once 

he  issued  orders  that  the  legionaries  should  fortify 
their  camp  at  Treves.  The  general,  says  the  Roman 
historian,  was  blamed  by  many  for  allowing  the 
natives  to  collect  together  on  the  hills  undisturbed. 
But  only  the  sending  out  of  flying  columns  could  have 
hindered  this,  and  this  method  not  only  demanded 
more  men  than  Cerialis  had  as  yet  at  his  disposal, 
but  it  also  was  far  too  dangerous ;  for  such 
columns  might  easily  have  been  entrapped  and  cut 
to  pieces  among  the  hills  by  the  swift  tribesmen. 
No  blame  attaches  to  the  Roman  on  this  account ; 
but  he  can  hardly  escape  censure  for  his  serious 
failure  to  appreciate  beforehand  the  suddenness  and 
ferocity  of  that  favourite  Gallic  device,  a  surprise.  /- 
The  city  of  Treves,  Colonia  Augusta  Treverorum, 
was  founded  as  a  Roman  town  perhaps  by  Augustus, 
and  owed  its  colonial  status  to  the  Emperor 
Claudius.  It  lay  on  the  right  bank  of  the  Moselle. 
Its  Roman  remains,  covered  some  of  them  with 
creepers  and  greenery,  far  surpass  in  beauty  those  of 
any  other  Roman  town  in  Europe  whose  picturesque- 
ness  has  been  spoilt  for  ever  by  the  excavator's 
spade.  These,  however,  date  to  a  later  time  than  the 
first  century  of  our  era.  But  still  to  the  spectator, 
standing  on  the  vine-clad  hills  to  the  north-west  of 
the  city,  and  looking  down  on  its  towers  and  rose- 


302       REBELLION  ON  THE  RHINE     ch.  m 

red  walls  in  the  rich  plain  at  his  feet,  where  one 
busy  bridge  spans  the  rapid  river,  it  is  easy  to  see 
again  the  little  slumbering  settlement,  the  Roman 
bridge,  the  camp  of  the  Roman  army  on  the  left 
bank  beyond  the  bridge,  the  drowsy  sentinels, 
hardly  aware  that  dawn  is  already  breaking  on 
the  surrounding  hills,  and  the  wild  onrush  of  the 
Gauls,  striking  their  last  blow  in  history  for 
freedom  from  the  Roman.  The  enemy  rushed  to 
the  assault  in  triple  column.  The  Batavians  came 
swarming  down  from  the  heights  which  overhung 
the  camp  on  north  and  west ;  Ubii  and  Lingones 
hastened  up  the  road  which  led  from  the  camp 
northwards  down  the  river's  left  bank ;  Bructeri 
and  Tencteri  rushed  through  the  gap  left  between 
road  and  river.  A  sketch  may  serve  to  illustrate 
the  onslaught : — 


On  the  night  of  the  attack  Cerialis  was  sleeping, 
not,  as  was  his  duty,  in  the  general's  quarters  in 
centre  of  the  camp,  but  in  the  town.     The  sentries, 


Sec.  vi    REBELLION  ON  THE  RHINE      303 

probably  on  that  account,  were  the  more  careless. 
The  enemy  were  upon  them  and  over  the  ramparts 
before  any  alarm  was  given.  Then,  as  at  the 
surprise  of  Gelduba,  followed  a  scene  of  wild  con- 
fusion, slaughter,  and  plunder.  The  legionaries 
fought  desperately  enough  in  little  knots  of  men, 
and  their  officers  sought  to  cheer  them  on  to  a 
stout  resistance.  But  their  exultant  and  agile  foes 
rushed  through  the  camp  and  seized  the  bridge, 
driving  over  it  a  mob  of  terrified  fugitives.  At  the 
town  end  of  the  bridge  their  general  met  them. 
Hastily  roused  from  sleep,  Cerialis  had  hastened  to 
the  noise  of  the  fighting,  and  now  played  verily  the 
man.  To  rally  some  of  the  fleeing,  and  with  them 
make  a  fierce  attack  on  the  bridge,  was  the  work 
of  a  moment.  The  bridge  was  retaken,  and  the 
general,  at  the  head  of  such  troops  as  he  could 
muster,  crossed  it  to  the  camp.  His  coming  saved 
the  day,  by  that  time  well-nigh  lost.  His  entreaties 
and  rebukes,  his  energy,  as  he  hastened  from  post 
to  post  reckless  of  his  life,  restored  discipline  and 
courage  to  the  legionaries.  The  steadiness  of  the 
Twenty-first  legion  shamed  the  wavering  men  of 
the  two  unlucky  traitor  legions,  but  newly  restored 
to  the  rank  of  Roman  soldiers.  The  Germans  and 
Gauls,  thinking  that  the  victory  was  won,  were 
already  scattered  far  and  wide  through  the  camp, 
gleefully  gathering  up  the  spoil.  But  their  chief 
spoil  that  day  was  the  saving  of  their  lives — by 
the  few  who  at  the  last  escaped  back  over  the  ram- 
parts to  the  hills,  leaving  the  camp  strewn  thick 
with    the   bodies   of  their  slain.      Treves  and  the 


304      REBELLION  ON  THE  RHINE     ch.ih 

Roman  army  were  saved.  That  same  day  the 
enemy's  camp  on  the  heights  was  stormed,  and  the 
foe  melted  away  to  the  north.  Civilis,  Tutor,  and 
Classicus  saved  themselves  by  flight.  A  handful 
of  the  leading  men  of  the  Treveri1  still  followed 
their  leaders'  fortunes.  But  the  resistance  of  this 
tribe  was  now  finally  at  an  end.2 

C.  The  Advance  to  Cologne. — But  Cerialis  could 
not  tarry  longer  at  Treves.  He  and  his  army  were 
needed  urgently  at  Cologne,  and  it  was  now  safe 
for  him  to  continue  his  advance  down  the  Rhine 
towards  that  city.  Its  inhabitants,  the  "  Agrippi- 
nenses,"  easily  the  most  cultured  and  Romanised  of 
all  the  Germans  on  the  river,  had  remained  loyal  to 
the  Roman  cause  so  long  as  they  dared,  and  their 
city  had  hardly  escaped  destruction  at  the  hands 
of  the  angry  Germans  across  the  Rhine  in  their 
hour  of  victory.  A  timely  recognition  by  the 
Agrippinenses  of  facts  and  the  humble  answer  which 
turns  away  wrath  had  saved  their  city.  But  now 
again  the  tide  of  German  triumphs  had  turned  and 
was  sweeping  fast  out  to  sea.  The  citizens  were 
eager  to  return  to  their  old  faith,  and  to  propitiate 
the  wrathful  Romans  by  a  sweet-smelling  sacrifice. 
Twenty  miles  away  to  the  south-west  of  Cologne, 
at  Tolbiacum,  the  modern  ZiAlpich,3  Civilis  had 
placed  in  garrison  one  of  his  most  warlike  and 
valued  cohorts,  of  Chauci  and  Frisii  combined. 
Now,  fresh  from  the  scene  of  his  bitter  defeat  at 

1  Tacitus  is  careful  to  give  a  most  precise  number,  viz.  1 13  !  (v.  19). 

2  Tac.  iv.  75-78. 

3  Famous  later  for  the  defeat  here  in  a.d.  496  of  the  Alemanni  by  the 
Franks,  and  Clovis'  consequent  conversion  to  Christianity. 


sec.v!    REBELLION  ON  THE  RHINE      305 

Treves,  the  German  leader,  sore  and  angry,  was 
hurrying  towards  Cologne,  where  his  own  wife  and 
sister,  together  with  Classicus'  daughter,  had  been 
left  as  pledges  of  the  alliance.  The  anxious  citizens 
resolved  to  carry  out  at  once  their  desperate  resolu- 
tion. The  unsuspecting  Germans,  scattered  through 
the  houses  in  the  city,  were  massacred.  The 
famous  cohort  at  Ziilpich  was  invited  to  a  banquet 
and  there  largely  entertained,  while  wine  flowed 
freely.  As  the  guests  lay  buried  in  drunken 
slumber,  their  hosts  stole  from  the  hall  of  feasting, 
made  fast  the  doors,  and  burnt  the  whole  with 
fire  to  the  ground.  Then  the  Agrippinenses  sent 
begging  Cerialis  to  march  instantly  to  their 
succour,  and  save  them  from  the  vengeance  of 
Civilis. 

By  forced  marches  the  Roman  general  out- 
stripped the  Germans  and  reached  the  town  in 
time.  Civilis,  at  the  bitter  news,  turned  aside,  and 
retreated  northwards,  sorrowing  for  his  lost  cohort. 
But  though  Gaul  was  also  lost  to  him,  though  his 
women-folk  were  prisoners  in  the  Romans'  hands, 
the  courage  of  the  rebel  general  never  failed  him. 
The  Gauls  must  be  let  make  their  peace  with  the 
foe.  The  Treveri  and  Lingones  had  at  last  been 
quelled.  The  Nervii  and  their  German  neighbours, 
the  Tungri,  followed  the  example  ;  for  the  Four- 
teenth legion,  landing  at  Boulogne  from  Britain, 
marched  through  the  territory  of  these  tribes  on  its 
way  to  the  Rhine,  and  scared  them  into  submission. 
Yet  for  all  this  Civilis  was  not  moved.  Still  he  had 
his  Germans,   Batavians,  and  Cannenefates  left  to 

x 


306      REBELLION  ON  THE  RHINE     CH.  m 

him.  With  these  he  had  begun  the  revolt.  With 
these  he  had  driven  the  Romans  from  the  Rhine. 
With  these  he  would  yet  maintain  his  cause.  A 
couple  of  small  reverses  soon  "  spoilt  the  fame  of 
the  victory  "  at  Treves,  and  showed  to  Cerialis  that 
the  Germans  were  yet  to  be  subdued.  A  small 
Roman  flotilla,  known  as  the  Classis  Britannica, 
which  kept  guard  in  the  North  Sea  and  Channel, 
was  attacked  on  the  German  coast  by  the  Can- 
nenefates  and  dispersed.1  And  one  of  Cerialis' 
advance  squadrons  of  cavalry  was  cut  to  pieces  by 
Civilis  at  Novaesium.  The  Gallic  bid  for  Empire 
had  failed ;  but  the  German  army  mustered  north 
of  Vetera  still  dauntless.  There  were  mutineers 
yet  in  arms,  and  Cerialis'  hardest  task  lay  before 
him.2 


§  7.    The  Submission  of  the  Germans3, 

Cerialis  found  no  difficulty  in  advancing  from 
Cologne  down  the  Rhine  until  he  came  again  to  the 
neighbourhood  of  Vetera,  where  Civilis  had  col- 
lected together  his  largest  possible  army,  intending 
to  make  a  resolute  stand  at  this  place.  Both  the 
Roman  and  the  German  general  had  received  re- 
inforcements since  they  had  met  in  their  first 
encounter  at  Treves.  Cerialis'  strength  had 
been  more  than  doubled  by  the  arrival  of  three 
new  legions — the  Fourteenth  from  Britain,  and  the 

1  Probably  it  had,  as  Mommsen  suggests,  just  conveyed  the  Fourteenth 
legion  from  Britain  to  Gaul. 

2  Tac.  iv.  79.      Tacitus'  narrative  is  now  continued  first  in  v.  14. 

3  Tac.  v.  14-26. 


sec.  vii    REBELLION  ON  THE  RHINE       307 

Second  Adjutrix  with  the  Sixth  Victrix  from  Spain. 
Numerous  auxiliary  troops,  both  horse  and  foot, 
were  now  added  to  the  legionaries.  Civilis  had 
persuaded  men  of  the  Cugerni  and  Transrhenane 
tribes  to  join  his  banner.  Germany  was  in  the  field 
against  Rome. 

It  was  by  this  time  autumn.  The  great  river, 
which  in  the  preceding  year  had  caused  such  trouble 
to  Vocula  and  the  other  Roman  commanders  by 
its  scanty  stream  and  shallows,1  at  this  time 
rolled  a  full  flood  to  the  sea,  and  the  low  ground 
round  Vetera  was  a  great  morass  of  mud  and 
swamp.  In  such  ground  for  battle  the  lightly-armed 
Germans  delighted.  Used  to  swimming  from  their 
childhood  up,  strong-limbed  and  tall  in  stature,  the 
natives  of  the  Lower  Rhine  had  no  fear  of  sudden 
plunges  into  treacherous  pools,  of  quaking  ground, 
or  hurrying  stream.  But  the  short,  sturdy  legionary 
of  Rome,  encumbered  with  heavy  armour,  and  easily 
lured  in  his  ignorance  on  to  treacherous  ground, 
fought  but  badly,  because  with  little  confidence,  on 
any  but  firm  soil.  Certainly  the  German  river  was 
striving  to  help  its  children. 

The  spot  first  selected  by  Civilis  in  the  neigh- 
bourhood of  Vetera  on  which  to  offer  battle  was 
carefully  chosen  and  prepared.  Swampy  by  nature, 
a  dam  cleverly  thrown  into  the  Rhine  from  the 
bank  had  impeded  the  flow  of  the  stream, 
and  added  thereby  to  the  depth  of  water  on 
the  ground.  The  river  was  on  one  side  of  it — a 
refuge   for   swimmers   if  they   were   driven  off  the 

1  See  above,  p.  269. 


3o8      REBELLION  ON  THE  RHINE     ch.  m 

field.  And  the  memory  of  triumphs  already  won 
at  Vetera  spurred  the  Germans  on  to  emulate  their 
former  deeds.  The  conflict,  skilfully  provoked  by 
the  German  and  rashly  accepted  by  the  Roman, 
ended  in  a  bad  reverse  for  the  latter.  The  swamp 
was  like  that  road  of  historic  fame  in  Virginia  on 
which  the  Federal  officer,  sent  to  reconnoitre  it, 
reported  that  the  road  was  there,  but  "  he  guessed 
the  bottom  had  fallen  out."  Water  and  mud  suc- 
cessfully worsted  Cerialis'  struggling  men,  and  for 
the  moment  ill-fortune  seemed  again  to  haunt  the 
Romans  on  the  Rhine.  But  the  check  was  only 
for  a  day.  Next  morning  Cerialis  renewed  the 
battle ;  and,  after  a  stubborn  contest,  a  deserter 
showed  the  cavalry  a  path  by  which  they  could 
skirt  the  morass  on  firm  ground  and  fall  upon  the 
Germans'  flank.  This  decided  the  battle,  and  the 
enemy  fled.  But  the  Roman  fleet,  which  Cerialis 
had  expected  to  appear  to  cut  off  the  German  flight 
across  the  Rhine,  did  not  come.  The  cavalry 
pursuit  was  checked  by  heavy  rain  and  nightfall. 
And  the  Germans,  therefore,  made  good  their 
retreat  without  serious  loss.1 

Civilis  by  this  defeat  was  compelled  to  cross  the 
Waal  into  his  last  refuge,  the  "  Island  of  the 
Batavians."2  He  therefore  evacuated  the  Oppidum 
Batavorum  (which  was  built  probably  on  the  site 
of   Lohengrin's    town,    Cleve,    some    seven    miles 

1  This  part  of  Tacitus'  narrative  is  made  both  dull  and  unreal  by  a 
number  of  invented  speeches.  Their  pretty  rhetorical  tropes  do  but  hinder 
the  military  narrative,  and  I  omit  them  all. 

2  For  the  whole  of  the  following  narrative,  the  plan  of  the  "  Island  "  on 
p.  233  must  be  consulted. 


sec.  vn    REBELLION  ON  THE  RHINE       309 

south-east  of  the  parting  of  the  channels),  carried  off 
all  that  he  could  from  it,  and  burnt  it  to  the  ground. 
To  add  to  the  security  of  his  position  in  the  island, 
he  destroyed  Drusus'  mole,  and  thus  diverted  the 
greater  bulk  of  the  waters  of  the  Rhine  from  its 
northern  arm,  the  Lek,  into  its  southern,  the  Waal.1 
The  Romans,  he  judged,  had  not  enough  vessels 
to  bridge  the  greatly  swollen  waters  of  the  latter 
channel,  and  his  own  communications  over  the 
shallow  northern  branch  with  his  friends  beyond 
were  made  both  safe  and  easy.  Once  in  the  Island, 
though  driven  back  like  a  hunted  beast  to  its  lair, 
he  turned  savagely  and  stood  at  bay.  Now,  too, 
the  Chauci  sent  him  men  besides  to  help  him  in  the 
defence  of  his  "  Island  "  home. 

Cerialis  and  his  army  ~  pushing  northwards  found 
themselves  stopped  by  the  Waal.  There  was  no 
help  for  it.  The  river  must  be  bridged  before  they 
could  get  to  grips  with  the  foe.  And  the  year  was 
growing  old ;  the  river,  swollen  by  autumnal  rains, 
was  rising  ever  higher.  There  seemed  little  pro- 
mise of  a  speedy  finish.  Cerialis  distributed  his 
army  along  the  southern  bank  of  the  river  ;  sent 
emphatic  messages  bidding-  the  tardy  fleet  come  at 
once  to  his  help ;  and  ordered  the  winter  quarters 
for  the  troops  to  be  rebuilt  at  Novaesium  and 
Bonn.3  His  main  army  was  divided  between  four 
camps  on  the  Waal.     The  two  eastern  camps  were 

1  See  above,  pp.  236-237. 

2  Leg.  X.  Gemina  from  Spain  now  takes  the  place  of  Leg.  XIV.  sent  to 
Upper  Germany.      See  note  on  p.  292. 

:!  Destroyed,  after  their  victory  over  Vocula,  by  the  Germans  in  the  pre- 
vious year. 


310      REBELLION  ON  THE  RHINE     ch.hi 

allotted  to  the  legionaries.  These  were  Arenacum, 
given  to  the  Tenth  legion,  which  perhaps  was 
situated  at  Ryndern  by  Cleve  ;  and  Batavodurum, 
the  camp  of  the  Second  legion,  almost  certainly 
at  Nymwegen.  Here,  too,  the  soldiers  began  to 
attempt  the  building  of  a  bridge.1  The  auxiliaries' 
camps  were  at  Grinnes  and  Vada,  but  these  places 
cannot  now  be  identified. 

The  river,  however  embarrassing  it  might  be  to 
the  Romans,  was  small  obstacle  to  the  movements 
of  Batavians.  The  fourfold  distribution  of  the 
Roman  army  seemed  to  give  to  Civilis  a  notable 
opportunity  of  a  simultaneous  attack  upon  all  four 
camps.  The  Germans  sallied  out  over  the  Waal, 
and  fell  at  one  and  the  same  time  upon  them  all, 
but  not  with  equal  vigour.  The  legionaries  had 
little  trouble  in  driving  off  the  assailants.  But 
Grinnes  and  Vada  were  attacked  with  great  deter- 
mination, and  it  was  not  until  the  Roman  com- 
mander-in-chief himself  came  to  his  hard-pressed 
auxiliaries'  help  at  the  head  of  a  picked  troop  of 
horse  that  the  Germans  were  forced  here  also  to 
fall  back  again  over  the  river.  By  evening  the 
enemy  had  all  again  crossed  the  Waal,  swimming 
or  by  boats,  and  the  attempted  surprise  had  this 
time  failed. 


1  It  is  abundantly  clear  at  least  that  all  four  camps  were  south  of  the 
Waal.  The  proposed  identifications,  therefore,  of  Arenacum  with  Arnhem, 
and  Batavodurum  with  Wyk-by-Durstede  are  quite  impossible,  as  both 
Arnhem  and  Wyk  are  on  the  Lek.  At  Nymwegen  there  are  to-day  a 
railway  bridge  and  also  a  swing-bridge.  I  have  not  found  on  the  map 
another  bridge  over  the  Waal  between  it  and  Bommel,  thirty  odd  miles  to 
the  west.  If,  however,  Noviomagus  is  identified,  as  is  usually  the  case, 
with  Nymwegen,  Batavodurum  must  be  sought  elsewhere. 


sec.  vii    REBELLION  ON  THE  RHINE      311 

But  the  Romans  could  do  very  little  without 
their  fleet.  Cerialis,  therefore,  left  the  army,  and 
journeyed  up-stream  to  look  for  it  and  to  super- 
intend the  building  of  the  camps  at  Novaesium  and 
Bonn.  The  Roman  historian  gives  a  number  of 
reasons  to  explain  the  fleet's  delay.  The  sailors 
were  afraid  ;  they  were  employed  elsewhere  ;  they 
had  not  been  given  time  enough  in  which  to  arrive. 
One  of  these  three  explanations  would  have  been 
enough.  The  effect  of  the  three  combined  is 
somewhat  ludicrous.  But  Cerialis  did  manage  to 
discover  his  navy,  and  brought  it  back  with  him 
rejoicing.  His  was  a  short-lived  joy.  It  happened 
on  a  night  black  with  clouds  that  Cerialis  and  his 
escort  were  encamped  beside  the  river  on  the 
return  journey.  The  ships,  the  object  of  his  toils, 
lay  moored  in  the  stream  beside  the  camp.  The 
sentries  gazed  sleepily  out  into  the  night.  The 
general  was  not  on  the  admiral's  galley,  where 
his  ensign  flew,  but  dallying,  so  scandal  said,  on 
shore  with  a  native  woman.  A  band  of  Germans 
silently  crossed  the  river  a  short  way  above  the 
camp,  and  stole  down  beneath  the  ramparts.  The 
sentries  noticed  nothing,  and  the  foe  clambered 
quietly  over  the  defences.  In  a  moment  the  sleep- 
ing troops  found  their  tents  falling  upon  their  heads. 
The  Germans  had  cut  the  ropes.  Then  hideous 
and  ferocious  cries  rent  the  stillness  of  the  night, 
as  the  natives  fell  with  joy  to  slaughtering  the 
Romans,  recumbent  and  struggling  beneath  the 
fallen  canvas,  or  emerging  bewildered  and  half- 
naked  into  the  open.     Others  of  the  foe,  coming  in 


312       REBELLION  ON  THE  RHINE     ch.hi 

boats  down-stream,  hurled  grappling-hooks  aboard 
the  ships  and  towed  them  away. 

Their  chief  triumph  was  the  capture  of  the 
general's  ship  and  ensign.  The  squall  had  been  a 
sharp  one,  and  had  broken  with  fury  over  the  hap- 
less Romans.  Then  it  passed  away.  The  Germans 
vanished.  The  Romans  were  left  to  straighten 
their  disordered  camp  and  peg  their  tents  again. 
But  when  morning  broke,  the  angry  and  mortified 
legionaries,  gazing  disconsolately  out  over  the  river, 
saw  their  own  ships,  crowded  with  the  laughing  foe, 
moving  over  to  the  opposite  bank.  Only  the  flag- 
ship was  not  there :  it  was  being  towed  up  the 
Lippe  river,1  yet  another  offering  to  Veleda,  the 
maiden  prophetess.  But  to  the  Germans'  grief, 
they  had  found  no  Cerialis  asleep  on  board  the 
ship. 

Autumn  was  passing  into  winter,  and  no  progress 
had  of  late  been  made.  The  Waal  still  rolled 
between.  Civilis,  exulting  in  his  new-won  ships, 
thought  the  time  come  for  a  naval  display.  In  the 
broad  channel  of  the  Maas,  hard  by  the  modern 
Rotterdam,2  he  gathered  together  all  his  motley 
crowd  of  vessels,  gay  with  every  kind  of  bunting 
and  parti-coloured  sails.  A  favouring  breeze  sent 
them  merrily  up-stream,  until  they  hove  in  view  of 
Cerialis  and  his  astonished  men.  The  Roman  was 
not  to  be  outdone.      He  had  collected  other  craft, 

1  The  Lippe  flows  into  the  Rhine  at  Wesel,  a  few  miles  above  Xanten. 

2  The  review  is  held  in  a  "spatium  velut  aequoris  electum,  quo  Mosae 
fluminis  os  amnem  Rhenum  Oceano  adfundit."  The  broadest  part  of  the 
Maas  would  be  that  where  the  De  Noord  channel  enters  the  Lek  above 
Rotterdam.      From  this  point  to  the  sea  the  Lek  takes  the  name  of  the  Maas. 


sbc. vn    REBELLION  ON  THE  RHINE      313 

and  still  had  the  remnants  of  his  former  fleet  which 
had  escaped  the  grappling-hooks.  His  pilots,  too, 
were  now  experienced  men.  The  Roman  fleet 
therefore  put  out  from  the  shore,  and  drifted 
slowly,  with  sails  furled,  down  the  stream  to  meet 
the  advancing  foe.  Each  gallant  fleet  passed  the 
other,  moving  in  column  of  line  ahead.  As  they 
sailed  by  a  few  missiles  flew  between.  And  there 
ended  the  last  great  naval  engagement  of  the 
war. 

Heavy  rain  had  fallen,  and  the  campaigning 
season  in  those  inclement  barbarian  northern  wilds 
was  fast  drawing  to  a  close.  Even  Civilis  must 
have  found  his  position  in  the  Island  uncomfortably 
damp.  For  he  quietly  evacuated  it,  and  drew  all 
his  men  with  him  over  the  Rhine }  to  the  northern 
bank.  Then  at  last  Cerialis  and  his  men  struggled 
over  the  Waal  without  resistance,  and  the  Island, 
so  long  the  object  of  their  patient  striving,  lay  at 
their  mercy.  Plunder  was  pleasant,  but  of  a  truth 
the  land  was  very  damp.  The  entire  Roman  camp 
bid  fair  to  be  washed  for  good  and  all  away.  And 
now  once  more  their  errant  fleet  had  lost  itself. 
Behind  them  the  swollen  waters  of  the  Waal  rolled 
heavily  seawards.  Their  position  was  uncomfort- 
able. 

So  uncomfortable  in  fact  was  it  that,  in  Tacitus' 
opinion,  a  third  onslaught  by  the  Germans  must 
now  have  put  an  end  once  and  for  all  time  to 
Cerialis  and  his  men,  in  which  case  the  Yorkshire 
wolds  might  have  preserved  their  independence  of 

1  I.e.  the  Lek. 


3i4      REBELLION  ON  THE  RHINE     ch.ih 

Rome  a  few  years  longer.1  But  now,  for  his  part, 
Civilis  had  had  enough  of  war.  For  some  eighteen 
months  he  had  led  an  enjoyably  exciting  life,  pur- 
suing and  pursued,  defeated  and  victorious,  triumph- 
ant and  fugitive,  faring  now  up,  now  down,  the 
valleys  of  the  Moselle  and  the  Rhine.  He  had  won 
fame,  and,  at  least  for  a  brief  time,  liberty  from 
Rome  for  himself  and  his  tribe.  But  by  this  time 
his  own  people  showed  signs  of  restiveness.  The 
Germans,  too,  across  the  Rhine  were  grumbling, 
not  liking  the  thought  of  having  to  support  the 
rebel  army  through  the  winter  months.  Messages 
reached  him  from  Cerialis  offering  life  on  submis- 
sion, but  full  of  threats  if  he  refused  compliance. 
Even  among  Batavians  there  were  traitors.  Civilis 
judged  it  more  profitable  to  surrender  than  to  be 
surrendered  to  the  Romans.  His  faithful  Batavians 
were  quite  capable  of  making  a  scapegoat  of  their 
general.  Civilis  therefore  sent  word  to  Cerialis, 
and  a  meeting  was  arranged  between  the  two  com- 
manders, to  take  place  on  the  river  Nabalia.'2  A 
bridge  over  the  river  was  broken  in  the  middle, 
and  at  the  two  ends  of  the  pieces  left  in  place  the 
two  opponents  stood  to  exchange  speeches  and 
arrange  terms  over  the  gap.  But  what  thereafter 
took  place  remains  unknown  to  us.  For  Civilis 
has    hardly    made    a   fair    beginning    of  what    was 

1  Cerialis  was  Governor  of  Britain  under  Vespasian's  principate  from 
A.i).  71-73.  For  his  successful  expeditions  against  the  Brigantes  cL  Tacitus, 
Agrnola,  17. 

2  This  river  is  nowhere  else  mentioned,  and  Tacitus,  of  course,  gives  no 
clue  which  leads  to  any  certain  identification.  It  may  have  been  the  Lek, 
or  the  Kromme  Rhyn,  or  the  New  (or  Guelders)  Yssel.  The  last  is  the 
favourite  choice.      See  the  plan  on  p.  233. 


sec.  vii    REBELLION  ON  THE  RHINE      315 

doubtless  to  be,  at  least  in  Tacitus,  a  long  excul- 
patory harangue,  than  the  text  of  the  historian  is, 
like  the  bridge,  abruptly  broken  off.  And,  with 
this,  Julius  Civilis,  the  Batavian  prince  and  rebel, 
vanishes  for  ever  from  our  ken.  History  speaks  ot 
him  no  more. 

But  the  war  in  Lower  Germany  was  certainly  at 
an  end.  Of  that  in  Upper  Germany,  with  which 
Annius  Gallus  was  entrusted,  no  record  remains. 
Gallus  had  four  legions  under  his  command,  and 
besides  the  reduction  of  the  Lingones,  of  which  a 
Roman  military  writer  makes  one  passing  mention,1 
the  German  tribes  in  the  lower  Main  valley,  who 
had  raided  up  to  the  walls  of  Mainz,  merited 
chastisement.  If  they  did  receive  this  it  had 
no  very  lasting  effect,  for  the  Emperor  Domitian 
thirteen  years  later  found  it  necessary  to  conduct 
a  serious  campaign  against  the  most  redoubtable 
of  these  very  offenders,  the  Chatti.  But  of  any 
military  operations  on  the  Upper  Rhine  in  a.d.  70, 
after  the  coming  of  Gallus,  there  exists  no  story. 

Rome  treated  the  rebels,  both  Gauls  and 
Germans,  with  politic  mercy.  Only  submission 
was  demanded  of  them,  and  a  return  to  their  old 
condition  of  subjection.  The  Batavians  and 
Cannenefates  still  paid  no  tribute.  No  Roman 
tax-gatherer  was  to  plague  their  soil.  They  must 
continue  to  furnish  troops  to  the  Roman  army  as 
had  been  their  duty  before  the  mutiny.  And  so 
the   German  "  honour "  suffered  no  infringement.2 

1  Frontinus,  Strat.  iv.  3.   14. 
2  "  Manet  honos  et  antiquae  societatis  insigne,"'  Tacitus,  German  in,  29. 


316      REBELLION  ON  THE  RHINE     CH.  tit 

Rome  had  no  desire  for  vengeance  on  the  common 
folk,  in  spite  of  the  losses  which  she  had  suffered  at 
their  hands.  Veleda,  the  fierce  prophetess  maiden, 
was  indeed  captured  later  and  brought  a  prisoner 
to  Rome.  But  it  is  unlikely  that  she  endured  any 
worse  fate  than  to  become  a  subject  for  mediocre 
poets'  verse.1 

But  Rome's  attitude  to  some  at  least  of  the 
tribal  chieftains  was  sterner :  for  it  was  always 
the  ambition  of  such  men,  both  in  Gaul  and 
Germany,  which  was  dangerous  to  the  peace  of 
those  lands  and  to  her  own  supremacy.  Cerealis 
explained  very  clearly  to  the  Treveri  the  reason  of 
Rome's  presence  on  the  Rhine.  "  We  have  not 
planted  ourselves  upon  the  Rhine,"  the  Roman 
historian  makes  him  declare,  "to  guard  Italy,  but 
for  fear  lest  some  second  Ariovistus  should  make 
himself  lord  of  the  kingdom  of  the  Gallic  lands. 
Men  talk  of  liberty,  and  use  other  such  specious 
words.  There  was  never  a  man,  if  he  sought  for 
power  and  dominion  for  himself  at  cost  of  others' 
slavery,  who  did  not  use  such  language.  Should 
we  Romans  ever  be  expelled,  which  may  Heaven 
avert !  what  remains  for  all  the  world  save  a  never- 
ending  war  of  nation  against  nation?"2  It  was 
true,  every  word  of  it.  The  right  of  Rome  to 
control  Gaul  and  "  Germany "  was  to  this  extent 
precisely  the  same  as  our  right  to  govern  India 
and  its  many  peoples.  The  Pax  Romana  had  to 
be  preserved  upon  the  Rhine  lest  bloody  war 
should  in  due  course,   after  long  years  of  savage 

'   Statius,  Silvae,  i.  4.  90.      See  Note  H,  p.  352.  -  Tac.  iv.  73,  74. 


sec.  vii    REBELLION  ON  THE  RHINE       317 

horror,  beget  one  chief  as  Tyrant.  This  reason  had 
made  Julius  Caesar  in  old  days  march  his  trembling 
army  against  King  Ariovistus  the  German. 
Chieftain  after  chieftain  in  Gaul  in  the  Caesarian 
period  between  59  b.c.  and  50  B.C.  had  dreamt 
this  dream  of  lordship  for  himself.  Vercingetorix 
had  all  but  made  of  the  dream  a  waking  vision. 
Peril  of  disturbance  from  such  ambition  in  Gaul  or 
Germany  had  haunted  Tiberius,  Caligula,  Claudius, 
Nero.  In  their  last  sore  straits  Civilis  and 
Classicus  had  even  sent  proffering  Cerialis  the 
"  Imperium  Galliarum "  ; 1  so  strong  and  deep- 
rooted  has  been  the  notion  of  kingship  on  the 
Rhine. 

Therefore,  her  honour  at  last  splendidly  vindi- 
cated, and  the  mutiny  finally  quelled,  Rome  showed 
mercy  to  the  peoples ;  but  some  at  least  of  the 
leaders  of  the  revolt  felt  the  power  of  her  wrath. 
Civilis  almost  certainly  bargained  for  his  life. 
Classicus  and  Tutor  are  never  heard  of  more.  It 
is  idle  to  speculate  upon  their  fate.  But  the  boy 
leader  Valentinus,  who  had  revived  a  dying  cause, 
was  sent  to  Vespasian,  and  by  him  executed. - 
And  most  significant  of  all  was  the  fate  of  Julius 
Sabinus  of  the  Lingones.  After  his  defeat  by  the 
Sequani 3  he  disappeared  entirely  from  the  Roman 
sight.  The  house  to  which  he  had  fled  was  shortly 
afterwards  burnt  to  the  ground,  and  men  commonly 
thought   that  he   had  sought  death  for  himself  in 

1  Unless,  indeed,  this  offer  was  but  a  ruse  to  lull  Cerialis  into  security  at 
Treves  while  the  foe  were  gathering  in  the  hills  for  their  night  attack  on  the 
camp.      Cerialis,  of  course,  sent  no  answer  to  the  offer  (Tac.  iv.  75). 

-  Tac.  iv.  85.  3  See  above,  p.  294. 


318       REBELLION  ON  THE  RHINE     ch.hi 

the  flames.  But  for  nine  long  years,  so  runs  the 
romantic  tale  which  impressed  the  imagination  of 
Greek  and  Roman  writers,  Sabinus  lay  hid  with  his 
faithful  wife  Epponina  in  a  secret  cave.  There  she 
gave  birth  to  two  boys,  and  the  little  lads  grew  up 
with  father  and  mother  in  their  dark  and  gloomy 
cavern.  For  nine  years  they  were  hidden.  After 
nine  years  they  appealed  to  Vespasian  for  pardon. 
After  nine  years  the  Emperor  ordered  their  execu- 
tion in  cold  blood,  though  he  spared  the  boys.  So 
should  we  English  have  treated  the  villain  Nana, 
could  we  but  have  caught  him.  Sabinus'  deeds 
were  no  atrocities  of  a  Nana,  nor  even  of  a  Civilis. 
But  he  had  dared  to  call  himself  Caesar,  and  for 
him  there  remained  no  mercy.1 


§  8.    The  Lessons  of  the  Mutiny 

Gauls,  Germans,  and  Romans,  all  had  learnt 
lessons  taught  them  by  the  great  mutiny. 

A.  The  Results  in  Gaul  and  Germany. — The 
cause  of  the  rebellion  had  been  the  natural  aspira- 
tion for  freedom  felt  by  all  men  worthy  of  the 
name.  This  aspiration  had  been  fed,  in  case  of 
the  Gauls,  by  a  century  of  striving  for  liberty,  by 
memories  both  new  and  old  of  risings  and  struggles 
on  its  behalf;  in  case  of  the  Germans  subject  to 
Rome,  by  more  recent  recollections  of  independ- 
ence, by  the  lively  example  of  their  kinsmen  over 

1  Tac.  iv.  67  ;  Dio,  lxvi.  16  ;  Plutarch,  Amat.  25.  Tacitus  promises  to 
give  the  whole  story  when  he  comes  to  the  year  in  question  (i.e.  A.D.  79) : 
but  this  part  of  his  work,  if  it  was  ever  written,  is  of  course  lost  to  us. 
Plutarch  says  that  he  was  personally  acquainted  with  one  of  the  boys. 


skcviii  REBELLION  ON  THE  RHINE      319 

the  Rhine  who,  once  like  themselves  in  servitude 
to  the  Roman,  had  boldly  struck  for  freedom,  won 
it,  and  retained  it ;  and  by  the  harsh,  careless,  and 
unjust  treatment  which  they  had  endured  at  the 
hands  of  the  military  agents  of  the  Roman  Govern- 
ment. Moreover,  there  was  not  a  race  or  people, 
save  the  effeminate  and  worthless  subjects  of  the 
old  Syrian  monarchy,  who,  conquered  by  Rome, 
had  not  at  least  once  risen  against  their  masters  in 
desperate  rebellion  before  they  had  learnt  to  receive 
and  to  enjoy  the  yet  prouder  and  more  ennobling 
position  of  citizens  of  the  Roman  Empire.  To 
the  Batavians  their  broad  streams  flowing  to 
the  boundless  unknown  ocean,  and  their  pathless 
wastes,  by  themselves  spoke  of  wide  unfettered 
liberty.  They  were  not  men  tamely  to  bow  their 
necks  beneath  the  yoke  of  alien  domination,  if  so  be 
that  they  could  break  it  and  cast  it  off. 

And  never  had  so  fair  an  opportunity  been 
given  to  both  Gauls  and  Germans  to  strike  a  blow 
for  freedom  as  during  these  terrible  months  of  Civil 
War  following  on  the  death  of  Galba,  when  the 
Empire,  which  subdued  them  and,  as  they  said, 
oppressed  them,  seemed  rent  utterly  in  pieces  by 
the  fury  of  contending  selfish  factions.  The  mal- 
contents indeed  had  not  been  men  had  they  let 
their  chance  pass  unheeded. 

But  the  great  effort  miserably  failed.  And  no 
like  opportunity  occurred  again.  Moreover,  the 
course  of  mutiny  had  shown  the  tribes  their  weak- 
ness, while  Rome  at  the  end  was  wisely  merciful, 
and    trampled   neither  on  their  lives  nor  on  their 


320      REBELLION  ON  THE  RHINE 

honour.  The  Imperial  State  remembered  her 
duty  to  her  subjects,  lessons  learnt  by  her  long 
since  in  Italy  and  Spain  and  Macedonia,  and  now 
taught  to  her  again  on  the  Rhine.  The  meanest 
Roman  henceforth  who  did  outrage,  were  it  but  to 
a  child  of  the  conquered  peoples,  was  a  greater  and 
more  despicable  traitor  to  his  country  than  he  who 
risked  his  life  in  mutiny  against  her. 

The  war  had  proved  that  neither  Gauls  nor 
Germans  could  for  long  combine  in  a  national 
war  against  the  Roman.  It  was  easy  to  preach 
eloquently  on  nationalism ;  it  was  impossible  for 
the  nationalists  to  overcome  for  any  length  of  time 
the  local  jealousies  which  proved  the  ruin  of  the 
movement  to  the  end.  Civilis,  who  openly  dis- 
claimed with  patriotic  indignation  any  desire  on 
part  of  Batavians  or  Treveri  to  rule  any  other 
tribe,  is  found  contemplating  a  war  for  this  object 
with  calm  confidence  in  the  prospect  of  a  German 
victory.1  Civilis'  own  nephew  fought  fiercely  to 
the  death  against  him.2  Tencteri  hated  Ubii ; 
Ubii  massacred  Chauci  and  Frisii ;  Sequani  fought 
Lingones.3  The  Gallic  tribes  rose  in  the  "  hope  of 
liberty."  But  the  desire  for  rule  over  their  neigh- 
bours was  a  more  powerful,  if  more  secret,  motive 
with  them.  There  was  to  be  "  One  Gallic  Em- 
pire." This,  doubtless,  was  excellent.  But  whose 
district  should  be  chosen  as  the  seat  for  that 
Empire?4  Indeed  this  was  an  apple  of  discord 
cast  in  upon  the  banquet  of  the  victors  by  Mischief 

1  Cf.  Tac.  iv.  61  with  iv.  66.  2  Julius  Briganticus,  iv.  70;  v.  21. 

3  Tac.  iv.  64,  67,  79.  4  Tac.  iv.  25,  69. 


sec.vui  REBELLION  ON  THE  RHINE      321 

smiling.  And  most  Gauls  thought  that  Rome's 
impartial  if  alien  rule  was  better  after  all.  And 
therefore  the  mutiny  of  the  years  a.d.  69  and  70 
was  the  last  rising  on  behalf  of  independence  which 
affected  Gaul  or  the  German  tribes  on  the  Roman 
side  of  the  Lower  Rhine.  The  page  on  which  is 
written  the  record  of  Gallic  rebellion  against  the 
Roman  conqueror  contains  the  history  of  a  hundred 
and  twenty  years.  But  now  at  last  the  leaf  was 
turned,  and  no  story  of  Gallic  self-sacrifice  or 
treason  on  behalf  of  liberty  embellishes  or  sullies 
the  chapter  any  more.  "  This  was  the  last  blood 
shed  for  the  cause  of  ancient  Gaul,  the  last  act  of 
devotion  to  a  social  order,  a  government,  a  religion, 
the  return  of  which  was  neither  possible  nor 
desirable."  ' 

Whether  a  mutiny  in  the  "native  army"  is 
caused  by  "  outward  and  accidental  causes  "  or  by 
the  "  inner  necessity  of  things"  is  a  question  likely 
to  be  always  debated.  Some  would  still  maintain 
that  there  existed  no  such  inner  necessity  for  the 
sepoy  mutiny  in  India,  and  would  perhaps  still 
point  confidently  to  the  "  apparently  complete 
quiet "  which  has  prevailed  in  the  Indian  army 
since  the  Mutiny  as  proof  of  their  contention. 
In  like  manner  the  German  historian  ascribes 
the  Batavian  mutiny  to  "  outward  and  accidental 
causes,"  and  cites  the  peace  prevailing  after  it  upon 
the  Rhine  as  the  evidence  for  his  contention.  Yet 
surely  it  was  no  mock  plea  for  liberty  which  called 

1  Thierry,   translated    and    quoted   by  Merivale,  History   of  the  Romans 
under  the  Empire,  vi.  p.  527. 

Y 


322      REBELLION  ON  THE  RHINE     ch.hi 

so  many  German  tribes  and  peoples  round  Civilis' 
standard,  and  made  them  faithful  to  their  leader 
well-nigh  to  the  very  last.  We  honour  the  bar- 
barian and  the  mutineer  too  greatly  if  we  ascribe 
to  him  the  pure  feeling  of  passionate  devotion  to 
an  ideal  such  as  sent  thousands  of  Italy's  sons 
joyfully  to  the  dungeon,  to  the  gibbet,  and  to  death 
on  the  field,  sixty  years  ago.  Liberty  to  the 
ancient  German  may  have  spelt  little  save  revelling 
in  lust,  in  plunder,  and  in  butchery.  Therefore 
civilised  man  applauds  the  victory  of  Rome,  of 
peace,  order,  government,  and  law.  The  barbarians' 
temper  was  wild  and  passionate  ;  their  deeds  were 
treacherous  and  foul ;  their  cruelty  was  savage. 
Yet  the  seed  which,  when  planted  in  so  rude  a  soil, 
sprang  up  a  rank  and  poisonous  growth,  in  kinder 
and  more  congenial  climes  has  borne  the  noblest 
fruit  which  ever  glorifies  Man  and  marks  him  apart 
from  the  brute.  Not  liberty,  but  the  use  men 
make  of  it,  is  its  sole  justification.  But  the  fierce 
Batavian  mutineer,  the  veriest  German  savage, 
who  dreamt  perhaps  vague  dreams  of  freedom, 
deserved  indeed  no  victory,  yet  his  merited  failure 
was  not  utterly  barren  of  honour.  In  the  history 
of  races  as  well  as  of  individuals  the  child  is  father 
of  the  man. 

But  the  German  children  had  learnt  their  lesson. 
They  were  quick  at  least  to  see  that  petulance 
brought  punishment,  and  their  manhood  was  not 
yet.  Though  we  may  venture  to  think  their  rising 
due  to  other  causes  besides  those  of  the  recruiting 
officer  and  the  happiest  of  opportunities,  yet  for  its 


sec.  v..,  REBELLION  ON  THE  RHINE      323 

issue  it  is  enough  to  cite  the  same  historian's 
words  :  "  The  Roman  Germans  were  merged  in  the 
Empire  no  less  completely  than  the  Roman  Gauls ; 
of  attempts  at  insurrection  on  the  part  of  the  former 
there  is  no  further  mention.  At  the  close  of  the 
third  century  the  Franks,  invading  Gaul  by  way  of 
the  Lower  Rhine,  included  in  their  seizure  the 
Batavian  territory.  Yet  the  Batavians  maintained 
themselves  in  their  old,  though  diminished,  settle- 
ments, as  did  likewise  the  Frisians,  even  during 
the  confusions  of  the  great  migrations  of  peoples, 
and  so  far  as  we  know,  preserved  allegiance  even 
to  the  decaying  Empire  as  a  whole."  : 

B.  The  Results  hi  the  Ro?nan  Army.  —  The 
Roman  Government  showed  that  it  too  had  learnt 
lessons  from  the  mutiny  in  its  treatment  of  both 
the  legions  and  the  auxiliary  forces  on  the  Rhine. 

(1)  The  Legions. — The  outbreak  of  the  mutiny 
had  revealed  serious  defects  in  the  prevailing  system 
of  "clan-regiments,"  taken  from  the  native  tribes 
to  serve  as  auxiliaries  in  the  Roman  army,  when  at 
least  these  were  stationed  in  the  country  of  their 
birth.  The  rapidity  of  the  spread  of  the  movement, 
the  feeble  resistance  to  it  offered  by,  and  even 
the  mutinous  tendencies  shown  in,  the  regular 
legionary  regiments  along  the  whole  course  of  the 
river,  made  manifest  that  some  defect  of  organisa- 
tion existed  also  in  this,  the  more  important,  branch 
of  the  Roman  Imperial  military  system. 

It  seemed  evident  that  the  legions  on  the  Rhine 
in  a.d.  69  were   tainted  with  native  German  sym- 

1  Mommsen. 


324      REBELLION  ON  THE  RHINE     ch.im 

pathies.  This  pointed  to  the  fact  that  the  recruit- 
ing system  was  to  blame.  The  Emperor  Augustus 
had  sought  to  establish  the  general  practice  that 
recruits  for  legions  serving  in  the  western  part  of 
the  Empire  should  be  drawn  from  the  eastern 
provinces,  and  that  legions  on  duty  in  the  latter 
should  be  recruited  from  the  west.  The  principle 
was  the  same  as  that  in  the  modern  Kingdom 
of  Italy,  where  Lombard  and  Tuscan  regiments 
tend  to  be  quartered  in  the  south,  Sicilian  and 
Calabrese  in  the  north  of  the  peninsula.  By 
such  means  the  army  itself  becomes  a  means  of 
promoting  unity  and  unification.  Moreover,  in 
the  event  of  a  disturbance,  the  troops  on  the  spot 
are  more  likely  to  be  utterly  true  to  their  military 
discipline  because  they  do  not  share  in,  or  perhaps 
even  realise,  local  feelings  and  aspirations.  But 
to  combine  this  sound  principle  of  recruiting  with 
the  equally  wise  system  of  permanent  military 
camps  upon  the  frontiers  of  the  Empire  was 
proving  a  very  hard  task  for  the  Government. 
These  camps,  again  devised  by  the  extraordinary 
foresight  of  Augustus,  were  invaluable  to  the  peace 
of  the  Empire,  the  popularity  of  the  army,  the 
prosperity  of  the  provinces,  and  the  Romanisation 
of  the  outlying  districts  of  the  Empire  and  the 
tribes  beyond  its  limits.  But  as  time  went  on, 
and  the  children  of  the  legions  grew  to  manhood, 
the  regiment,  stationed  for  years  together  at  the 
same  frontier  camp,  could  not  but  gather  its 
recruits  from  the  sons  of  its  soldiers,  many  of  whom 
married   the  women   of  the  district,  and  from   the 


skc.vii.  REBELLION  ON  THE  RHINE      325 

native  auxiliaries  serving  side  by  side  with  them. 
The  result  was  that  Augustus'  principle  of  recruiting 
was  not  to  be  reconciled  with,  and  had  to  yield  to, 
his  system  of  permanent  cantonments.  And  hence 
a  legion  and  a  locality  became  identified  so  closely 
that  the  interests  and  hopes  of  the  latter  became 
those  of  the  former.  To  the  Roman  Empire  the 
introduction  of  the  territorial  system  into  the  army 
had  grave  disadvantages. 

The  six  legions  engaged  at  first  in  the  German 
rising  of  a.d.  69  were  the  First,  Fourth,  Sixteenth, 
and  Twenty-second,  whose  men  proved  mutinous 
and  treacherous,  the  Fifth  Alaudae  and  Fifteenth 
Primigenia,  some  of  whose  men  at  least  fought 
most  gallantly  and  died  for  Rome.  Of  these  six 
legions  it  appears  that  the  First  had  been  in  Lower 
Germany,  the  Sixteenth  in  Upper  Germany,  since 
the  days  of  Augustus  ;  the  Fourth  and  Twenty- 
second  since  a.d.  43.  Thus  the  leaven  of  local 
sympathy  had  had  time  to  work  with  these  men. 
It  is  true  that  the  Fifth  legion  Alaudae  had  also 
been  in  Lower  Germany  since  Augustus'  time, 
save  for  a  passing  excursion  to  Britain  under 
Claudius.  But  this  regiment  had  great  traditions 
of  bitter  fighting  with  the  Germans,  in  former  days 
under  Lollius,  and  recently  under  Corbulo.  The 
small  heroic  vexillum  left  behind  at  Vetera  when 
the  bulk  of  the  legion  marched  for  the  Vitellian 
cause  to  Italy  was,  therefore,  proud  to  preserve  its 
regimental  tradition.  Its  comrades  of  the  Fifteenth 
legion  were  probably  swayed  by  the  example  of 
the   men   of  the   Fifth  who  were  in  garrison  with 


326      REBELLION  ON  THE  RHINE     ch.hi 

them.  It  is  possible,  too,  that  the  Fifteenth  legion 
was  but  newly  raised,  and  had  been  but  a  few  years 
in  Germany,  so  that  its  loyalty  in  this  case  would 
be  a  striking  illustration  (from  opposites)  of  the 
thesis  advanced  above  that  the  mutinous  tendencies 
might  be  produced  by  a  long  stay  in  Germany.  If, 
however,  this  legion  also  had  been  in  the  country 
since  a.d.  43,  as  some  suppose,  the  loyalty  of  their 
comrades  must  be  held  responsible  for  the  bravery 
of  its  own  soldiers.1 

When  the  revolt  was  ended,  and  the  new 
Emperor,  Vespasian,  dealt  with  the  question  of 
the  garrison  of  the  Rhine,  he  made  sweeping 
changes  in  its  composition.  So  shrewd  a  soldier 
as  was  this  sagacious  Prince  might  approve  the 
pardon  granted  by  his  general  to  the  mutineers 
who  repented,  but  their  regiments  had  stained  their 
reputation  beyond  forgiveness.  Vespasian  promptly 
struck  three  of  the  "  traitor  legions  "  from  the  roll 
of  the  army.  The  First,  the  Fourth  Macedonica, 
and  Sixteenth  Gallica  ceased  henceforth  to  exist. 
The  place  of  the  last  two  in  the  army  list  was 
taken  by  two  new  legions,  the  Fourth  Flavia 
firma,  and  the  Sixteenth  Flavia  felix.  In  the  year 
a.d.    82    a    third    new   legion   was  also  added,   the 

1  Legio  XV.  Primigenia  seems  to  make  its  first  appearance  in  Tacitus, 
Histories,  i.  55,  a.d.  69.  Pfitzner  thinks  it  was  created  in  a.d.  62  ; 
Grotefend,  that  both  XV.  Primigenia  and  XXII.  Primigenia  were  created 
as  separate  from  the  two  legions,  XV.  Apollinaris  and  XXII.  Ueiotariana, 
on  the  occasion  of  the  invasion  of  Britain  in  a.d.  43.  The  name  Primigenia 
implies  "first  existing";  i.e.  when  a  legion  was  duplicated,  the  part  which 
retained  the  old  eagle  was  Primigenia  ;  the  other  legion  of  the  same  number 
received  a  new  eagle  and  retained  the  old  distinctive  title.  But  there  are 
other  explanations  of  the  title,  for  which  see  Pfitzner,  Geschichte  der  rom. 
Kaiser-legionen,  p.  8. 


sFx.vm  REBELLION  ON  THE  RHINE      327 

First  Flavia  Minervia  pia  fidelis,  and  was  encamped 
at  Bonn,  where  it  still  lay  as  late  as  the  year 
a.d.  295. 1  Only  the  Twenty-second  legion  of  the 
four  disgraced  regiments  was  spared,  and  kept  still  in 
the  country  in  camp  at  Vetera  from  a.d.  71  to  a.d.  90.2 
Vespasian's  reasons  for  such  a  difference  of  treat- 
ment between  this  regiment  and  its  three  partners 
in  dishonour  are  not  preserved  to  us. 

Some  uncertainty  attaches  to  the  fate  of  the 
two  loyal  legions,  the  Fifth  and  the  Fifteenth.  It 
is  quite  possible  that  these  were  also  both  dis- 
banded.3 If  so,  Vespasian  had  evidently  deter- 
mined to  make  an  almost  entirely  clean  sweep  of 
the  regiments  of  the  former  garrison  of  the  Rhine. 
It  is,  however,  a  greater  pleasure  to  suppose  that 
the  proud  and  valiant  Fifth  did  continue  to  exist, 
and  sealed  its  long  services  to  the  Empire  on  the 
dire  field  of  battle  in  Moesia  against  the  Sarmatians 
about  a.d.  92,  perishing  there  to  the  last  man.4 

One  fact  at  least  is  certain — that  the  garrison 
of  the  Rhine  after  a.d.  71  was,  with  the  two  excep- 
tions of  the  Twenty-first  and  Twenty-second  legions, 

1  Cf.  Dio,  lv.  24,  3  ;  Ruggiero,  Dizionario  Epigrafico,  ii.  p.  513; 
Ritterling,  W.-D.  Zeitschrift,  xii.  3,  p.  234. 

2  Tiles  and  tombstones  of  the  legionaries  of  XXII.  are  found  at  Xanten  and 
Nymwegen,  without  the  addition  of  pia  fidelis  to  the  name  of  the  legion  which 
it  subsequently  gained.  Cf.  Brambach,  C.I.R.  ;  Weichert,  W.-D.  Zeitschrift, 
1902  ;  Ritterling,  ibid.  xii.  3,  p.  230;  Diz.  Epig.  ii.  p.  514  ;  Hardy,  Studies, 
p.  2IO  sqq. 

3  This  is  Mommsen's  view.  For  a  discussion  of  the  question  cf.  Hardy, 
Studies,  p.  213,  note. 

4  The  legion  destroyed  by  the  Sarmatians  (cf.  Suetonius,  Domitian,  6) 
has  been  supposed  to  be  either  V.  Alaudae  or  XXI.  Rapax.  So  Riese  in 
the  Diz.  Epigrafico,  ii.  p.  514,  believes  it  to  have  been  the  latter.  Ritter- 
ling (IV.-D.  Zeitschrift,  xii.  3,  p.  234)  believes  that  V.  Alaudae  did  continue 
to  exist,  but  was  destroyed  under  Domitian  by  the  Dacians,  while  XXI. 
endured  this  fate  at  the  hands  of  the  Sarmatians  a  few  years  later. 


328      REBELLION  ON  THE  RHINE     ch.hi 

composed  of  entirely  different  regiments  from  those 
which  had  hitherto  been  encamped  on  the  river.1 
In  both  Lower  and  Upper  Germany  there  were 
again  four  legions.      In  the   former  district    Legio 

X.  Gemina  was  stationed  at  Noviomagus  (possibly 
Nymwegen2);  XXII.  Primigenia  at  Vetera  (by 
Xanten) ;  VI.  Victrix  at  Novaesium,  till  about 
a.d.  105,  when  it  was  moved  to  Vetera  to  take  the 
place  of  the  Twenty-second;  and  XXI.  Rapax  at 
Bonn  up  to  a.d.  82,  when  it  was  moved  to  Mainz, 
and  its  place  at  Bonn  was  taken  by  the  new  legion, 
I.  Flavia  Minervia.  This  allotment  of  the  legions 
shows  that  Vespasian  thought  it  desirable  to  keep 
one  legion  in  immediate  touch  with  the  Batavian 
land  (at  Nymwegen  ?),  whereas  hitherto  there  had 
been  no  legionary  camp  on  the  river  lower  down 
than  Vetera.  In  Upper  Germany,  Legions  I. 
Adjutrix  and  XIV.  Gemina  were  encamped  together 
at  Mainz,  the  defences  there  being  strengthened  ; 

XI.  Claudia  was  stationed  at  Vindonissa ;  and, 
in  between  the  two  places,  VIII.  Augusta  was 
stationed  at  Strassburg  (Argentoratum),  a  camp 
which  had  not  been  occupied  since  its  legion,  the 
Second  Augusta,  had  been  sent  to  Britain  in,  or 
perhaps  before,  a.d.  43. 3 

In    the    system    of    recruiting    for    the    legions 
Vespasian  is  not  known  to  have  made  any  change. 

1  But  Legio  XXL,  usually  in  garrison  at  Vindonissa,  had  been  absent  at 
the  time  of  the  mutiny. 

2  See  note  above,  p.  310. 

3  For  these  legionary  arrangements  in  Germany,  cf.  Riese,  ap.  Kuggiero, 
Diz.  Epigrafuo,  sub  voc.  "  Germania,"  vol.  ii.  pp.  513,  514.  One  recently 
found  tile  of  Legio  II.  at  Strassburg  is  said  to  be  earlier  than  a.d.  43 
(IVestd.  Zeit.  1905,  p.  330). 


sKc.vm  REBELLION  ON  THE  RHINE      329 

No  change,  in  fact,  seemed  possible.  Neither  were 
the  legions  shifted  at  short  intervals  from  camp 
to  camp,  or  summoned  away  save  when  urgent 
wars  elsewhere  made  demands  on  the  Rhine  army. 
For  the  loyalty  of  that  army  Vespasian  must  have 
relied  on  the  change  of  regiments  for  the  immediate 
present,  and,  for  the  future,  on  the  absence  of  such 
local  discontent  as  had  excited  the  sympathy  of  the 
former  Vitellian  army.  He  was  not  disappointed 
in  his  trust,  and  the  legions  on  the  Rhine  gave  him 
no  cause  of  concern.  Thus,  after  all,  Augustus' 
system  of  permanent  camps  was  justified. 

(2)  The  Auxilia. — The  practice  of  using  clan- 
regiments  of  auxiliaries  in  their  native  country  had 
proved  disastrous  during  the  mutiny.  Doubtless  it 
had  had  the  effect  of  popularising  this  branch  of 
the  service  and  of  making  recruits  easy  to  obtain. 
In  the  case  also  of  war  in  the  adjoining  districts 
such  regiments  might  be  expected  to  be  well 
acquainted  with  the  enemy's  methods  of  fighting 
and  the  ground,  and  could  obtain  information  and 
supplies  far  more  easily  from  the  country  than  could 
auxiliaries  who  were  aliens  and  strange  to  the  land. 
As  the  Roman  army  relied  upon  its  auxiliaries  for 
the  all-important  duties  of  reconnaissance  and  scout- 
ing, the  advantages  of  using  clan-regiments  in  their 
own  country  were  indeed  very  great,  quite  apart 
from  the  great  saving  of  expense  of  transport  and 
maintenance  which  this  system  secured,  and  which 
counted  not  a  little  in  the  careful  financial  organisa- 
tion of  the  wiser  and  more  thrifty  of  the  early 
Emperors.     There  was,  therefore,  every  reason  for 


33©      REBELLION  ON  THE  RHINE     ch.hi 

the  choice  of  this  system  by  Augustus  and  his 
immediate  successors.  But  the  mutiny  had  opened 
the  eyes  of  the  Roman  Government  to  its  risks. 
By  the  Flavian  Emperors  its  many  advantages  were 
counted  as  nothing  compared  to  its  dangers,  at  least' 
upon  the  Rhine.  No  attempt  was  made  to  abolish 
clan-regiments.  This  indeed  would  have  been  far 
too  sweeping  a  measure,  and  might  have  destroyed 
the  auxiliary  system  completely.  But,  with  very 
few  exceptions,  such  regiments,  both  infantry  and 
cavalry,  are  found  serving  in  countries  other  than 
those  of  their  origin. 

Thus  the  indigenous  cohorts  and  alae,  which 
before  a.d.  70  had  served  on  the  Rhine,  after  that 
date  were  either  disbanded  or  sent  far  afield  to 
Britain,  Raetia,  Pannonia,  Dacia,  Moesia,  even 
Mauretania.1  In  Lower  Germany  neither  Germans 
nor  Gauls  seem  to  have  been  employed  as  troops 
under  the  Flavians.  In  Upper  Germany  there  is  still 
found  under  Vespasian  and  Domitian  a  squadron  of 
Cannenefates,  significantly  removed  from  their  own 
district  at  the  mouth  of  the  Rhine,  and  solitary 
cohorts  of  "  Gauls,"  "  Germans,"  and  Bituriges. 
But  the  great  majority  of  auxiliary  troops  in  both 
provinces  come  from  other  countries.  Cavalry 
squadrons  of  men  from  Africa,  Ituraea,  Moesia, 
Noricum,  Thrace,  serve  in  the  Lower  province. 
Cohorts,    either    infantry    or    mixed    infantry    and 

'  E.g.  in  Britain  are  found  Baetasii,  Batavi,  Cugerni,  Frisii,  Lingones, 
Menapii,  Morini,  Suebi,  Tungri,  Vangiones  ;  in  Pannonia,  Batavi,  Cannene- 
fates, Helvetii ;  in  Raetia  (from  a.d.  103)  and  Dacia,  Batavi ;  in  Moesia, 
Mattiaci,  Ubii,  Tungri ;  in  Mauretania,  Sugambri.  (List  from  Riese,  cf.  next 
note. ) 


sEc.vni  REBELLION  ON  THE  RHINE      331 

cavalry,  of  Dalmatians,  Spaniards,  Lusitanians, 
Vindelicians,  are  found  in  the  Lower  province  ;  and 
in  the  Upper,  of  Aquitanians,  Asturians,  Dalmatians 
(at  Wiesbaden  and  Bingen),  Ituraeans  and  Damas- 
cenes from  the  East,  Raeti  (at  Wiesbaden  and 
Vindonissa),  Pannonians  (at  Bingen),  men  of  Cyrene 
(at  Neuenheim),  Thracians,  and  Vindelicians.  The 
mere  list  shows  how  complete  was  Vespasian's 
reversal  of  the  former  practice,  how  utterly  different 
was  the  army,  as  well  of  auxiliaries  as  of  legionaries, 
which  garrisoned  the  Rhine  in  his  own  and  his  sons' 
days  from  that  which  was  submerged  by  the  great 
flood  of  the  mutiny  or  helped  to  swell  its  volume.1 
Auxiliaries  as  well  as  legionaries  are  henceforward 
loyal  to  the  Empire.  "  Vespasian  was  a  soldier  of 
sagacity  and  experience  ;  it  is  probably  in  good  part 
a  merit  of  his  if  we  meet  with  no  later  example  of 
revolt  of  the  auxilia  against  their  legions."  2 

The  garrison  of  the  Rhine  of  men  of  all  arms  in 

1  These  details  concerning  the  auxilia  I  take  from  Alex.  Riese's  valuable 
article  "  Germania  "  in  the  new  Dizionario  Epigrafico  of  Ruggiero,  published 
last  year.  The  whole  subject  of  the  auxiliary  troops  of  the  Imperial  military 
system  is  as  yet  in  a  most  dishevelled  condition,  and  no  good  and  complete 
treatment  of  it  as  a  whole  has  yet,  to  my  knowledge,  been  published.  See 
Note  H,  "The  Flavian  Army  of  the  Lower  Rhine." 

2  Mommsen,  Provinces,  Eng.  Trans,  vol.  i.  p.  144,  who  also  comments  on 
the  disappearance  from  the  auxilia  after  this  date  of  native  officers,  such  as 
Arminius,  Civilis,  Classicus.  But  the  clan-regiments  surely  continued  to 
exist,  though  Mommsen  seems  to  doubt  this.  "The  men  serve,  without 
distinction  as  to  their  descent,  in  the  most  various  divisions."  This,  of  course, 
is  true  of  such  corps  as  the  First  and  Second  ala  Flavia  Gemina,  the  ala 
Singularium,  the  numerous  cohortes  voluntariorum  civium  Romanorum,  etc., 
found  in  Flavian  times  on  the  Rhine.  And  the  "special"  corps,  e.g.  Cohors 
Sagittariorum,  continue  to  be  raised.  But  the  great  majority  of  regiments  in 
Riese's  long  list  bear  at  least  clan  names,  and  presumably  continued  to  be 
composed,  at  least  largely  (?  entirely),  of  the  natives  of  those  tribes  or  countries 
whose  names  they  bear. 


332       REBELLION  ON  THE  RHINE 

Flavian  times  has  been  reckoned  at  some  sixty-nine 
thousand  men,  of  whom  thirty -four  thousand  be- 
longed to  the  Upper,  thirty-five  thousand  to  the 
Lower,  German  province.1  But  as  time  went  by, 
and  men's  memories  of  the  mutiny  grew  dim,  the 
Lower  province,  the  scene  once  of  its  greatest  fury 
and  carnage,  was  found  so  peaceful  that  part  of  its 
troops,  urgently  needed  by  wars  elsewhere,  could 
with  safety  be  withdrawn.  Vespasian's  annexation 
of  the  district  of  the  Agri  Decumates  and  the  valley 
of  the  Neckar  in  a.d.  73-74,  Domitian's  warring 
with  the  Chatti  some  ten  years  later,  employed  the 
troops,  without  disturbing  the  peace,  of  the  Upper 
province.  And  soon  the  storm-clouds  came  rolling 
up  black  and  threatening  ruin  to  the  Roman  Empire 
upon  the  Danube  frontier.  Under  Marcus  Aurelius 
two  legions  only  were  left  in  garrison  upon  the 
Lower  Rhine.  The  tide  of  war  had  swept  steadily 
eastwards,  carrying  with  it  the  line  of  Roman  forti- 
fications on  the  Upper  Rhine  and  the  military  camps 
upon  the  Danube.  The  Teuton  was  yet  to  be  at 
death-grips  with  the  Roman.  But  the  desperate 
struggle  was  to  be  waged  upon  both  banks  of  the 
greater  river  and  with  new  invading  tribes.  Few 
storms  of  war  disturbed  the  calm  surface  of  the  Rhine 
after  a.d.  70,  and  those  were  of  brief  duration.  And 
the  German  tribes  upon  the  Roman  bank  joined  with 
Rome's  troops  upon  the  river  in  accepting  loyally 
and  placidly  her  sway.  The  great  mutiny  left  no 
heritage  of  ill  -  will  behind  it  to  any  generation. 
Rome  was   always   truly  victor   because   she   knew 

1   Ritterling    W.-D.  Zeitschrift,  xii.  3.  p.  242. 


sbc. vni  REBELLION  ON  THE  RHINE      3t>3 

how  to  use  victory  well.  Her  citizens  shirked  no 
military  duty  for  pleasure  or  for  any  folly  of 
humanitarian  sentiment  which,  if  indulged  in, 
defeats  its  own  ends.  In  the  strength  as  in  the 
valour  of  her  "  National  army "  she  defied  her 
enemies.  Where  she  conquered  she  civilised. 
To  those  whom  she  defeated  she  taught  the  use 
of  arms  on  her  own  behalf,  as  well  as  order  and 
law.  Her  very  rebels  and  subject  races  learnt  the 
patriotism  of  Romans,  a  patriotism  of  self-sacrifice 
and  deeds,  not  of  boasts  and  empty  words.  And 
therefore,  still  in  the  days  of  her  Emperors  as  in 
those  earlier  days  when  the  citizen-soldier,  trained 
from  boyhood  to  the  use  of  arms,  crushed  his  Mace- 
donian or  Carthaginian  enemy,  and  hurled  the 
Asiatic  back  behind  the  barriers  of  Taurus  and 
Euphrates,  Rome  was  an  Imperial  State.  Still  her 
patriotism  was  no  mock  patriotism,  loud-tongued, 
afraid  of  burdens.  Still  the  rock  of  her  strength, 
though  fiercely  assaulted  by  the  jealous  hatred  of 
her  enemies,  stood  firm,  because  it  was  not  yet 
undermined  by  cowardice  and  pleasure-seeking  on 
the  part  of  her  citizens. 


POSTSCRIPT 

During  the  time  of  the  writing  of  this  book  I 
chanced  to  have  been  reading  again  part  of  the  story 
of  the  making  of  modern  Italy,  that  great  epic  of  the 
nineteenth  century.  The  contrast  between  the  two 
periods  of  war  in  Italy  and  struggle  with  the  German 
ene?ny,  that  of  my  writing  and  that  of  my  reading, 
could  not  fail  to  present  itself  vividly. 

Ancient  Rome  won  the  unity  of  Italy,  and  then,  in 
due  course,  her  Empire,  by  the  unflinching  heroism 
and  pure  devotion  to  their  country  of  her  sons.  Then 
it  came  to  pass  that  greed  and  selfishness,  ambition 
and  passion,  triumphed  over  patriotism,  self-sacrifice, 
si7nphcity.  In  this  book  we  have  seen  rival  Italian 
leaders  contending  in  furious  struggle  for  the  personal 
mastery.  And  all  the  while  Italy  lay  unheeded,  sorely 
wounded.  Her  life-blood  was  draining  away ;  her 
sons  slew  one  another  remorselessly  ;  while  the  da?igcr 
from  the  northern  barbarian  gathered  ever  more 
gloomily  upon  her  frontiers. 

Now  it  is  scarcely  a  generation  since  men  have 
seen  Italy  won  at  last  again  to  unity  by  the  bravery 
and  the  endurance  of  her  children.     Mere  boys  and 

334 


POSTSCRIPT  335 

youths  in  the  pride  of  their  strength  faced  the  cannon 
and  the  executioner  with  smiles  on  their  lips  ;  women 
endured  all  agony  of  pain  and  loss ;  men  battled 
forward  to  victory  in  spite  of  peril,  failure,  and 
disaster.  So  unselfishness  and  patriotism  won  here 
their  most  renowned  victory  of  modem  times. 

Hurtful  indeed  and  well-nigh  ruinous  to  Italy  ivas 
the  "year  of  the  four  Emperors,"  when  men  fought 
for  the  sake  of  greed,  or,  more  nobly,  as  in  the  case  of 
our  own  Wars  of  the  Roses,  for  personal  devotion  to 
so?ne  leader,  but  not  for  love  of  country.  For  any 
cause  other  than  the  highest  a  man,  it  might  be  thought, 
would  not  willingly  die.  Yet  these  men  in  Italy  of 
olden  days  did  face  death  cheerfully  for  causes  lower, 
and  many  of  the?n  base  enough.  And  this  is  a  glory, 
albeit  a  lesser  glory,  of  Roman  manhood. 

Salve,  magna  parens  frugtim,  Saturnia  tellus, 
magna  virum. 

For  there  are  diseases  of  the  body  politic  which 
cost  a  nation  the  loss  of  strength  and  manhood,  and 
these  are  more  injurious  than  is  Civil  War.  Such 
were  in  due  course  to  inflict  upon  Italy  yet  greater 
miseries  than  did  even  the  masterful  strivings  of  the 
rivals  for  Empire,  before  the  time  of  her  redemption 
came  at  last.  Greater  perils  to  a  land  even  than 
armed  ambition  and  cruelty  are  that  craven  self- 
regarding  sloth  and  that  veritable  diruTr)  rod  irXovrov, 
deluding  rich  and  poor  alike  by  its  enchantment, 
which,  however  fair- seeming  may  be  the  titles  of  peace 


336  POSTSCRIPT 

and  humanity  under  which  they  seek  to  disguise 
themselves — and  what  nobler  names  than  these  could 
ever  be  so  misused? — would  yet  surrender  the  country 
indolent,  poorly -armed,  tim'eady,  to  the  sudden  on- 
slaught of  a  jealous  and  a  vigorous  foe. 

"  Does  the  red  stand  for  rose-leaves  on  our  flag  ?  " 


NOTES 

A. — The  Movements  of  the  Danube  Legions  (p.  79) 

Tacitus  (ii.  11)  says:  "  Laeta  interim  Othoni  principia  belli, 
motis  ad  imperium  eius  e  Dalmatia  Pannoniaque  exercitibus.  Fuere 
quattuor  legiones,  e  quibus  bina  milia  praemissa ;  ipsae  modicis 
intervallis  sequebantur."  Again,  in  his  enumeration  of  Otho's  troops 
engaged  in  the  battle  of  Locus  Castorum  (ii.  24)  he  writes  :  "Tertiae 
decumae  legionis  vexillum,  quattuor  auxiliorum  cohortes  et  quingenti 
equites  in  sinistra  locantur  ;  aggerem  viae  tres  praetoriae  cohortes 
altis  ordinibus  obtinuere  ;  dextra  fronte  prima  legio  incessit  cum 
duabus  auxiliaribus  [MS.  vexillaribus]  cohortibus  et  quingentis 
equitibus." 

The  "  Thirteenth  "  legion  was  one  of  the  two  in  Pannonia  ;  the 
"First"  was  the  legion  from  Rome  belonging  to  the  Army  of  Italy. 

It  has  been  argued  {e.g.  by  Pfitzner  in  his  Geschichte  der 
Kaiserlegionen)  that  by  the  time  of  this  battle  all  four  vexilla  from 
the  Pannonian  and  Dalmatian  legions  had  arrived  on  the  scene  and 
not  the  one  only  of  the  Thirteenth  legion.  Hence  for  the  unmeaning 
MS.  reading  "vexillaribus"  in  ii.  24  he  would  substitute  "  vexillariis  " 
and  take  these  two  "  vexillariae  cohortes  "  to  be  two  of  the  remaining 
three. 

This  view  is  to  be  rejected.  It  is  clear,  as  Gerstenecker  {pp.  cit. 
pp.  20-22,  and  note  47,  p.  64)  points  out,  that  the  vexillum  of  a  legion 
cannot  be  properly  described  as  a  vexillaria  cohors.  The  right  form 
of  description  is  already  given  in  ii.  24  in  the  phrase  "tertiae 
decumae  legionis  vexillum  "  ;  i.e.  the  legion  to  which  such  a  vexillum 
belonged  is  named.  Moreover  Tacitus,  after  his  enumeration  of  the 
Othonian  forces  present  at  the  battle,  says  expressly  that  there  was 
no  reserve,  "  nullum  retro  subsidium "  (ii.  26).  But  he  has  not 
mentioned   a   fourth  vexillum   at  all,  supposing   the   two  vexillariae 

337  Z 


338  NOTES 


cohortes  were  numbers  2  and  3.  The  proper  inference  is  that  there 
was  no  fourth  vexillum  present.  Hence  in  ii.  24  we  should  accept 
the  proposed  "  auxiliaribus "  for  the  MS.  "  vexillaribus,"  and  take 
these  two  cohorts  to  belong  to  the  auxiliaries,  not  to  be  two  of  the 
vexilla  of  the  Danube  legions.  In  this  case  only  one  of  the  four 
vexilla  has  arrived  on  the  scene  by  April  6,  and  hence  my  statements 
in  the  text  are  based  on  this  conclusion. 


B. —  The  Capture  of  Cremona  by  the  Vitellians  (p.  81) 

Plutarch  {Otho,  c.  7)  believes  that  Cremona  remained  in 
Othonian  hands  until  after  Caecina's  vain  attack  on  Placentia  ;  that 
Caecina  then  marched  on  the  town  to  take  it  ;  that  Gallus  marched 
thereupon  to  defend  its  garrison  ;  and  that  the  battle  of  Locus 
Castorum  was  the  result. 

This  is  certainly  not  Tacitus'  view.  According  to  him  Cremona 
must  have  been  occupied  by  the  advance  guard  which  Caecina  sent 
/  over  the  Alps.  It  is  certainly  included  in  the  phrase  "  florentissimum 
Italiae  latus,  quantum  inter  Padum  Alpesque  camporum  et  urbium, 
armis  Vitellii  tenebatur"  (ii.  17),  for  it  was  "at  the  same  time" 
(iisdem  diebus)  as  the  assault  on  Placentia  that  the  Vitellian 
auxiliaries,  fleeing  before  the  foray  of  Macer's  gladiators  over  the 
""fijiver,  took  refuge  in  the  city  (ii.  23).  When  then  Caecina,  on  his 
jjtmarch  from  Placentia,  "Cremonam  petere  intendit "  (ii.  22),  the 
words  are  not  to  be  taken  in  the  sense  of  a  hostile  movement. 
Probably  Cremona  was  taken  at  the  same  time  as  the  "  Cohors 
Pannoniorum  apud  Cremonam,"  which  may  have  been  its  garrison. 
Gerstenecker  adopts  Plutarch's  view  and  supports  it  by  a  truly 
quaint  argument,  befitting  an  arm-chair  student  rather  than  any  one 
with  a  knowledge  of  geography  and  military  history.  He  maintains 
that  had  Cremona  been  captured  before  the  attack  on  Placentia  this 
I  attack  need  never  have  been  delivered  at  all.  "  Placentia  would 
have  been  completely  paralysed  by  Cremona"  {op.  at.  p.  17).  And 
he  thinks  this  proved  by  the  fact  that  subsequently  Otho  recalled 
Spurinna  from  Placentia  when  Cremona,  without  doubt,  was  occupied 
by  the  enemy  (ii.  36).  This  last  fact,  however,  proves  nothing  in  his 
favour,  as  it  is  the  eve  of  the  last  struggle  and  Otho  must  have 
every  man  available.  Yet  even  so  a  garrison  is  still  left  by  him 
in  Placentia  sufficient  to  hold  the  town.  One  fortress  does  not 
"  paralyse  "  another.      Placentia  was  invaluable  to  Otho  when  there 


NOTES  339 

was  no  bridge  over  the  Po  at  Cremona.  Does  the  German  student 
expect  Placentia  to  take  the  field,  like  Birnam  Wood  ?  It  was  held 
to  guard  the  great  crossing  of  the  river,  and  served  this  purpose 
equally  well  even  if  fifty  Cremonas  on  the  north  bank  were  in  the 
hands  of  the  enemy. 


C. — The  Site  of  Bedriacum  (p.  86) 

The  exact  site  of  ancient  Bedriacum  has  been  a  matter  of  dispute. 
There  are  three  pieces  of  evidence  : — 

(a)  Tacitus. 

In  ii.  23  Tacitus  calls  it  a  vicus  "inter  Veronam  Cremonamque 
situs."  Naturally,  therefore,  it  would  lie  on  the  Roman  road  between 
these  two  towns.  In  iii.  1 5  he  implies  that  it  was  situated  where 
this  road  from  Verona  to  Cremona  was  joined  by  the  road  along  the 
north  bank  of  the  Po,  which  road  he  calls  the  Via  Postumia  (iii.  2  1  ; 
cf.  C.I.L.  i.  540).  This  was  probably  the  road  from  Hostilia  to 
Cremona. 

This  information  is  very  precise,  but,  unfortunately  for  purposes 
of  modern  identification,  the  actual  course  of  the  Roman  roads  in 
this  district  has  not  been  determined.  The  country  here  to-day  is 
very  flat,  and  well  drained  by  modern  ditches.  Centuries'  ruin  and 
flood  have  obscured  every  trace  of  the  old  roads  ;  nor  do  any  such 
traces  remain,  it  seems,  even  of  the  Roman  bridges,  e.g.  over  the 
Oglio,  which  would  greatly  assist  the  inquiry.  If,  then,  we  do  not 
know  the  actual  course  of  the  two  roads,  we  are  not  greatly  helped 
by  learning  that  the  village  of  our  search  lay  at  the  point  where  they 
joined. 

(b)  The  scholiast  to  Juvenal,  ii.  99  and  106. 

Here  Bedriacum  is  said  to  have  been  twenty  miles  from  Cremona, 
and  to  have  lain  between  Cremona  and  Hostilia. 

(c)  The  "  Peutinger  Table."  l 

1  Conrad  Peutinger  of  Augsburg  (a.d.  1565-1647)  possessed,  it  seems,  a 
thirteenth-century  copy  of  a  Roman  map  of  the  third  century  A.  D. ,  which  he 
published.  This  gives  a  picture  of  the  roads  of  the  Roman  Empire,  naming 
the  chief  stations  on  them,  and  giving  in  figures  the  distances  between  them. 
It  does  not  attempt  to  give  any  accurate  delineation  of  shape,  or  to  draw 
distances  to  scale.  The  result  is  a  series  of  long  strips  of  country  which 
presents  a  very  quaint  appearance.  But  its  information  on  vexed  questions 
of  site  may  of  course  be  valuable,  as  in  this  case. 


34Q  NOTES 

This  reads  as  follows  : — 

Cremona — xxii. — Beloriaco — ,, — Mantua — xl. — Hostilia. 

Beloriaco  is  obviously  Bedriacum.  But  there  are  two  difficulties 
here  :  (i.)  No  distance  is  given  between  it  and  Mantua  ;  (ii.)  The 
distance  from  Mantua  to  Hostilia  is  twice  the  correct  number  of 
miles.  Mommsen's  suggestion  (op.  Hermes,  v.  p.  163  n.)  is  that  forty 
was  the  number  of  miles  between  Bedriacum  and  Hostilia,  and  that 
Mantua  lay  on  a  branch  road  which,  "  as  often  happens,"  has  dis- 
appeared from  the  map.      Hence  the  right  reading  should  be 

Cremona — xxii. — Bedriaco — xl. — Hostilia. 

This  would  explain  why  the  Juvenal  scholia  placed  Bedriacum 
between  Cremona  and  Hostilia  rather  than  between  Cremona  and 
Mantua. 

The  difficulty  in  Mommsen's  view  is  that  it  is  hard  to  see  why 
Mantua,  an  important  town,  should  lie  on  a  side  road  which  would 
have  been  an  exceedingly  short  one,  and  why,  even  on  grounds  of 
military  exigency  (as  Mommsen  urges),  it  should  have  been  avoided 
by  the  main  road.  The  explanation,  however,  of  the  difficulty  of  the 
Peutinger  Table  datum  is  the  only  one  which  we  have,  and  holds 
the  field. 

The  distance  of  the  modern  village  of  Calvatone  from  Cremona 
tallies  practically  to  a  yard  with  that  given  by  the  Table  for 
Bedriacum  (33.02  kiL  =  2oi  English  =  22  Roman  miles).  The 
scholiast's  "twenty"  is  less  trustworthy.  The  distances  from  Cremona 
of  other  villages  suggested — Carneto,  Cividale,  S.  Lorenzo  Guazzone 
— suit  neither  of  the  figures.  I  have  little  doubt  that  Calvatone  lies 
on  the  site  of  Bedriacum,  and  that  the  Roman  road  ran  a  quarter 
of  a  mile  north  of  the  modern  high-road  from  Cremona  to  Mantua. 

D. —  The  '■'■Distances"  in  Taciitis,  ii.  39,  40  (p.  114) 

Two  statements  of  distance  in  Tacitus'  account,  when  put 
together,  present  very  great  difficulties  : — 

(1 )  ii.  39.  "  Promoveri  ad  quartum  a  Bedriaco  castra  placuit,  adeo 
imperite  ut  quamquam  verno  tempore  anni  et  tot  circum  amnibus 
penuria  aquae  fatigarentur." 

According  to  this  the  troops  reach  the  fourth  milestone  from 
Bedriacum,  i.e.  a  point  eighteen  miles  from  Cremona.  See  above, 
Note  C,  Position  of  Bedriacum. 


NOTES  341 


(2)  ii.  40,  41.  When  the  troops  resume  their  march  from  this 
point — 

"  Non  ut  ad  pugnam  sed  ad  bellandum  profecti  confluentes  Padi 
et  Aduae  fluminum  sedecim  inde  milium  spatio  distantes  petebant." 

Celsus  and  Paulinus,  in  the  next  sentence,  remonstrate  against 
exposing  their  tired  troops — "  militem  itinere  fessum  " — to  an  enemy 
who — "  vix  quattuor  milia  passuum  progressus  " — would  be  likely  to 
attack  them  with  vigour. 

The  crux  here  is  that  the  confluence  is  seven  miles  west  of 
Cremona,  and  thus  some  twenty-five  miles  from  the  position  in  (1), 
not  sixteen  miles. 

Other  difficulties  of  explanation  are  added,  e.g.  : — 

(i)  How  can  a  four-mile  march  distress  the  troops  for  want  of 
water  ?      It  seems  absurd,  however  hot  the  sun. 

(2)  Why  do  the  troops  want  to  get  to  the  confluence  at  all  ?  No 
motive  is  given  for  this  extraordinary  objective. 

(3)  How  can  the  troops  possibly  expect  to  get  to  the  confluence 
without  fighting,  as  the  road  lies  via  Cremona  ?  What,  then,  is  the 
meaning  of  "non  ut  ad  pugnam  sed  ad  bellandum  profecti"  ? 

Plutarch,  it  is  true  (Ot/to,  ii.),  has  a  different  and  straightforward 
account  which  presents  none  of  these  difficulties.  According  to  him, 
the  troops  march  first  day  fifty  stades  from  Bedriacum,  i.e.  about  six 
miles,  and  then  encamp.  The  want  of  water  is  due  to  the  "  ludicrous  " 
position  chosen  for  the  camp.  Next  day  Proculus  wishes  "  Trpodyew 
i~l  roi'5  TToAe/itovs  68bv  ovk  eXdrrova  (rraoYcoi'  €ko.t6i>."  1 00  stades 
=  1 2  miles.  This  would  bring  the  force  to  a  point  some  four  miles 
short  of  Cremona.  Plutarch  appears  to  have  thought  Cremona 
nearer  than  it  actually  was,  or,  more  probably,  the  battle  actually 
took  place  here.  (See  below.)  But  Suetonius  and  his  party  object- 
ing, nothing  is  done  until  Otho's  orderly  arrives  with  orders  "  fitj 
fievetv  fii]8e  8ia.Tpi/3et,v,  dW  dyew  evdvs  cttI  to  us  irokefuovs."  They 
therefore  "cipavres  €\(opovv."  Immediately  the  scene  shifts  to  Caecina 
and  Valens  and  the  battle,  without  any  further  hint  of  its  precise 
position  on  the  road. 

This  account  is  in  itself  not  altogether  satisfactory.  But  it  does 
not  raise  the  difficult  questions  presented  by  Tacitus.  There  is  no 
word  of  the  confluence  in  it,  no  hint  that  the  troops  were  not 
marching  out  to  make  a  frontal  attack  upon  the  Yitellian  position  at 
Cremona.  On  the  contrary  they  marched  along  the  road  to  fight, 
and,  reasonably  enough,  did  fight  as  soon  as  they  met  the  enemy, 


342  NOTES 

at  or  near  the  city :  four  miles  away,  if  we  insist  on  Plutarch's 
distances.  But  his  "  ovk  eAa-rrova"  allows  us  some  margin  to  play 
with. 

But  if  Plutarch  has  (on  the  whole)  a  simple  straightforward  view 
of  events,  this  does  not  help  us  in  the  least  to  solve  the  perplexities 
of  the  Tacitean  story,  unless  it  induces  us  to  reject  the  latter 
altogether  in  favour  of  the  alternative.  We  have  no  right,  and  small 
inclination,  to  adopt  this  heroic  course. 

Various  solutions  have  been  suggested  of  the  perplexities  of  the 
second  passage  in  Tacitus  : — 

(a)  Gerstenecker  proposes  a  remarkable  translation  for  it  : 
"  Nicht  wie  zu  einer  Schlacht,  sondern  wie  zu  einem  Feldzuge 
aufgebrochen,  befanden  sie  sich  auf  dem  Marsche  nach  der 
Mtindung  der  Adda  in  den  Po,  sechzehn  Milien  davon  entfernt " 
(op.  cit.  p.  32),  i.e.  "having  set  out,  they  found  themselves  at  a 
distance  of  sixteen  miles  from  the  objective  of  their  march,  the 
confluence." 

This  version  separates  "  confluentes "  and  "distantes";  makes 
"  distantes  "  agree  with  the  subject  of  "  petebant "  ;  and  translates 
"petebant"  as  the  equivalent  of  "arrived  at"  instead  of  "were 
making  for."  All  this  is  surely  impossible  as  a  mere  matter  of  the 
general  run  of  a  Latin  sentence.  It  also  gives  no  answer  at  all  to 
the  question  of  the  reason  for  the  objective,  or  to  the  question  why 
the  troops  did  not  expect  a  battle.  And  it  brings  the  army  nine  miles 
from  Cremona  ( =  sixteen  from  the  confluence)  for  the  battle  site,  which 
agrees  with  no  other  datum  of  any  kind  and  is  intrinsically  wildly 
improbable.  Gerstenecker  outrages  Latin  and  only  makes  confusion 
worse  confounded. 

(a)  The  new  Kiepert  map  of  North  Italy  (Berlin,  1902),  to  my 
amazement,  I  find  escorting  the  Adda  into  the  Po  through  the  very 
town  of  Cremona  itself.  I  can  only  infer  that  this  is  Kiepert's  effort 
to  solve  this  very  problem.  Bedriacum  to  Cremona  =  twenty  miles 
(according  to  the  scholiast's  account) :  the  troops  march  four  miles  : 
they  are  then  sixteen  miles  from  the  confluence.  It  must  follow  that 
Cremona  is  at  the  confluence  and  the  confluence  at  Cremona !  If 
Tacitus  will  not  suit  the  course  of  the  river,  then  the  course  of  the 
river  must  suit  Tacitus  !  Dr.  Grundy's  recent  map  of  Italy  (Murray, 
n.  d.)  avoids  this  error.  Of  course  this  solution  cannot  be  entertained 
for  a  moment. 

(c)  The  words  "  Confluentes  Padi  et  Aduae  "  give  all  the  trouble. 


NOTES  343 

How  simple,  then,  is  the  remedy  of  rejecting  the  words  "  Padi  et 
Aduae  "  entirely  as  an  addition  by  a  very  ignorant  scribe  which  has 
crept  unluckily  into  the  text !  The  troops  are  making  for  a  "  con- 
fluence "  sixteen  miles  away.  What  confluence  is  the  proper  distance 
from  their  first  camp  ?  At  this  point  a  stream,  the  "  Caneta "  by 
name,  is  produced  by  Nipperdey,  flowing  from  the  north  into  the 
Po.  Nipperdey's  "Caneta"  is  cheerfully  quoted  by  Professor  Bury 
(Students'  Rowan  Empire,  p.  349),  and  Mr.  Hardy  believes  in  "a 
small  stream  from  the  north  "  (P/utarch's  Lives  of  Galba  and  Ot/10, 
p.  254;  cf.  his  whole  note  on  the  matter,  pp.  253-55,  w'tn  its 
despairing  conclusion).  On  this  view  of  the  situation  the  troops 
intend  to  diverge  from  the  main  road  southwards  towards  the  Po 
"  to  get  into  touch  with  Macer's  gladiators."  But  they  all  mean  to 
fight,  and  the  words  "  non  ut  ad  pugnam  sed  ad  bellandum  profecti  " 
are  merely  meant  as  a  picturesque  description  of  the  general  careless- 
ness and  disorderliness  of  their  march. 

This  indeed  is  a  violent  remedy,  when  a  puzzled  scholiast  is  in- 
vented to  create  the  whole  difficulty.  Not  even  a  scholiast  would 
lightly  have  thought  that  the  Adda  joined  the  Po  east  of  Cremona. 
Why  in  the  world  should  he  have  inserted  an  impossible  river  into 
the  narrative  ?  Why  should  an  impossible  insertion  have  been 
accepted  ever  since  ?  Praestat  difficilius  ?  Not  a  bit  of  it !  We 
get  rid  of  the  words  which  cause  the  whole  difficulty  and  build  our 
hopes  on  the  river  Caneta  ! 

Veritably  they  are  built  on  sand  and  not  on  a  river  at  all !  The 
large  scale  Italian  ordnance  map  of  Cremona  lies  before  me  as  I 
write  (Fo.  61,  1897,  scale  1  :  100,000).  For  the  twelfth  time  I 
search  seventy-five  square  miles  of  country  east  and  south-east  of 
Cremona  to  find  the  name  of  Caneta.  There  are  plenty  of  "  Ca," 
but  this  presumably  stands  for  "Casa.!'  There  is  no  Caneta.  Very 
wisely  Mr.  Hardy  wrote  of  his  "small  stream  from  the  north"  with- 
out naming  it.  But  there  is  a  brooklet  or  a  ditch  called,  it  seems, 
Dugale  Pozzolo,  with  a  course  (so  far  as  I  can  track  it)  of  some 
seven  miles,  which  enters  the  Po  from  the  north  at  Isola  Pescaroli. 
Its  beginning  seems  to  be  four  miles  away  from  Cremona.  Its 
"confluence"  is  nine  miles  from  the  camp,  "fifty  stades  west  of 
Bedriacum."  This  distance  refuses  to  suit  anything.  Is  this  "Dugale 
Pozzolo  "  (if  that  be  its  name)  the  notable  stream  which  makes  a 
confluence?  Is  the  "confluence"  an  error  for  the  source  of  this 
noble   river   (9  +  7  =  16)?      What   then   befalls    Macer's   gladiators, 


344  NOTES 

who  could  have  jumped  this  ditch  here?  All  this  Caneta- erratic 
scholiast  story  is  a  mere  tissue  of  silliness. 

(d)  Mommsen  {Hermes,  v.  pp.  171-73)  proposes  to  regard  the 
distance,  sixteen  miles,  as  due  to  a  confusion  in  Tacitus'  own  mind 
between  the  ultimate  objective  of  the  whole  march,  viz.  the  confluence 
of  the  Adda  and  Po,  and  the  proposed  end  of  the  first  day's  march 
on  the  way  to  the  objective.  In  his  view  the  army  did  not  set  out 
to  fight,  nor  did  it  intend  to  march  for  the  confluence  by  way  of 
Cremona.  But  the  plan  was  that,  after  marching  for  some  distance 
along  the  main  road,  it  should  diverge  to  the  north-west  and  plant 
itself  astride  the  Cremona-Brescia  road  to  the  north  of  the  former 
town.  The  proposed  end  of  the  first  day's  march  is  sixteen  miles 
away  from  the  camp,  four  miles  from  Bedriacum.  [This  would  not 
have  reached  the  road  to  Brescia.]  But  the  troops  do  not  get  so 
far.  For  after  marching  along  the  road  twelve  miles  (Plutarch's  one 
hundred  stades)  to  the  point  where  they  mean  to  leave  it.  at  this 
point  they  meet  the  enemy  come  out  from  Cremona  to  fight. 
Tacitus'  whole  account  is  unintelligible  because  he  misunderstands 
the  military  situation.  The  army  is  ordered  to  provoke  a  fight,  not 
by  marching  straight  on  the  enemy  but  by  threatening  their  com- 
munications ;  and  the  distance  of  the  intended  end  of  the  first  day's 
march  is  confused  with  that  of  the  confluence,  the  ultimate  objective. 

Mommsen  thus  accepts  the  objective  and  propounds  briefly  the 
idea  of  a  flank  march,  but  rejects  the  number  sixteen  as  given  for 
the  distance  of  the  objective.  He  regards,  it  seems,  Brescia  as 
being  of  some  importance,  whereas  this  road,  as  I  have  explained  in 
the  text,  mattered  little  or  nothing  to  the  Vitellians.  And  the  whole 
march  is  treated  in  isolation  instead  of  as  part  of  one  great  strategi- 
cal idea.  The  criticism  of  Tacitus'  failings  as  a  military  historian 
must  win  the  acceptance  of  every  student  of  the  historian.  To 
Mommsen's  whole  paper  the  student  of  the  Histories  owes  much, 
and  not  least  in  the  consideration  of  the  difficulties  of  these  particular 
passages  in  Tacitus'  account. 

I  have  endeavoured  in  the  text  of  this  chapter  to  develop  my  view 
of  the  whole  strategy  of  the  battle.  But  with  regard  to  the  special 
difficulties  of  the  Tacitus  sentences  mentioned  at  the  beginning  of 
this  note,  I  would  suggest  that  one  simple  alteration  of  a  numeral  in 
the  first  passage  will  clear  most  of  them  up.  To  alter  the  "sedecim" 
(as  has  been  proposed),  of  course,  would  get  rid  also  of  the  main  crux 
of  the  second  passage,  but  leaves  the  "  water  "  difficulty  of  the  first 


NOTES  345 

unexplained.      But  an  alteration  of  "  quartum  "  to  "  quartum  decu- 
mum,"  if  allowed,  solves  both.      The  plan  works  out  as  follows  : — 

end  of  first  day's 
march.  (14  miles) 


Confluence  ,     ■,     Cremona    0     •,                                        ■,.      ■,               Bedriacum 
^  7 miles      %s  Smiles  k  74  miles £\ 

Cremona  to  Bedriacum  22  miles 

The  first  day's  march  is  fourteen  miles.  This  is  eight  miles 
from  Cremona  (22-14)  and  thus  fifteen  from  the  confluence.  The 
discrepancy  of  a  mile  (roughly)  need  not  trouble  us  greatly. 
The  force  is  then  "  fifteen  miles  from  the  confluence  "  in  a  straight 
line.  Perhaps  the  sixteenth  mile  allows  for  the  circle  round  the  city; 
for  of  any  wide  turning  movement  Tacitus  has  no  notion.  The 
troops  are  naturally  distressed  for  want  of  water  after  fourteen  miles, 
as  well  as  because  where  they  encamp  there  is  none  (Plutarch).  Next 
day  Suetonius  urges  that  the  enemy  will  have  "barely  four  miles  to 
march."  It  looks  as  if  he  thought  both  sides  would  set  out  to  march 
at  the  same  time,  and  so  would  meet  in  the  middle  of  the  eight  miles 
which  then  separated  the  Othonians  from  the  city. 

Tacitus  seems  to  believe  the  whole  march  and  fight  took  place  in 
one  day,  which  accounts  best  for  his  making  Suetonius  insist  on  the 
weariness  of  his  own  men.  Plutarch  definitely  says  the  fight  was  on 
the  second  day  after  leaving  camp.  This  is  more  probable.  But  1 
can  make  nothing  of  Plutarch's  figures.  They  disagree  entirely  with 
Tacitus'  whether  we  read  four  or  fourteen,  save  that,  if  we  read 
fourteen  and  then  make  the  force  advance  four  miles  farther  next 
day  for  the  battle  (the  "half  eight"  required  by  Suetonius'  state- 
ment of  the  distance  to  be  marched  by  the  enemy),  then  in  Tacitus' 
account  as  well  as  in  Plutarch's  the  battle  takes  place  eighteen  miles 
from  Bedriacum  (150  stades)  and  four  from  Cremona. 

My  whole  account  in  the  text  of  the  movements  of  the  troops  is 
based  on  my  "  fourteen  miles  "  suggestion  and  on  the  importance  of 
the  mention  of  the  confluence  of  Adda  and  Po.  Of  course,  the  view 
there  taken  of  Otho's  strategy  as  a  whole  is  entirely  independent 
of  the  difficulty  of  the  figures  names  are  far  more  valuable  than 
figures)  and  of  the  proposed  alteration  of  these. 


346  NOTES 

E. — Tacitus  as  Military  Historian 

In  my  Preface  I  have  stated  my  view  that  Tacitus'  information 
"in  matters  military  represents  the  common  gossip  of  the  camp,  the 
talk  of  the  private  or  of  the  subordinate  officer." 

The  narrative  and  the  notes  to  it  have  given  not  a  few  instances 
in  support  of  this  contention.  A  bare  summary  of  the  more  import- 
ant of  these  from  Chapter  I.  may  be  presented  here,  in  proof  of  the 
contention,  for  clearness'  sake  : — 

(i)  Caecina's  hesitation  as  to  his  route  to  Italy:  described  by 
Tacitus  as  a  choice  between  an  invasion  of  Italy  and  an  attack  on 
Noricum.      (Tacitus,  i.  70.)      See  above,  §  4,  pp.  66-69. 

(2)  Otho's  motive  for  sending  the  fleet  to  Narbonese  Gaul : 
Tacitus  fails  to  understand  its  bearing  upon  the  general  strategical 
situation.      (Tacitus,  i.  87  ;  ii.  12-16.)      See  above,  §  5,  pp.  73-79. 

(3)  Valens'  despatch  of  the  Batavian  cohorts  to  the  coast  to 
guard  his  flank  :  Tacitus  reproduces  the  complaints  of  the  common 
soldiers.  When  the  private  marching  east  sees  his  comrade  detached 
to  march  south,  he  does  not  think  of  strategic  necessities,  the  safety 
of  lines  of  communication  and  retreat,  etc.  etc.  When  he  murmurs 
secretly,  or,  in  the  time  of  civil  war  when  discipline  has  gone  to  the 
dogs,  grumbles  openly,  he  is  thinking  that  the  enemy,  of  unknown 
strength,  will  have  twice  as  good  a  chance  of  disposing  of  his  own 
valuable  life.  Hence  he  enunciates  in  exactly  the  wrong  place  a 
splendid  principle  of  strategical  concentration,  "  Let  the  whole  army 
stick  together,"  etc.  Tacitus  himself  sees  in  Valens'  action  only 
his  desire  to  "guard  his  allies  and  to  get  rid  of  mutineers  from  his 
army "  ;  both  only  primary  and  not  ultimate  strategical  reasons,  if 
true.     (Tacitus,  ii.  28.)     See  above,  §  4,  pp.  61-62. 

(4)  Spurinna's  reconnaissance  in  force  from  Placentia  :  according 
to  Tacitus  forced  by  a  mutinous  soldiery  on  a  reluctant  general. 
(Tacitus,  ii.  18,  19.)      See  above,  §  6,  pp.  82-83. 

(5)  Caecina's  attack  on  Placentia:  Tacitus  explains  it  as  due  to 
his  desire  to  score  an  opening  success— a  motive  of  very  secondary 
weight.      (Tacitus,  ii.  20.)      See  above,  §  6,  pp.  83-84. 

(6)  The  battle  of  Locus  Castorum :  a  most  confused  tale  in 
Tacitus.  But  he  does  seem  to  understand  and  appreciate  Suetonius' 
reasons  for  recalling  his  troops  from  pursuit.  (Tacitus,  ii.  24-26.) 
See  above,  §  6,  pp.  87-92. 

(7)  The  bridge-building  at  Cremona,  and— 


NOTES  347 

(8)  Otho's  order  "to  attack." 

According  to  Tacitus,  Caecina  and  Valens  hear  of  the  Othonian 
intention  to  attack,  and  so  "  like  wise  men  stay  quiet  to  take  advan- 
tage of  others'  folly":  "Caecina  ac  Valens,  quando  hostis  imprudentia 
rueret,  quod  loco  sapientiae  est,  alienam  stultitiam  opperiebantur." 
But  in  order  to  give  their  idle  troops  something  to  do,  they  begin 
their  mock  bridge-building — "ne  ipsorum  miles  segne  otium  tereret  " 
(Tacitus,  ii.  34).  Even  Mommsen  himself  agrees  that  the  bridge- 
building  "cannot  have  been  seriously  meant"  {Hermes,  v.  p.  165). 

This  entirely  fails  to  realise  the  whole  strategical  plans  of  both 
sides.  It  is  a  shallow  conclusion  from  results.  Because  the  bridge 
was  not  finished,  it  was  not  meant  to  be  finished.  Because  the 
Othonians'  advance  led  to  an  attack,  they  meant  to  attack.  Because 
they  were  defeated,  their  attack  was  folly. 

I  have  endeavoured  to  show  the  deeper  significance  of  the  plans 
of  both  sides,  the  evidence  for  them,  and  the  reasons  why  Tacitus 
failed  to  understand  them,  in  the  whole  of  §§  7,  8. 

(9)  Otho's  withdrawal  to  Brixellum  with  a  considerable  number 
of  troops  :  described  by  Tacitus  as  a  step  urged  on  him  and  agreed 
to  by  him  in  order  that  he  might  not  incur  the  risk  of  the  battle 
(Tacitus,  ii.  33).  I  have  commented  on  the  improbability  of  this  in 
§  7,  p.  107.  Apart  from  its  inconsistency  with  all  that  we  know 
besides  of  the  Emperor's  character  and  acts,  he  did  not  need  so  large 
a  bodyguard  if  all  that  he  and  others  were  thinking  of  had  been  his 
own  personal  safety. 

Tacitus'  whole  account  of  these  closing  scenes  of  the  campaign 
makes  the  Vitellians  act  like  babies  ("Satan  finds  some  mischief 
still,"  etc.),  and  Otho  a  strategical  idiot  and  a  craven  as  well.  He 
becomes  himself  again  only  just  in  time  to  die. 

The  picture  of  Vitellius,  however,  is  harmonious  throughout  and 
a  masterpiece.  So  simple  a  character  could  hardly  be  misunder- 
stood. 

In  the  following  two  chapters  Tacitus'  failings  are  those  of  the 
lack  of  strategical  insight  upon  more  than  one  occasion.  They  lend 
themselves,  however,  less  easily  to  the  purposes  of  a  catalogue,  and 
may  be  left  to  the  discovery  and  appreciation  of  the  reader.  See 
also  the  Index,  "Tacitus  as  Military  Historian." 


348  NOTES 

F. —  Valetis'  March  to  the  North  (p.  173) 

Tacitus'  account  of  this  (iii.  41,  42)  is  very  perplexing.  Accord- 
ing to  it,  Valens  receives  the  news  of  the  mutiny  of  the  Ravenna 
fleet  "while  on  the  march,"  i.e.  from  Rome.  He  was,  of  course, 
marching  by  the  Via  Flaminia.  This  road  runs  through  Southern 
Etruria  for  a  few  miles  after  leaving  Rome,  enters  Umbria  near 
Ocriculum,  and  continues  in  this  district  up  to  and  beyond  Ariminum, 
which  town  is  in  Umbria.  From  Ariminum  there  is  a  choice  of 
roads.  The  coast  road  runs  north  to  Ravenna,  crossing  the  Rubicon 
a  few  miles  from  Ariminum.  This  river  formed  the  northern 
boundary  of  Umbria.  The  main  road,  however,  strikes  away  north- 
west from  Ariminum,  leaves  Umbria  near  the  town  of  Cesena,  and 
continues  via  Faventia  to  Bononia,  whence  a  road  runs  north  to 
Hostilia,  while  the  highway  continues  in  a  straight  line  to  Placentia. 

The  puzzle  is  to  discover  where  Valens  was  when  he  heard  of  the 
mutiny  of  the  fleet.  Tacitus'  account  is  that,  when  the  news  came,  if 
Valens  had  hastened  he  might  have  reached  Caecina  while  the  latter 
was  still  wavering,  or  have  reached  the  legions  before  the  critical  battle. 
There  were  actually  some  who  advised  him  "  ut  per  occultos  tramites 
vitata  Ravenna  Hostiliam  Cremonamve  pergeret."  But  he  halts 
where  he  is,  and  sends  to  Rome  for  reinforcements.  When  these 
arrive  in  small  numbers,  viz.  three  cohorts  and  one  ala,  he  is  unable 
"to  force  his  way  through  the  enemy"  (vadere  per  hostes).  He  is 
fearful,  too,  of  their  scanty  loyalty.  "  Eo  metu  cohortes  Ariminum 
praemittit,  alam  tueri  terga  iubet  :  ipse  .  .  .  flexit  in  Umbriam 
atque  inde  Etruriam."  In  Etruria  he  hears  of  the  battle  at  Cremona, 
and  makes  his  way  to  the  sea  at  Pisa  (iii.  42). 

From  this  account,  studied  in  connection  with  the  geography,  I 
conclude  that  Valens  must  have  been  already  on  the  road  between 
Ariminum  and  Ravenna  when  the  news  of  the  fleet's  mutiny  came. 
Had  he  still  been  south  of  Ariminum  the  following  difficulties  present 
themselves  : — 

(a)  There  was  not  the  slightest  need  for  him  to  march  "per 
occultos  tramites "  if  he  desired  to  get  to  Hostilia  as  quickly  as 
possible  while  avoiding  Ravenna.  The  quickest  route  for  him,  if 
still  south  of  Ariminum,  was  by  the  main  road  via  Bononia.  But  if 
he  were  already  north  of  Ariminum  and  near  Ravenna,  then  the 
cross-roads  recommended  to  him,  e.g,  from  Cesenatico  to  Cesena, 
would  save  time. 


NOTES  349 

(b)  When  his  reinforcements  have  arrived  he  will  not  inarch  via 
Ravenna,  as  they  are  too  weak.  He  therefore  "sends  the  infantry 
before  him  to  Ariminum,  and  bids  the  cavalry  guard  his  rear." 
Now,  if  he  is  marching  in  a  hurry  north  to  Ariminum,  these  dis- 
positions are  too  ludicrous  for  words.  The  danger  is  all  in  his 
front  :  he  therefore  puts  his  cavalry  to  guard  his  rear  ;  the  need  for 
speed  is  urgent  :  he  therefore  places  his  infantry  at  head  of  the 
column  !  But  if  he  is  retiring  on  Ariminum  from  the  north  the 
arrangement  is  clear.  The  cavalry  are  properly  placed  to  guard 
the  rear  of  the  retreat.  He  himself,  however,  does  not  accompany 
the  column,  as  he  has  given  up  the  plan  of  getting  to  the  Po  at  all. 

(c)  He  himself  "turns  aside  first  to  Umbria,  next  to  Etruria." 
But  if  he  is  south  of  Ariminum  he  is  already  in  Umbria.  For  he 
cannot  have  been  still  within  a  few  miles  of  Rome.  (Why  should 
the  cavalry  guard  his  rear,  above  all,  in  this  case  ?)  If  he  has  alreadv 
crossed  the  Rubicon,  he  may  have  recrossed  it  into  Umbria  before 
leaving  that  district  again  for  Etruria  and  the  west.  But  even  this 
would  give  him  but  a  mile  or  two  in  Umbria,  as  he  clearly  did  not 
accompany  his  troops  to  Ariminum,  which  is  hard  by  the  boundary 
of  the  province.  The  words  "  flexit  in  Umbriam  "  are,  indeed,  almost 
unmeaning.  Valens'  obvious  route,  if  (for  reasons  (a)  and  (6))  we 
believe  him  to  have  been  north  of  Ariminum,  was  to  proceed  via 
Cesena  to  Faventia  (Faenza),  thence  cross  the  mountains  by  the 
well-known  road  to  Florence  and  the  valley  of  the  Arno,  and  so  to 
Pisa  at  the  mouth  of  this  river.  This  ridge  of  the  Apennines  is  not 
easily  crossed  even  by  a  small  company,  as  those  who  have  roamed 
about  the  Prato  Magno  chain  or  crossed  from  the  Casentino  and  La 
Verna  to  Badia  Prataglia  and  Urbino  can  tell.  Even  the  road  over 
the  Consuma  Pass  has  only  just  been  opened.  The  mountains  are 
truly  magnificent,  and  I  saw  no  snow  on  Falterona  in  September. 
But  the  going  is  hard.  I  doubt  if  Valens  and  his  small  faithful 
band  disturbed  the  autumn  leaves  in  Vallombrosa.  North  of  the 
Metaurus  and  the  Furlo  Pass  (whence  the  Arno  valley  can  be 
reached  by  Urbania  and  Borgo  San  Sepolcro,  or  by  Gubbio  farther 
south)  the  first  good  track  to  the  Arno  valley  is  the  route  now 
followed  by  the  railway  from  Faenza  via  Brisighella  and  Borgo  San 
Lorenzo  to  Florence.  If  Valens  was  in  the  neighbourhood  of  Rimini 
and  wished  to  get  speedily  to  Pisa,  this  was  his  natural  road. 

If,  then,  Valens  had  crossed  the  Rubicon  travelling  north,  his  near 
approach  may  account  for  the  sudden  mutiny  by  night  at   Ravenna, 


350  NOTES 

which  seems  to  have  surprised  Caecina  before  he  was  ready  also 
to  play  the  traitor.  For  Tacitus  says  that  Valens  could  still  have 
reached  him  by  rapid  movement  before  he,  too,  went  over. 

But  if  there  were,  as  Tacitus  suggests,  a  secret  agreement  between 
Caecina  and  liassus,  Valens  must  have  been  near  at  hand  indeed  if 
he  could  have  arrived  in  time  to  dissuade  Caecina  from  following  his 
fellow-traitor's  example. 

The  whole  discussion  may  at  least  serve  to  illustrate  Tacitus' 
inexcusable  vagueness  in  his  military  history.  His  interest  in  geo- 
graphy is  evidently  of  the  most  casual  description.  One  other 
example  of  this  blemish  in  the  history  may  here  be  appropriately 
mentioned.  Valens  is  presently  captured  by  the  Flavians  at  the 
lies  d'Hyeres,  near  Toulon  (iii.  43).  When  Tacitus  next  mentions 
him  it  is  to  remark  casually  that  he  is  in  prison  at  Urbinum  (iii.  62). 
Why  at  this  little  mountain  city  in  Umbria,  of  all  places  in  Italy  ? 
Why,  when,  and  how  did  the  Flavians  take  him  there  ?  Doubtless 
his  head  was  useful  to  the  enemy  marching  for  Rome,  and  it  must 
have  been  forwarded  to  them  for  use  at  Carsulae,  eighty  miles  to  the 
south.  After  all,  Hasdrubal's  head  travelled  a  longer  journey  down 
the  same  road.  But  why  the  unlucky  Valens  should  have  been 
escorted  for  execution  to  Urbino,  which  lies  up  in  the  hills  ten  miles 
from  Fossombrone  and  away  from  the  main  road  south,  is  a  problem 
to  which  I  can  give  no  answer,  and  Tacitus  vouchsafes  no  explana- 
tion. If  Valens  had  been  a  Garibaldi  fleeing  through  these  moun- 
tains for  dear  life,  Urbino  might  have  been  his  San  Marino,  and  the 
problem  would  be  easier.      But  Valens  was  a  prisoner. 

G. — Vetera  and  Harper's  Ferry  (p.  283) 
The  situation  on  the  Potomac  at  the  time  of  the  Confederates' 
invasion  of  Maryland,  following  upon  their  brilliant  victory  of  Second 
Manassas  in  September  1862,  was  so  similar  to  that  on  the  Rhine 
towards  the  end  of  a.d.  69  that  the  former  may,  it  seems  to  me, 
help  to  explain  the  latter  so  far  as  Vocula's  decision  not  to  evacuate 
Vetera  is  concerned.  Of  course  no  other  comparison  is  intended, 
for  never  did  the  brilliant  genius  of  the  Southern  commanders  in 
civilised  warfare  more  completely  outshine  the  mediocre  respect- 
abilities on  the  Federal  side  than  in  this  campaign,  and  the  mere 
thought  of  a  Stonewall  Jackson  and  a  Civilis  together  seems 
ludicrous.  But  a  brief  note  may  be  desirable  to  explain  the  point 
of  the  comparison  in  the  text. 


NOTES  35, 

A  bare  sketch  may  serve  to  illustrate  the  note  :  the  Roman 
"equivalents"  are  printed  in  capitals.  Points  of  the  compass,  of 
course,  are  of  no  matter  in  such  a  comparison,  and  the  Potomac- 
Rhine  may  be  taken  as  flowing  in  any  direction  : — 

_     ,     .  n  Baltimore 

Frederic  u  cologne: 

D-0ELDW6A y 

A 
! 


a 

Harper's  Ferry 

VETERA 


Potomac     River  —~ QWashington 

-    RHINE  mainz 


D 
Leesburg 


Dotted  lines  show  the  direction  of  attack  (expected  or  actual)  by  the 
Confederates  or  Germans. 

At  Harper's  Ferry  there  was  a  Federal  garrison  of  twelve  thousand 
men  ;  at  Washington,  M'Clellan's  main  army  of  ninety  thousand  men. 
The  Confederates,  fifty  thousand  strong,  invaded  Maryland,  from 
Leesburg  to  Frederic,  and  threatened  an  advance  on  Baltimore,  if 
not  on  Washington  itself.  The  question  for  the  Federal  authorities 
was  whether,  while  there  was  still  time,  the  garrison  at  Harper's 
Ferry  should  be  recalled.  M'Clellan  desired  this,  but  his  political 
superiors  overruled  him.  It  was  clear  that  so  strongly  garrisoned  a 
fort  must  seriously  threaten  the  line  of  communication  of  the  Con- 
federate invaders.  Might  it  not  be  expected  to  stop  their  projected 
inroad  entirely  ?  And  if  from  Frederic  they  turned  on  Harper's 
Ferry,  surely  they  could  be  caught  between  two  fires — the  garrison 
and  M'Clellan's  army — and,  outnumbered  (as  usual)  by  two  to  one, 
be  annihilated.      So  the  parallel  works  out  thus  : — 

_  Federal  .  Harper's  Ferrv  ...  . 

Cal1    thC    ROMAN    gamSOn    at  VETERA  ^^    °r    t™*™* 

..        ,  Baltimore        ,    ,      .     Confederates  .,  , 

the  advance  on „„„  made  by  the     „  if  these  propose 

Cologne  '  Germans  ^    v 

to  neglect  it  and  advance  via   „  ,  leaving  the  fort  in   their 

Gelduba'  ° 

rear  ?     And  if  not,  can  the  foe  not  be  driven  back  upon  the  hostile 

f    .   ,      .,  Federal  M'Clellan      ,  up 

iort    by  the  main   „  army  under    ,T  advancing    , 

'  Roman         '  Vocula  down 

..  Potomac  ,         Washington , 

the  river    _  from       »»  ? 

Rhine  Mainz 


352  NOTES 

In  the  result,  the  Federal  Government  risked  it  and  ordered 
the  garrison  to  remain.  The  Confederates  thereupon  dared  to 
detach  a  force  of  twenty-five  thousand  men  upon  Harper's  Ferry, 
despite  the  imminent  peril  from  M'Clellan,  and  the  garrison  of  the 
fort  (somewhat  ignominiously)  surrendered  after  some  show  of 
resistance.  Vocula  similarly  risked  it,  and  also  lost  his  fortress 
in  consequence,  under  the  circumstances  detailed  in  the  text.  But 
I  take  it  that  the  retention  of  both  Harper's  Ferry  and  Vetera  was 
dictated  by  much  the  same  military  considerations. 


H. — The  Flavian  Army  of  the  Lower  Rhine  (p.  331) 

A  newly-found  military  diploma,  of  date  April  1  5,  A.D.  78,  shows 
that  there  were  then  stationed  on  the  Lower  Rhine  these  six  alae, 
viz. :  ala  Noricorum  (placed  by  Ritterling  at  Burginatium  by  Calcar), 
ala  Singularium  (at  Vada),  ala  Moesica  (at  Asciburgium),  ala 
Afrorum  veterana  (at  Vetera  ?),  ala  Siliana,  and  ala  Sulpicia  (?  at 
Bonn,  Noviomagus,  or  Neuss),  and  also  the  Cohors  I.  Flavia 
Hispanorum.  To  these  must  be  added  the  ala  Indiana  (at 
Worringen).  Ritterling  supposes  that  all  these  troops  made  their 
first  appearance  on  the  Rhine  under  Vespasian.  The  actual 
recipient  of  the  diploma,  though  belonging  to  the  ala  Moesica,  is 
by  nationality  a  Trevir.  Otherwise  this  evidence  is  consistent  with 
the  view  taken  on  pp.  330-331. 

Besides  the  above  cohort,  Ritterling  gives  the  following  cohorts 
to  the  Flavian  army  of  the  Lower  Rhine: — II.  Asturum,  II.  Britt., 
I.  and  II.  civ.  Rom.,  III.  Dalmatarum,  VI.  ingenuorum,  II.  His- 
panorum, .  .  .  Lucensium,  III.  Lusitanorum,  I.  Tracum  (?),  II. 
Varcianorum,  I.  Vindelicorum,  and  XV.  voluntariorum.  This  is  a 
completer  list  than  Riese's,  but  the  conclusions  to  be  drawn  from  it 
are  the  same. 

Also  Legio  VII.  Gemina  seems  to  have  been  employed  on  the 
Rhine  by  Rutilius  Gallicus  in  A.D.  77-78  in  his  war  with  the  Bructeri. 
It  was  probably  in  this  war  that  the  prophetess  Veleda  was  captured 
(see  p.  316).  Cf.  E.  Ritterling  ap.  Korrespondensblatt  der  westd. 
Zeitschrift,  xxv.  (1906),  pp.  20-28,  and  von  Domaszewski  in  Linden- 
schmit's  Heidn.  Altert.  Bd.  v.  Heft  vi.  (1905). 


INDEX 


Abetone  Pass,  the,  54 

Adda,  the  river  (Adua),  101-114,  124> 

341-345 
Adige,  the  river,  169-172,  185,  187 
Aedui,  the,  59 
Agrestis,  Julius,  213,  214 
Agricola,  73 

Albenga  (Albingaunum),  76,  78 
Albinus,  Lucceius,  25 
Alexander,    Tiberius    Julius,     22,    23, 

137.  141 
Allobroges,  the,  60 
Alps,  the  Cottian.      See  Mont  Genevre 

Pass 
the  Julian  or  Pannonian,  147,  148, 

158,  159,  164,  165 
the  Pennine.     See  St.  Bernard  Pass, 

Great 
Passes  of  the,  39,  40,  42-45.  57.  58- 

67,  68,  291 
Altino  (Altinum),  169,  185 
Andernach,  58 
Annius  Gallus.     See  Gallus 
Antibes  (Antipolis),  76 
Antioch,  142,  144 
Antiochus  of  Commagene,  143 
Antonius   Primus,   M.,    138,    157-164, 

176,  178,  179,  182-213,  218-227, 
248,  249 

Antonius,  Saturninus,  rebellion  of,  294 
Apennines,    passes    of    the,    54,    209- 

212,  349 
Aponius  Saturninus,   M.,    31-32,   155- 

156,  160,  165,  189 
Aquae  Helvetiorum  (Baden),  63 
Aquae    Mattiacae    (Wiesbaden),    251, 

33i 
Aquileia,   37,   40,   46-50,    56,    66,    71, 
77>  79'  86,   101,    103,    136,    141, 
148,    158,    161,    164,    169,    170, 

177.  185 
Arenacum  (Ryndern?),  310 
Argentoratum  (Strassburg),  328 


Aricia,  212 

Ariminum  (Rimini),  39,  172,  183,  184, 

209-211,  348 
Arlberg  Pass,  the,  67,  293 
Army,  the  Roman  : 

the  Imperial  military  system,  16-19, 

329-331 
the  permanent  camps(Castrastativa), 

18-19,  266,  324-329 
the  system  of  recruiting,  17-18,  259, 

323-329 
the  cavalry,  17,  160 
the  Cohortes  praetoriae  (Praetorian 
Guards) 
under  Nero,  2 
under  Galba,  7,  13,  15 
under  Otho,  26,  27,  32-35,  54-55. 

70,  75,  88-89 
under  Vitellius,  130, 134-135,  198- 
200,     202,    208,    214,    2T5, 
217 
the  Cohortes  urbanae,   32-35,    130, 

214-215,  217 
the  Auxilia  : 

numbers,  etc.,  16,  17,  256 
system  of  "clan  regiments,"  257- 

259.  329-331 
Alae  : 

Afrorum,  352 
Batavorum,  257,  262 
Gallorum  Sebosiana,  17J 
Hispanorum  Auriana,  162 
Moesica,  352 
Noricorum,  352 
Petriana,  66 

Picentina,  257,  289-290,  296 
Siliana,  34,  35,  43,  65,  70,  352 
Singularium,  293,  331,  352 
Sulpicia,  352 
Tampiana,  154 
Tauriana,  27,  35,  60 
Treverorum,  75,  247,  256,  257, 
259,  262,  263 

!  2  A 


354 


INDEX 


Cohortes  : 

Asturum,  256,  331,  352 
Baetasiorum,  290,  330 
Batavorum,    59,   61,    131,    133, 
134,    248,    256-258,    263, 

33° 
Belgarum,  256,  257 
Breucorum,  256 
Cannenefatum,   256,   257,   263, 

33° 
Dalmatarum,  352 
Flavia  Hispanorum,  352 
Lusitanorum,  352 
Nemetum,  256 
Nerviorum,  256,  257,  290 
Pannoniorum,  80,  331,  338 
Raetorum,  256,  331 
Silaunensium,  256 
Thracum,  352 
Tungrorum,     240,     241,     256, 

257,  262,  290,  330 
Ubiorum,   256,  257,   262,  263, 

271.  33° 
Varcianorum,  352 
Vasconum,  256,  274 
Vindelicorum,  256,  331,  352 
the  Legions  : 

I.    6,    11,   35,    57,   88,    167,    198, 
199,   254,   256,   263,   268- 
276,   289,   297,   300,   325, 
326 
I.   Adjutrix,   33-35,  70,  n8,  132, 
133,   138,    166,   207,    214, 
292,  294,  328 
I.    Flavia  Minervia,  327,  328 
I.    Italica,   27,   35,  60,   167,    170, 

198,  199 

II.    Adjutrix,  214,  291,  292,  307 
II.   Augusta,    24,    166,    167,    198, 

199,  328 
III.    Augusta,  23,  166 
III.  Cyrenaica,  22,  143 

III.  Gallica,    30-32,    35,    iSS'^. 

141,    142,    155,    161,    165, 
189,  198,  199,  202,  203 

IV.  Flavia,  326 

IV.   Macedonica,  26,  35,  167,  198, 
199,    255,    256,    268-276, 

325-  32° 
IV.   Scythica,  21,  141 
V.   Alaudae,  26,  35,  57,  118,  167, 
181,    190,    198,    199,   253, 
256,    262,    264-276,    325, 

327 
V.    Macedonica,  21,  142 
VI.    Ferrata,  21,  141    163 
VI.    Vtctrix,  24,  166,  292,  307,  328 


VII.   Claudia,  30,  35,  135-137,  155, 

161, 189, 198, 199,  291,  292 

VII.   Galbiana  (or  Gemina),   6,   30, 

35,    101,    135,    136,    138, 

157,   158,    186,    198,    199, 

200,  203,  352 

VIII.   Augusta,  30,  35,  135-137,  161. 

189,    198,   199,   291,   292, 

294,  328 

IX.    Hispana,    24,    166,    167,    198, 

199,  223 
X.   Fretensis,  21,  142 
X.   Gemina,  24,  166,  292,  328 
XI.   Claudia,  30,  35,  101,  135,  136, 
13^>    155-   291,   292,    294, 
328 
XII.    Fulminata,  21,  141 

XIII.  Gemina,    30,    35,    86,    88,    95, 

118,    135,    136,    138,    186, 
198,  199,  291,  337 

XIV.  Gemina,    24,    30,    31,    35,    95, 

118,    134,    135,    138,    166, 
235-   257.   291,   292,   294, 
305,  306,  328 
XV.   Apollinaris,  21,  142,  326 
XV.    Primigenia,    26,    35,    57,    167, 
198,    199,   253,   256,    262, 
264-276,  325,  326,  327 
XVI.    26,  35,  57,  167,  198,  199,  254, 
256,    269-276,    289,    297, 
300,  325,  326 
XVI.   Flavia,  326 

XX.   Valeria  Victrix,   24,    166,   167, 

198,  199 

XXI.   Rapax,  26,  35,  57,  118,   167, 

170,    198,    199,   255.    291- 

294,  303,  327,  328 

XXII.    Deiotariana,  22,  142,  198,  199, 

326 
XXII.    Primigenia,    26,    35,    57,    167, 
255,256,  268-276,  325-328 
Arnhem,  231,  236,  262,  263,  310 
Arrius  Varus,  157,  163,  164,  189,  193, 

221 
Asciburgium(Asberg  by  Mors),  257,  273 
Ateste  (Este),  169,  185 
Augustus,    the   Emperor,    7,    19,    235, 

251,  266,  301,  324,  329 
Auspex,  Julius,  247 
Auxilia,  the.      See  Army,  the  Roman 
Aventicum  (Avenches),  64,  69 

Baetasii,    the,   241,  244,  330  ;   and  see 

Army,  the  Roman 
Bassus,  Annius,  155 
Bassus,  Sextus  Lucilius,  170,  179-181, 

183 


INDEX 


355 


Batavi,  the,   232-241,   245,    248,    261, 
262,    300,    302,    305,    314,    315, 

319.  32I-323.  33° 
Batavian    cohorts,    the,    59,    61-62, 

235,  248,  257,  258,  261 
Insula    Batavorum,    232-234,    236, 

253'  257.  262,  308,   309,  313 
Oppidum  Batavorum  (?  Cleve),   236, 
308 
Batavodurum  (?  Nymwegen),  232,  262, 

310,  328 
Bedriacum  (Calvatone)  : 
site  of,  85-86,  339-340 
Othonian  camp  at,  85,  95,  101-114 
first  battle  of,  114-121,  140 
second  battle  of,  192-204 
Belgae,  the,  241,  256  ;  and  see  Nervii, 

the 
Berenice,  Queen,  144 
Berytus,  144,  154 

Bingium  (Bingerbriick),  296,  298,  331 
Bolanus,  M.  Vettius,  24 
Bonn,  9,  26,  254,  257,  263,  264,  268, 

309,  311,  327,  328,  352 
Bononia  (Bologna),  39,  129,  135,  183, 

348 
Brenner  Pass,  the,   46,  56,  66-69,  86, 

169,  187 
Brescello  (Brixellum),    106,    107,    116, 

122,  124 
Brescia  (Brixia),  101,  203,  344 
Briancon  (Brigantio),  61,  78 
Bridge  at  Cremona,   the,   94-96,    104, 

105,  346-347 
Briganticus,  Julius,  247,  293,  320 
Brinno,  261-262 

Britain,  the  Army  of,    24,   36,  59,  96, 
99,    166  ;  and  see  Provinces,   the 
Roman 
Bructeri,  the,  239,  289,  300,  302,  352 
Brundisium,  147 
Byzantium,  146,  148 

Caecina  Alienus,  Aulus  : 
early  career,  11-12 
invasion  of  Italy,  57,  62-70,  78,  83- 

92 
first  campaign  of  Bedriacum,  92-95, 

117-120,  122 
under  Vitellius,  128,  137 
second  campaign  of  Bedriacum,  167, 

184,  186-190,  204-205 
generalship,  88-92,  100 
character,  63,  83,  93,  180 
Caeracates,  the,  295 
Cagli  (Cales),  209 
Calenus,  Julius,  247 


Caligula,    the    Emperor,    3,    10,    229, 

261,  317 
Calvatone.     See  Bedriacum 
Campania,  216-217,  230 
"Caneta,"  the,  343"344 
Cannenefates,  the,  234,  235,  237,  239, 

258,  261-262,  305,  306,  315,  330 
Cantiano,  209 
Capito,  Fonteius,  9 
Capitol,   the  burning  of  the,  224-225, 

244,  285 
Carmel,  Mount,  140 
Carsulae,  219,  221,  350 
Castel  by  Mainz,  251-252 
"  Castra  stativa,"  the,  18-19,  324-329 
Castra  Vetera.     See  Vetera 
Cerialis,  Q.  Petilius,  223-225,  292-317 
Chalon-sur-Saone  (Cabillonum),  59 
Chatti,   the,   232,   234,  239,  251,  284, 

294.  3!5>  332 
Chauci,   the,  236,  237,  239,  304,  309, 

320 
Cherusci,  the,  240 
Cimbri,  the,  240 
Civilis,  Julius,  245,  247-250,  257,  258, 

261-277,  284,  288-290,  295,  298, 

300,  304-315,  317,  320,  322 
Classicus,    Julius,    75,    247,   258,   285- 

288,  295,  298,  300,  304,  305,  317 
Claudius,  the  Emperor,  3,  10,  12,  19, 

22,  24,  236,  251,  252,  301,  317 
Claudius  Sanctus,  289 
Cleve  (?  Oppidum  Batavorum),  236,  308 
Cluvius  Rufus,  24,  25 
Coblenz  (Confluentes),  9,  58,  239,  241, 

297 
Colline  gate,  the,  223,  226-227 
Cologne,  11,  38,  58,  239,  240,  245,  254, 

264,  268-270,  285,  288,  304,  351 
"  Cologne  Column,"  march  of  the,  40, 

57-62 
Constance,  Lake,  7,  9,  28,  231,  293 
Consuma  Pass,  the,  349 
Corbulo,  Domitius,  157,  202,  236,  251, 

325 
Corsica,  76,  -]-j 
Cremona  : 

in  the  Vitellian  invasion,  62,  80-81, 
83-87,  90,  92-95,  101-124,  129, 
T-2>b-i?£>,  338-339 
in     the    Flavian    invasion,    169-173, 

183,  190-204 
sack  of,  204-206,  220 
Cugerni,  the,  240,  270,  307,  330 

Dacians,  the,  162,    163,  207,  285,  327 
Danube,  the,  28,  29,  161,  162 


356 


INDEX 


"  Danube  Army,"  the  : 
under  Otho,   26,   29-32,  36,  79,  95, 

101-106,  337-338 
under  Vitellius,    135-138,    153,  154, 
161,  164-165,  177-179 
"  Delmona,"  the,  194 
Die  (Dea  Vocontiorum),  61 
Domitian,    the    Emperor,    9,    20,    22, 
156,    220,    224,    251,    253,    294, 

3I5.  327-  332 
Doria  Riparia,  the,  61 
Drome,  the  river,  60 
Druids,  the,  244-245,  285 
Drusus  Claudius  Caesar,  236 
Drusus'  Agger,  236-237,  309 
Drusus'  Fosse,  236-237 
Dubius  Avitus,  251 
Dyrrhachium,  147,  148 

"  Eastern  Army,"  the,  21-22,  137,  141- 

144 
Egypt,  21-23,  34,  143,  146 
Elbe,  the,  8 

Emmerich,  231,  232,  236 
Eporedia  (Ivrea),  65,  70 
Epponina,  318 

Fano    (Fanum    Fortunae),     209,    211, 

212,  215,  219 
Faventia  (Faenza),  348-349 
Ferento  (Ferentinum),  13 
Finance,  Roman,  7,  131 
Firmus,  Plotius,  126 
Flaccus,   Hordeonius,  9,  36,  166,  248, 

259,  262,  263,  268,  269,  284 
Fleet,  the  Roman  : 

under  Otho,  50-52,  61,  70,  73-79 
under  Vitellius,    149,    180-181,  214, 

216,  348-350 
the  Flavian,  146-149 
the  Rhine  flotilla,  237,  262,  308-313 
the  "  Classis  Britannica,"  306 
Florus,  Julius,  243,  247 
Foligno  (Fulginium),  209,  210,  219 
Fonteius  Agrippa,  248,  292 
Forum    Alieni    (Legnago),    170,    171, 

185,  187 
Fossombrone  (Forum  Sempronii),  209 
Frejus  (Forum  Julii),  51,  52,  61,  75 
Frisii,   the,    234-237,    239,    251,    262, 

304,  320,  323,  330 
Frontiers  : 

the   Danube,    29-32,    157,    161-163, 

285,  292,  332 
the  Eastern,  21,  143,  145 
the  Rhine,   8-9,    28,    142,    236-237, 
251-252,  266,  332 


Furlo  Pass,  the,  349 

Fuscus,  Cornelius,   154,  156-157,  180, 


Galba,  Servius  Sulpicius,  the  Emperor: 
descent  and  early  career,   5,  24-25, 

251 
election  as  Emperor,  5-6 
principate  of,   6-15,   23,  27,  33,  6o, 

139,  157,  248 
death,  15,  22 
Galeria,  129 
Galerius  Trachalus,  129 
"Gallic  Independence,"  242-245 
Gallic  tribes  : 

See  Aedui,  Allobroges,  Baetasii, 
Belgae,  Caeracates,  Helvetii, 
Leuci,  Lingones,  Mediomatrici, 
Nemetes,  Nervii,  Remi,  Se- 
quani,  Treveri,  Triboci,  Van- 
giones 
Gallus,  Annius,  70,  71,  81,  85,  86,  96, 

114,  292-294,  297,  315 
Gallus,   Herennius,  254,  263,  269,  270 
Gaul.     See  Provinces,  Roman 
Gelduba  (Gellep),   269-271,   273,   275- 

278,  284,  303,  351 
German  tribes  : 

See  Batavi,  Bructeri,  Cannenefates, 
Chatti,  Chauci,  Cherusci,  Cim- 
bri,  Cugerni,  Frisii,  Marsaci, 
Mattiaci,  Suebi,  Sugambri, 
Tencteri,  Tungri,  Ubii,  Usipi 
Germany,     Lower    and    Upper.       See 

Provinces,  Roman 
"  Germany,  the  Army  of"  : 

under  Galba,  9-12,  26-28,  130-131 
under  Otho  and  Vitellius,  159,  164- 
169,    181-184,     196-197,     250- 
260,  323-332 
Germany,   Independent,  7-8,   239-240, 

307,  318-319 
Gladiators  as  troops,  70,  84,  94,   119 
Goch,  270 

"Golden  House,"  the,  131 
Grinnes,  310 

Hadrian,  the  Emperor,  132 
Helvetii,  the,  28,  63-64 
Herod  Agrippa  II.,  King,  143,  144 
Hordeonius  Flaccus.     See  Flaccus 
Hostilia  (Ostiglia),  39,  46,  71,  86,  102, 

103,  106,  169-172,    185-187,  190, 

192,  348,  349 

"  Imperium  Galliarum,"  the,  243-245, 
29c  3T7.  320-321 


INDEX 


;57 


Inn,  the  river,  28,  67-69,  162,  187 

Interamna  (Terni),  218,  219,  221-223 

Intimilium  (Ventimiglia),  74 

Isola  Pescaroli,  343 

Istria,  30,  158 

Italicus,  Prince  of  Suebi,  161 

"  Italy,  Army  of,"  37,  42,  43,  164 

Ivrea.     See  Eporedia 

Javolenus  Priscus,  251 

Jazyges,  the,  162 

Jerusalem,  143,  146 

Jewish  war,   the,  21-23,    138-i39>  142- 

r43.  I4S.  M6,  281-282 
Josephus,  140 

Julius  Caesar,  241,  242,  244,  247,  317 
Jura,  the,  64,  242 

Labeo,  Claudius,  257,  259,  290,  295 
Langres  (Andemantunnum),  58,  59 
Legions,   the  Roman.     See  Army,   the 

Roman 
Legnago    (Forum    Alieni),    170,    171, 

185,  187 
Lek,    the  river,    231,   232,   236,    309, 

312 
Leuci,  the,  58 
Lingones,   the,   25,   58,   242-244,  248, 

285,  293-295,  297,  300,  302,  305, 

3i5.  320,  330 
Lippe,  the  river,  250,  289,  312 
Locus  Castorum,  battle  of,  88-92,  338 
Luc-en-Diois  (Lucus  Augusti),  61 
Lugdunum  (Lyon),  26-28,   34,  59,  60, 

128,  131 
Lusones,  the,  25 

Maas,  the  river  (or  Meuse),  232,  240, 

241,  290,  312 
Macer,  L.  Clodius,  23 
Macer,  Martius,  84,  94,  95,  108 
Mainz  (Moguntiacum),   26,   239,   251- 

257,    263,    268,    284,    285,    288, 

290,  296,  297,  315,  328,  351 
Mantua,   46,   71,   81,  85,  86,  190-191, 

339-340 
Marcodurum  (Diiren),  270 
Marius    Celsus,     86,    88-89,    °6.  106, 

116,  119,  121 
Marne,  the  river,  58 
Marsaci,  the,  239 
Mattiaci,  the,  239,  251,  284,  330 
Mauretania.    See  Provinces,  the  Roman 
Maximus,  M.  Trebellius,  24 
Mediomatrici,  the,  58,  241,  294,  297 
Merwede,  the,  232 
Messalla,  Vipstanus,  189,  195 


Metaurus,  the  river,  209,  349 
Metz  (Divodurum),  58,  241,  298 
Mevania  (Bevagna),  209,  214-219 
Milan  (Mediolanum),  65 
Mirebeau-sur-Beze,  294 
Moguntiacum.      See  Mainz 
Monaco,  184 
Mont  Genevre  Pass,  the,   51,   57,  61, 

291 
Moselle,   the,   58,  241,  265,  286,  294, 

297-304 
Mucianus,    C.    Licinius,    21,   23,    129, 

137.  139.  141.  142,  146-153.  »S7. 

163,  179,  210,  211,  214,  290-291 
Munius  Lupercus,  253,  262,  289 
Mutina  (Modena),  71,  191 

Nabalia,  the  river  (?  Yssel),  314 

Nar,  the  river,  218,  219 

Narnia,  217-222 

"Nationalism,"  242-245,  320,  321 

Nava,  the  river  (Nahe),  296,  298 

Neckar,  the,  252,  332 

Nemetes,  the,  256 

Nero,   the  Emperor,    1-6,  n,    13,    14, 

19,  21-24,  34,  73,  131,  139,  145, 

242,  248,  317 
Nervii,  the,    237,  241,  244,  256,  257, 

305 

Nocera  Umbra  (Nuceria),  209 

Novaesium  (Neuss),  26,  254,  257,  269, 
271,  276-278,  284,  286,  288, 
306,  309,  311,  328 

Novara  (Novaria),  65 

Noviomagus  (?  Nymwegen),  310,  328 

Numisius  Rufus,  254 

Nymwegen  (Batavodurum  or  Novio- 
magus), 232,  262,  310,  328 

Ocriculum  (Otricoli),  222-224,  34^ 

Oglio,  the  river,  85,  86,  115,  194 

Opitergium  (Oderzo),  185 

Ostia,  34 

Otho,  Marcus  Salvius,  the  Emperor  : 
birth  and  earl}'  career,  13-15,  24 
principate,    15,    22-27,    30-32,    131, 

157 
campaign  against  the  Vitellians,  71, 

95"I24 
as  strategist,  70-73,  77-81,  100-114, 

116,  120-124 
death,  124-127 
character,  14,  107,  123,  127 

Padua  (Patavium),  170,  180,  185-188 
Pannonian  Alps.      See  Alps 
Parma,  191-192 


358 


INDEX 


Parthians,  the,  2,  145,  157 
Passau  (Castra  Batava),  28,  163 
Paulinus,  Pompeius,  236 
Petronius  Urbicus,  68,  69 
"  Peutinger  Table,"  the,  339 
Piadena,  119 
Pieve  Delmona,  115 
Pincian  Hill,  the,  225-226 
Pisa,  184,  348,  349 
Piso,  L.  Calpurnius,  12-15 
Pizzighettone,  113 

Placentia   (Piacenza),    39,    45-50,    56, 
70,   80-84,   87,   93,   96,    98,    102, 

113.  338-339 
Plutarch  : 

on  Otho,  126-127 

on  the  first  battle  of  Bedriacum,  341- 

345 
Po,  the,  39,  45-46,  49,  50,  102,  169 
Poetovio,  37,  136,  149,  154,  158 
Pomponius  Secundus,  P.,  251 
Pontic  fleet,  the,  146 
Poppaea  Sabina,  14,  15 
Portents,  belief  in,  140 
Praetorian  Guards,  the.     See  Army,  the 

Roman 
Proculus,  Licinius,  106,  116,  119,  121, 

128 
Provinces,  the  Roman  : 

Africa,  10,  22,  23,  65,  146,  166 
Britain,   2,  22,  24,  59,  96,  98,  133, 

206,  252,  314 
Dalmatia,  24,  37,  59,  154,  207 
Gallia  Belgica,  9,  26,  206,  240-242, 

250-251 
Gallia  Lugdunensis,  26 
Gallia  Narbonensis,  26,  70,  96 
Germany,  Lower,  9,  26,  250 
Germany,  Upper,  6,  9,  26,  250 
Mauretania,  25,  98 
Moesia,   28-32,   37,    141,    154,    162, 

163,  207,  292 
Noricum,  28,  29,  68,  69,  161,  162, 

187,  293 
Pannonia,  28-32,  37,  154,  161,  163, 

186 
Raetia,   28-29,  63-64,  68,  161,   162, 

187,  293 
Spain,   2,    5,    14,    24,    25,   97,    166, 

206,  207 
Syria,  21,  22,  141,  142 

Ravenna,  170-172,  179,  180,  183,  185, 

348-350 
Recruiting,  system  of,  17-18,  259,  323- 

329 
Remi,  the,  242,  244,  294 


Republicanism,  Roman,  n,  20,  23 

Rhine,    the,    7,    8,    28,    231-232,    251- 
252,  264-265,  269,  307 

Rhone,  the,  26,  59,  60,  242,  293 

Rieti  (Reate),  21 

Rigodulum  (Riol),  298-300 

Rome  : 

importance  in  the  civil  wars,  97,  151 

under  Galba,  6 

under  Otho,  30,  40 

under  Vitellius,  129-130,  215 

burning  of  the  Capitol,  224 

capture  by  the  Flavians,  225-229 

Rotterdam,  231,  232,  312 

Roxolani,  the,  31-32,  162,  163 

Sabinus,  Flavius,  220,  223-225 
Sabinus,   Julius,    247,    285,   294,  317- 

3i8 
Sacrovir,  Julius,  243,  247 
St.   Bernard  Pass,   Great,   58,   65,  66, 

68,  70,  291,  293 
St.  Bernard  Pass,  Little,  291,  293 
Salii,  the,  72 
Salonina,  83 
Saone,  the,  26,  59,  242 
Sardinia,  76-77 
Sarmatians,  the,  31-32,  162,  285,  292, 

327 
Savile,   Sir   Henry,    translation   of  the 

Histories,  74,  292 
Saxa  Rubra,  225 
Sea,   command  of  the.      See  Strategy 

and  Tactics 
Sequani,  the,  242-244,  294,  317,  320 
Sextilius  Felix,  162,  293,  295,  296 
Sido,  Prince  of  Suebi,  161,  162 
Silvanus,  M.  Pompeius,  154-155 
Singidunum  (Belgrade),  28 
Sohaemus,  Prince  of  Sophene,  143 
Spoleto  (Spoletium),  218 
Spurinna,  Vestricius,  70,  80-85,  95 
Stoichades     Islands    (lies     d'Hyeres), 

184,  350 
Strategy  and  Tactics  : 

ambushes  and  surprises,  87-92,  273- 

274-  3OI"304 
annihilation,    strategy    of,    154-165  ; 

and  see  Offence,  strategy  of 
artillery  and  siege-engines,  200,  203, 

272 
base  of  operations,  double,  47 
cavalry,  use  of,  88-89 
commander  -  in  -  chief,     position     in 

battle,  108-109 
commissariat     and     supplies,     211, 

212,  269,  277-278 


INDEX 


359 


communications,  line  of,  47-49,  61, 

71,  78,  84,  113,  178,  185 
concentration  on  the  field  of  battle,  55 
councils  of  war,  99 
defence,  strategy  of,  42,   53-55,   96- 

100,  148-153.  I73"l84 
defence,  tactical,  169-173,  178-179 
double  objective,  strategy  of,  148 
envelopment,    strategy  of,   53,    100- 

114,  120-124 
exhaustion,  strategy  of,  148-153,  164 
flank    marches,    risks    of,    102-105, 

110-112,  190-191 
fortresses,  use  of,  266,  283-284 
front    to    a    flank,    tactical    position 

of,  48-49,  178 
Initiative,  value  of  strategical,  38,  152 
minor  operations  of  war,  51,  76-77 
obstacles,  strategical  : 

mountains,  42-45,  52,  219 
rivers,  50 
offence,   strategy  of,  38,  52,  54,  55, 

146-153,  i73"l84 
outposts  and  sentries,  171 
penetration,  strategy  of,  38,  56,  66- 

69,  92-95 
penetration,  tactical,  100,  120 
pivot  of  manoeuvre,  a,  45 
pursuit,  91,  92,  280-281 
rearguard  action,  281-282 
reconnaissance  in  force,  82-83 
rivers,  defence  of,  50 
scope  of  strategy,  53 
sea,  command  of  the,  50-52,  61-62, 

75-79-  149.  IS2-IS3.  i59-i6o 
speed,  importance  of,  55,  63,  72,  79 
surprises.     See   Ambushes   and  sur- 
prises 
tactics,  relation  of  strategy  to,  91,  99 
Suebi,  the,  161,  162 
Suetonius  the  biographer,  228 
Suetonius  Paulinus,  71,  86,  88-92,  95- 

100,    106,    116,    119,  121,  128, 

282 
Sugambri,  the,  270,  330 
Syria.     See  Provinces,  the  Roman 
Syrian  legions,  the,  141,  142 

Tacitus  : 

characteristics,  168,  173,  206,  286, 
289,  348-350 

as  military  historian,  69,  76-77-  82, 
90-92,  99,  107,  115,  121-124, 
161,  191-192,  278-281,  289, 
344,  346-347  ;  and  see  Preface 

quoted,  3,  7,  15,  30,  62,  74,  119, 
132,  138,  189,  203,  215,  217 


Tampius  Flavianus,  L.,  154-155 

Tarracina,  216 

Tartaro,  the  river,  169,  172,  173,  181- 

183,  186,  190,  192 
Taunus,  232,  234,  239,  251 
Tencteri,  the,  239,  300,  302,  320 
Theiss,  the  river,  161,  162 
Tiber  flood,  the,  72 
Tiberius,    the   Emperor,   3,   9,   10,  19, 

235.  243.  270,  317 
Ticinum  (Pavia),  62,  82,  92 
Tiridates  of  Armenia,  145 
Titianus,  Salvius,   95,    106,    116,  119, 

121,  129 
Titus,  the  Emperor,  22,  23,  139,  144, 

M5 
Tolbiacum  (Ziilpich),  304-305 
Tongres,  240 
Toul  (Tullum),  58 
Trajan,  the  Emperor,  20,  162 
Treveri,   the,    58,   241,   243-245,   257, 

285,  293-305 
Treves,  58,  241,  289,  297-304,  317 
Triboci,  the,  295 

Tungri,  the,  240-241,  245,  285,  305,  330 
Turin  (Augusta  Taurinorum),  61,  133, 

134,  258 
Tutor,  Julius,  247,  285-288,  295,  296, 

298,  300,  304,  317 

Ubii,    the,    240,    270,    285,    300,    302, 

320,  330 
Urbinum,  184,  350 
Usipi,  the,  239,  284 

Vada,  310 

Vahalis,   the  river   (Waal),    231,   232, 

308-310,  312,  313 
Valence  (Valentia),  60 
Valens,  Fabius  : 

birth  and  early  career,  1 1 
invasion  of  Italy,  37-62,  74-75,  92 
first  campaign  of  Bedriacum,  92-95, 

117-120 
under  Vitellius,  128,  137 
second  campaign  of  Bedriacum,  167, 
172-173,    180,    183-184,    211, 
213.  348,  35o 
death,  222,  350 
character,  60,  93,  180 
Valentinus,  Julius,  247,  297,  298,  317 
Valerius  Festus,  166 
Vangiones,  the,  295,  330 
Vatican,  the,  130 
Vedius  Aquila,  95 

Veleda  the  prophetess,  250,  289,  312, 
316,  352 


;6o 


INDEX 


Vercellae  (Vercelli),  65 
Vercingetorix,  242,  317 
Verginius  Rufus,  L. ,  6,  9,  10,  243 
Verona,  39,  46,  66,  67,  154,  169,  178, 

187-192,  212 
Vesontio  (Besancon),  243,  293 
Vespasian,  the  Emperor,  6,  20-22,  64, 

129,    137-153'    214,    230,    248, 

252,   259,   264,   290,  317,  318, 

326-329,  332 
as  strategist,  147-153 
character,  22,  138-140,  230 
Vetera,  Castra,  26,  239,  244,  249,  253, 

254.  2S7.    262-289,   306-308, 
325.  327.  328,  3S°-352 

Via  Egnatia,  147 

Via  Flaminia,   72,   209-210,  214,  218, 

222-223,  348 
Via  Postumia,  87,  102,  192,  194,  198, 

199.  339 
Via  Salaria,  210,  223,  226 
Vicenza  (Vicetia),  11,  188 
Vienne,  27,  60,  133 
Vindex,  the  revolt  of,  5-6,  10,  27,  34, 

242-243,  248-249 
Vindonissa  (Windisch),  26,  28,  38,  62, 

255,  290,  293-295,  328,  331 


"  Vindonissa  Column,"   inarch  of  the, 

40,  41,  57,  62-70 
Vitellius,  the  Emperor  : 

birth  and  early  career,  9-1 1,  65 

in  Germany,  11,  12,  15 

invasion  of  Italy,  24,  36,  57-58,  99, 

112,  120,  128-129 
in  Rome,  129-132 
defensive  measures,    165-166,    212 

218,  258,  261 
death,  228-229 

character,  11,  58,  120,  129,  131-132, 
208,    212-216,   220,   221,  229- 
230 
Vitellius,  Lucius,  215,  217,  230 
Vocetius,  Mons,  64 
Vocula,  Dillius,  255,  268-284,  3°7 

death,  284-288,  290,  350-352 
Vologeses  of  Parthia,  145 
Volusius,  C. ,  203 

Xanten.     See  Vetera,  Castra 

Yssel,  the  river,  237 

Zuider  Zee,  the,  232-234,  237 


«.*> 


THE   END 


Printed  by  R.  &  R.  Clark,  Limited,  Edinburgh. 


Hende:  Llia 

Gi       and  rebellio: 

K5 
.2 


PLEASE  DO  NOT  REMOVE 
CARDS  OR  SLIPS  FROM  THIS  POCKET 

UNIVERSITY  OF  TORONTO  LIBRARY 


■MitJWOWUBlW 

8gBi«iilaHB8i§WBi