Skip to main content

Full text of "The Use of the Adjective as a Substantive in the De Rerum Natura of Lucretius"

See other formats


This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http: //books .google .com/I 


a ( 



^ 7 ^/ 


<: :":■■ 

/ •■ 

,; -- 




• • V . 



i ,\. 

If 7 ti,£ 

y ■' 



\ ■- 

:i -^ 











MACMILLAN & CO., Limited 










(Jlew ©orft 



All rights reserved 



' i 



5 ^7 


^ The transfer of a word ordinarily used as an adjective to 

r^ the function of a substantive, involves one of the most com- 
Q mon shifts of category that occur in language, and is abun- 
dantly illustrated by all the Indo-European languages. So 
among the Romans, as new substantive concepts arose new 
expressions developed for them. Of the various new forms 
that thus arose the substantivized adjective is one of the 
most important and most interesting, ,whether viewed from 
the point of view of general linguistics or treated as a factor 
in the historical development of the Latin language. 

The use of the Latin adjective as a substantive has been 
made the subject of a great deal of study. Those who de- 
voted attention to the subject in the earlier part of the past 
century approached it with minds seriously prejudiced in 
two respects. First, as they were interested in it mainly for 
the light it threw on " good usage ^^ and " bad usage," they 
looked at the phenomenon mainly from the point of view of 
the stylist. In the second place they were still largely 
swayed by the " logical " method of language study, which 
had sharply and clearly defined the respective provinces of 
the adjective and the substantive, and they accordingly 
viewed with disapproval any ^^ intrusion" of the former 
upon the field of the latter. It was an offense against the 
principles of grammar as well as a violation of good style. 

* Additions by the editor are enclosed in single quotation 
marks reversed, thus: , * 


O ^^ o 


180 University of Michigan Studies 

Even a man of the type of Christian Karl Eeisig makes the 
following remarkable statement : ^ " Es hat die Spraehe eine 
Menge Bedensarten, wo das neutnim adjectivi fiir ein sub- 
stantivum gesetzt ist, erst almahlich gebildet. Zuerst war 
dies mehr dichterisch; seit dem ersten Jahrhnndert der 
Kaiser warde es in der Prose immer allgemeiner; z. B. in 
levi habendum bei Tac. Ann. 3,54, primas dominandi spes 
in arduo {esse) Ann. 4,7. Auch friiher schon hat Sallust, 
der iiberhaupt manches Dichterische anwendet, in incerto 
Cat. 41,1.^' 

In 1837 C. G. Dietrich published a brief paper in the 
Zeitschrift f. Alterthxmiswissenschaft, Nr. 44, pp. 367ff., 
and treated the same subject in greater detail in the Easter 
program of the gymnasium at Freiburg in 1842. This 
paper was reprinted in N'eue Jahrb. f . Phil. u. Paed. suppl. 
vol. 8, pp. 487-503 (= Archiv. f. Phil. u. Paed.). While 
Dietrich still viewed the question with the eyes of a stylist 
and bases his conclusions almost exclusively on the usage of 
Cicero (he cites scarcely more than half a dozen passages 
from Sallust, N'epos and Livy), yet to him is due the credit 
of having pointed out the frequency of this usage, which 
others had regarded as confined to a comparatively small 
number of words and to a few special phrases. He took a 
decided stand : '' omnia fere (sc. adiediva) pro substantivis 
usurpari posse existimarem, si quidem ex ipsa verhorum 
compositione satis intelligitur adjectiva habere vim stib- 
stantivorumf During the succeeding thirty years Nagels- 
bach's Stilistik (Ist ed. 1846, 3d 1858, 4th 1861), Holtze, 
Syntaxis priscorum scriptorum Latinorum usque ad Teren- 
tium, 1861, 1862 and Draeger^s Historische Syntax der lat. 

^Vorlesungen fiber lat. Sprachwissenschaft (first published 
in 1839 by his pupil Fr. Haase and re-edited in the 80's by 
Heerdegen, Schmalz and Landgraf ) , vol. 3, pp. 159ff . 

Use of the Adjective as a Substantive 181 

Spr. vol. 1, 1872 (2d ed. 1878) added considerable new ma- 
terial from other writers than Cicero and introduced a more 
elaborate classification of the material. 

In 1874 appeared two works which made substantial con- 
tributions to the subject: Haase^ Vorlesungen iiber lat. 
Sprachwissenschaft, vol. 1, ed. by Eckstein and Ott, Die 
Substantivierung des lat. Adjectivum durch Ellipse, Pro- 
gram, Eottweil. The former dealt a heavy (and final) 
blow to those who had made extravagant use of the ellipsis 
as an explanation of origin of substantivized adjectives, 
while Ott pointed out a large group of instances, in the case 
of which one is justified in assuming that an ellipsis has 
been involved in the development of the usage. Ten years 
later Panhoff,^ Barth/ and Wueseke ' enriched the available 
material by somewhat exhaustive papers on the usage of 
Tacitus, Terence and Plautus, and in 1890 Hirt added the 
material supplied by Quintilian.* The special line of work 
opened up by Ott in the above-mentioned program was fol- 
lowed out by Wolfflin, Die Ellipse von navis'^ and Eolfe, 
Die Ellipse von ars * and The Formation of Latin Substan- 
tives from Geographical Adjectives by Ellipsis ^ in the thor- 
ough and exhaustive maimer characteristic of the school of 

^Panhoff, De neutrius generis adiectivorum substantive usu 
apud Taciturn, Diss. Halle, 1883. 

^Barth, Die Bleganz des Terentius im Gebrauch des Adjecti- 
vums, in Jahr. Class. Phil., vol. 129 (1884), pp. 177-182. 

•Wueseke, De Plauti et Terentii usu, adjectiva et participia 
substantive ponendi, Diss. Marburg, 1884. 

* Hirt, Ueber die Substantivierung des Adjectivums bei Quin- 
tilian. Program des Sophiengymnasiums, Berlin, 1890. 

"Archiv. Lat. Lex. vol. 9 (1896), pp. 285-291. 

•Archiv. Lat. Lex. vol. 10 (1898), pp. 229-246. 

^ Trans. Amer. Phil. Assoc, vol. 30 (1899), pp. 5-23. 

182 Univeesity of Michigan Studies 

The aim of the present paper is to throw additional light 
upon the subject by the examination of the usage of Lu- 

Since the publication of Ott's paper it has been customary 
to distinguish rather sharply between two processes by 
which substantives are developed from adjectives. Ott 
(p. 1) defines them as follows: "Die Substanvierung des 
lateinischen Adjectivs voUzieht zich auf einem doppelten 
Wege: entweder durch unbewusste Subsumption 
eines persohlichen oder sachlichen (abstracten) Begriffes 
unter einem obersten Allgemeinbegriff oder durch fiihlbare 
Ellipse eines ebenfalls generellen, aber enger begrenzten 
Begriffes von weit uberwiegend concreter Natur/^ e. g. first 
type : consvlaris ^ an exconsul/ honi ^ the good/ docta ^ a 
lady of culture/ honestum ^integrity.' .... "Der Her- 
gang bei der erster Art Substantivierung ist dem nach ein 
innerer, in den Elementen des Wortes selbst gelegener. 
Diese Elementen sind (1) die im Stamm ruhende Bedeu- 
tung des Wortes, (2) bei abgeleiteten Adjectiven zugleich 
die zum Ausdruck bestimmter Verhaltnisse dienenden Suf- 
fixe, (3) die Geschlechtsbezeichnung .... Was nun die 
zweite Art der Substantivierung betrifft, so ist hier der 
Hergang ausserlich, nicht in Elementen des Adjectivum 
selbst begriindet, er besteht, namlich in dem Wegfall eines 
allgemeinen Substantivbegriffes, der sich zum Adjectivbe- 
griff verhalt wie das Genus zur Species.^' 

,This division of substantivized adjectives into two 
classes, while it is in some respects convenient for the pur- 
poses of systematic classification and description, has been 
rather too sharply drawn heretofore, and has tended to ob- 
scure at least partially the real nature of the processes in- 
volved in the change. In the first place we should never 
lose sight of the fact, that the grammatical function of a 
word is only one of the many elements or groups of ele- 


Use op the Adjective as a Substantive 183 

ments of thought and feeling that constitute the complex 
meaning it bears in any given sentence in which it may 
occur. For example, in handling the problem now before 
us, we should distiixguish carefully between the processes 
involved in the shift of grammatical category (from adjec- 
tive to substantive) on the one hand and those involved in 
a change of what is ordinarily called the " meaning '^ ^ of a 
word on the other. While the two are concomitant in proba- 
bly all cases, yet the one is not an indispensable prerequisite 
to the existence of the other. The word molaris from con- 
tinued use in connection with dens " absorbed ^^ a part of 
its meaning, that is, certain elements in the meaning of 
dens became closely associated with the phonetic symbol 
molaris, a semantic change of very common occurrence and 
variously called VerdicMung, contagion, fusion or satura- 
tion,* itself a process involving several successive stages. 
We should be much in error, however, if we should suppose 
that it was through association with the word dens alone 
that molaris came to be associated with the objects with 
which English molar is associated when used of the teeth. 
In that case we should be overlooking the fundamental truth - 
that in each and every instance the meaning of a word has 
its origin in the sentence as a whole, or even in the wider 
context, and that not simply one element of the sentence 
(e.g. in this case the word dens) is responsible for its mean- 
ing. Take for example Juvenal, Sat. 13,213. 

Perpetua anxietas nee mensae tempore cessat 
Faucibus ut morbo siccis interque molares 
Difficili crescente clbo. 

^The shift of grammatical category is of course in this in- 
stance fundamentally a change in meaning. 

'See Wundt, Volkerpsychologle, 1,2, chap. 8,§5a (=pp. 
537ff. 1st ed.); Br^al, Essai de semantique, p. 221 (^pp. 
200fF. Engl, translation) ; Darmesteter, La vie des mots, p. 
124, 2d ed. 

184 University of Michigan Studies 

It is by the general meaning of the sentence (note the words 
mensae, faucibus, crescente cibo) that the meaning of mo- 
lares is determined. This is made clear by a comparison 
of Pliny, Nat Hist. 36A74f. 

Calcem e vario lapide Cato censorius inprobat; ex albo melior. 
.... Utilior eadem effosso lapide quam ex ripis fluminum 
collecto, utilior e molari, quia est quaedam pinguior natura 
eius; 36,137 Molarem quidam pyriten vocant, quoniam pluri- 
mum sit ignis illi . . . .; Celsus, 5,28 lapide molari contrito; 
Virgil, Aen. 8,249f. (description of the slaying of Cacus by 

Desuper Alcides telis premit (sc. Cacum) omniaque 

Advocat et ramis vastisque molaribus instat, 

'huge rocks'; cf. Ovid, Met. 3,59f. (description of the 

slaying of the dragon by Cadmus) 

Dixit, dextraque molarem 
Sustulit et magnum magno conamine misit; 

and Apuleius, Met. 7,17, where the meaning of molares cir- 
cuitus is made clear by the words of chap. 15 mulier molae 
machinariae subiugum me dedit and mercenariis discursi- 
biLS meis. 

In these six passages molaris is used in at least five dis- 
tinctly different senses. Four different meanings of the 
word are known as applied to stones; i. e. ^mill stone,' any 
^ large stone/ ^ flint/ ^ marcasite/ Clearly association with 
the word lapis alone would not explain these various mean- 
ings. Once the word has come to be closely associated with 
the groups of qualities possessed by the various objects sym- 
bolized by it in the above sentences, it may be used either 
as an adjective or substantive, just as the Romans said 
either homines consvlares ' men of consular rank/ ^ excon- 
suls ^ or considares. The shift from the former meaning to 
the latter (i. e. the shift of grammatical category) involves 
a different problem from that discussed above and appears 

• • • • 

Use of the Adjective as a Substantive 185 

to be essentially the same in both the classes of words de- 
scribed by Ott. In any case it can not be said that the sub- 
stantivizing is the result of the omission of the substantive, 
any more than that the omission of the substantive is due to 
the substantivizing of the adjective. 

In the statement of Ott as to the manner in which consvr 
laris, boni, docta, and honestum take on substantive mean- 
ing he leaves out of account in the same manner the part 
played by the context in the problem. The " elements of 
the word itself '^ are not more important than the other ele- 
ments of the unit of thought (sentence) of which it is a 
part. The conditions under which a word of this class 
appears with a specific " meaning ^^ are only slightly dif- 
ferent from those under which molaris and words of its 
class take on theirs. In the latter class some specific word 
as denSj lapis plays a large part^ in the former class this is 
not so likely to be true. The shift from adjective to sub- 
stantive is not conditioned by this change of meaning, but 
is a concomitant process, which may or may not take place. 

Professor Eolf e appears to have recognized the real error 
of Ott^s method, for he says in Trans. Amer. Phil. Assoc, 
vol. 30, p. 6, " To ellipsis clearness is sometimes apparently 

sacrificed With argentaria, for example, we may 

think of the mine (fodina), of the workshop {officina or 
tdberna)y of the bank (mensa), or of the banking business 
(ars) . The sacrifice of clearness is, however, only apparent, 
since the particular substantive which is to be supplied in 
each case is plainly indicated by the situation in which the 
word is used." This apparent ambiguity, or to speak more 
accurately, this variety of meaning exists, of course only in 
the mind of the grammarian or lexicographer who is study- 
ing the word. In the mind of the Eoman who was employ- 
ing the word for the purpose of expressing his ideas, nor- 
mally there was in each specific instance but one meaning; 

186 Univbrsity op Michigan Studies 

and it is only with the mental processes involved in that one 
meaning that the student of Latin syntax and semantics 
has to do/ 

In the present paper in dealing with the adjective con- 
verted into a substantive as found in the De rerum natura 
of Lucretius the subject is treated for the sake of conven- 
ience under the two heads mentioned above. For the first 
mentioned class, that is, " without ellipsis," I follow the 
classification of Nagelsbach amplifying it where necessary. 
In the second division I have made such a classification as 
seemed appropriate. 


Adjectives by inflection in gender and number, in con- 
nection with the context in which they stand, denote, in the 
masculine, males or animate beings (especially persons) 
generally; in the feminine, females; in the neuter, things. 
Such adjectives acquired the force of substantives (e. g. 
amicus, immicus, bonum, docta) and did not derive their 
meaning chiefly from an omitted substantive, the morpho- 
logical elements and the situation furnishing the conditions 
necessary to make evident the specific meaning and the sub- 
stantive character. When such a usage comes to be the pre- 
vailing or exclusive one, the adjective becomes a substan- 

There are various stages in this transition from one usage 
to the other, from that in which the substantive use is sug- 
gested by a real substantive or some other word standing in 
close proximity to the adjective, while, in the common 
usage, the adjective retains its full adjectival force, to that 
in which the adjective has become a real substantive and is 
only sporadically or never used with its adjectival force. 

^ See Rolfe, Trans, p. 7. 

•• • 



•-: ••. : : 

•! • • • • 

• • • • 

•Use of the Adjective as a Substantive 187 

To illustrate these facts in a concrete form I will take the 
adjective pyblicasrarum. In the early language these 
forms were prevailingly, if not exclusively, used with sub- 
stantives of the masculine, feminine and neuter genders to 
designate males, females, and objects of the neuter gender 
respectively as possessing the attributes signified by the ad- 
jectives. Thus, while publicus, because of its connotation 
of sex might be used of any public man, that is, an officer, 
a magistrate or a slave, the adjective was used to such an 
extent in the phrase, publicus servus, that, when a Soman 
said publicus in a certain context, a friend understood what 
he meant even if he failed to add servus. This usage con- 
tinued, the word servus being dropped more and more fre- 
quently, until finally, publicus came to mean ^ public slave/ 
[In Plautus, True. 557 metuit publicos, although the pas- 
sage is a disputed one, we probably have publicus used with 
the meaning of ^public officer^ or ^policeman.* Orelli, 
Inscr. 3,6,7,3 publicus sacerdotalis, a sacris, a sacrario dim 
Augusti gives publicus as meaning a lower servant of a 
priestly college.] 

N"ow just as publicus conveyed the proper meaning by its 
masculine termination supported by the situation, so pub- 
lica with a feminine ending is seen to have been used as a 
substantive and to have been associated with the idea of a 
public woman. Seneca, Epist. 88,37 has Sappho publica 
fuerit where undoubtedly such is the correct interpretation. 
Publica in addition, however, through the association with 
the word via, on the subsequent ellipsis of that word, came 
to express the idea embraced in the combined adjective and 
substantive publica via; e. g. Gromatici Lat. p. 334,16. This 
ellipsis becomes intelligible to us when we know the environ- 
ment of the substantive. 

Publicum when used as a substantive does not indicate so 
clearly as in the other instances the idea by means of its 

188 TJxiTEBSiTT OP Michigan Studies 

inherent chaiaeieristics. Here the meaning generally de- 
pends more npon the situation. So in some cases ierriio- 
rium, agrum sure the words (or concepts) which must have 
heen in the mind of the speaker, in others vectigalia, tonum 
etc. for example, Cic. Agr. 2,82 in publicum Campanum 
where publicum is equivalent in meaning to the state's 
purse, expense. In Cic. Verr. 3,105 Apronium .... im- 
perasse, ut in medio faro sibi ledi stemerentur, cotidie 
solitum esse non m^do in publico sed etiam. de publico con" 
vivari; we have publico in two senses, hut each equally in- 
telligihle thru the relation of the word to the rest of the 

The earliest example of publicum in the sense of ' a pub- 
lic place,' ' publicity,' is S. C. de Bacchan. Seve in popli- 
cod neve in preivatod. 

1. Neuter Singular. 
The neuter singular of the adjective is used as a substan- 
tive in all cases except the vocative. In the nominative and 
accusative cases such substantives are used, for the most 
part, in scientific writings. Nagelsbach attributes this to 
the evident influence of the Greek mode of expression. So 
many words of abstract meaning occur in this group, that I 
feel justified in calling especial attention to them in my 
classification. N'agelsbach notes among other examples 
alium, inane ("vom Lucr. aufgebracht"), beatum, dHud- 
dum, breve, prohabUe, illustre, suave, omne, cerium, fai- 
sum, honestum, verum, magnum. From this list I would 
omit inane. This is undoubtedly sometimes used in an ab- 
stract sense, but I think not in Lucretius. The word will 
be discussed in full, however, under ellipsis (Class B.) 
NTagelsbach does not call attention to the fact that the above 
words express abstract ideas, but Schmalz in Ms Lateinische 
Stilistik, p. 434, says, " Durch das Xeutrum werden all- 

Use of the Adjective as a Substantive 189 

gemein sachliche, zumeist abstrakte Verhaltnisse, z. B. 
honestum, iustum, immensum, und in Hural Dinge, die 
ihrem Wesen nach eine Eigenschaft besonders hervortreten 
lassen, z. B. digna, vera, suinma bezeichnet/* 

In studying the neuter singular as used in Lucretius I 
have found an abstract meaning in certain substantive ad- 
jectives (although various visual or auditory images may 
have been present in some of the cases cited) . A number of 
other adjectives of this gender and number imply the ellip- 
sis of a substantive. These will be treated under Class B. 

(1) I will first cite those examples in which an abstract 
idea is paramount. 

a. NTominative. 

2,1043 * . . . . perpende et, si tibi vera videntur, 
Dede manus, aut, si faUum est, accingere contra. 

b. Accusative. 

1,615 Praeterea nisi erit minimum, parvissima quaeque 

Corpora constabunt ex partibus infinitis; 
1,959 Namque extremum debebat habere; see also 1,752; 

960; 964; 4,266. 
1,409 Et verum protrahere Inde; also 4,794; 5,704. 
3,525 Ancipltique refutatur convincere falsum; also 

5,540; 4,764; 
3,800 Quippe etenim mortale aeterno iungere || Disi- 

4,477 duhium; 4,1119 malum; 5,958 tonum, (also 6,26). 

c. Genitive. The genitive case, especially the 
" partitive *' genitive, is often found, according to the gram- 
marians. Nagelsbach cites Cic. Verr. 4,12 nihil neque pri- 
vati neque pvhlid neque profani neqvs sacri. Other geni- 
tives are found even in prose writers, e. g. Cic. De nat. deor. 

^ 2,79 lex, quae est recti praeceptio 'praA)ique depuLsio, 

^ The passages quoted from Lucretius in the following pages 
conform to the text and numbering of Munro's fourth and 
revised edition, 1896. 

190 University of Michigan Studies 

In Lucretius I find the partitive genitive in 1,497 soUdi 
nil; 3,294 calidi plus; 3,915 Mali hoc; 5,176 quidve mali 
(also 6,29; 811) ; 6,663 Satis mali; 3,909 quid sit amari; 
4,1134 amari aliquit; 4,474 Veri nil; 5,168 quidve novi; 
5,172 nil acddit aegri. 

The examples of other genitives than the partitive are: 
3,1056 mali; 6,1178 requies m^li (Mali adopted by Brix 
from Maerobius, Sat. 6,2,13) ; 2,1052 Veri simile; 3,646 
mohilitate mali; 4,476 notitiam veri falsique (also 4,479). 

For totius see translations of Greek expressions, p. 19i, 

Nagelsbach states (p. 101) that the dative case furnished 
few examples of substantive-adjective usage, chiefly scien- 
tific terms. Lucretius furnishes only two datives of this 
sort : 3,804 .... quid dinjersiu^ esse putandumM 1 1 Quam 
mortaJe quod est immortali atque perenni \ | lunctum, 

d. Ablative. The ablative case furnishes com- 
paratively few examples. Nagelsbach cites recto recti/us, 
bono melius (Cicero), aequ/) et bono (Sallust). In the 
following examples of the ablative case the substantive ex- 
presses an abstract idea: 1,370 vero; 3,313 aequo; 3,953 
aequo; 3,800 aetemo; 4,477 certo (cited above) ; 4,557 
aequo. Perhaps we should place here 1,257 pingui (for 
MSS. pinguis)y although it would seem to mean ^fat' 
rather than ^ fatness.' 

In this connection mention should be made of quantum, 
tantum, tantundem, m/altum, common to all authors and 

At this point attention should be called to those substan- 
tives which have unquestionably been influenced by Greek 
usage, being translations of Greek expressions. As is well 
known,^ " Neuter adjectives and participles are freely em- 

^ Gildersleeve, Syntax of Classical Greek, part I, p. 13, §36. 

* » 4. - . 

Use of the Adjective as a Substantive 191 

ployed as substantives in almost any relations, t6 napt^riXvBos 
t6 ficXXov, TO rrapoy Dem. 18,19,2 ^ the past,* ^ the future,* ^ the 
present/ ** In t6 nav, * the whole,* ' the universe,* the sub- 
stantive character is shown at once by the article. This 
expression finds an exact counterpart, as far as meaning is 
concerned, in the omne of Lucretius in such phrases as 1,74 
omne immensum pergravit. In fact, it is highly probable 
that Lucretius used omne as a translation of the t6 nop of 
Epicurus. In the Epicurea of Ilsener p. 211, frg. 296 (= Plu- 
tarch, Adv. Coloten 13, p. 1114a) t6 nav and Tov.yrrn'TOi occur. 
A comparison of the Greek text with Cic. De div. 2,103 
will show the close relation existing between the Greek and 
Latin terms. In the passage cited above omne is used to 
convey the meaning of t6 nap Munro calls Lucretius 
l,958ff. Omne quad est, etc. almost a translation of Epicu- 
rus as given in Diog. 10,41. Here again t6 nav is rendered 
in Latin by omne. Omne is found in the De rerum natura 
at 1,521; 523; 975; 1024; 958; 967; 987; 2,305; 547; 
1049; 4,1620; 5,527; 530; 6,1121. On the analogy of 
oi(nne, or perhaps as an equivalent of the Greek t6 nhv^ Lu- 
cretius also used totiAis in 2,90, totum in 5,321. 

(2) We pass now to the use of the substantive in prepo- 
sitional phrases (see Nagelsbach, pp. 102ff.). This use was 
widespread in Latin literature, Cicero and Livy having used 
the phrase frequently. These prepositional phrases gen- 
erally denote relations of place. Sometimes they denote 
relations of time, while in some instances they metaphor- 
ically portray a condition or a situation as a place from 
which or to which something is taken. In all the phrases 
found in Lucretius the adjectives seem to have been closely 
associated with some substantive (subsequently passing out 
of use in the phrase), excepting in the temporal expression 
and also, perhaps, in the phrase per omne. 

192 IIniveesity op Michigan Studies 

Analogous to such a phrase as per omne are the meta- 
phorical phrases such as that in livy 4,43,3 ex trcmqvillo, 
where there is in all probability no ellipsis. Only three ex- 
amples are found in Lucretius exemplifying this use : 1,711 
Magno opere a vero longe derrasse videntur; cf. 1,758 
(a vero) ; 1,370 (a vero). This metaphorical relation is 
expressed also in similar prepositional phrases in which 
there is an ellipsis of a substantive, see pp. 201f. 

(3) Nagelsbach (p. 107) states that adjectives of the 
third declension are not used as substantives to any great 
extent. Most of the examples given below fall also under 
other special divisions, but for the sake of uniformity and 
of adherence to Nagelsbach^s classification I give them here 
also: 4,616 Hahent in se rationis plus operaeve; also in 
1,365 ; 366 ; 2,200 ; 3,1184; 294; 1,521 Omne foret soKdum; 
cf. references to omne just cited; 3,804 immortali atque 
perenni (cited above) ; 1,527 pingui (cited above) ; 5,1089 
Quanto mortolis m^gis aecumst tum potvisse (also in 3,800, 
etc.) ; 1,367 minm; 3,382 aegri (also in 5,172). 

Since animans, though originally a participle, is used 
only as an adjective or substantive, I place it here instead of 
under the division of participles. 

a. Nominative (singular number) : 2,573 cor- 
pus enim atque animans erit aer. 

b. Genitive: 3,388 animantis; also in 2,938; 
3,97; 388; 4,740; 859. 

The substantive adjective is accompanied in 3,97; 388; 
4,740 ; 859 by a modifying adjective or pronoun. 

c. Accusative: 

2,944 Praeterea quamvis animantem grandior ictus, 

Quam patitur natura, repente adfligit; see also 
2,669; 3,666. 

d. Ablative: 2,943. In 2,669 and 943 the 
substantive is modified by an adjectival word. 

Use of the Adjective as a Substantive 193 

The comparative and superlative degrees of the adjective 
used as a substantive may be thus grouped in Lucretius: 

plus 1,365; 366; 2,200; 3,294; 4,616; 1184; 
minus 1,367; extremum quod haberit 1,752; 1,959; 

960; 964; 4,266; minimum 1,615; 752. 

(4) Up to this point I have been treating of the adjec- 
tive substantive without special reference to words closely 
associated with it in the sentence. In the examples cited 
the adjective had taken on the characteristics of the sub- 
stantive, in other words it had received the rank of a sub- 
stantive. N'evertheless it had not, so to speak, as yet made 
use of all a substantive's "rights."^ This was the next 
step in the evolution, in consequence of which the adjective 
is found governing a genitive in the same way in which a 
substantive does, next a pronoun is found with the substan- 
tive and then even another adjective or participle. 

!N"agelsbach (p. 110) states that the use of the neuter 
singular with the genitive in Cicero and Caesar is very lim- 
ited, but is freer in Sallust, Livy and Curtius, such sub- 
stantivized neuters either expressing something concrete, 
e. g. commune, 'community,' as in Cic. Verr. 2,114, or 
superlative conceptions like extremumj vltimum, summum, 
plurimum. The usage is almost entirely confined to adjec- 
tives of the first and second declensions and to those which 
like multum denote grade, measure and part relations. 
Therefore most of the genitives are partitives. In Lucre- 
tius there are but few examples falling under this head. I 
have found only: 

1,1052 Illud in his rebus longe fuge credere, Memml, 
In medium summae^ quod dicunt, omnia niti. 

In the superlative: 1,959 extremum nuLlivs; 4,266 extre- 
mum saxi; the genitive with the comparative: 1,365 plus 

* Nagelsbach, p. 109. 

194 University of Michigan Studies 

inarm; 1,368 plus in se corporis esse; 1,367 vacui miwus; 
plus and the genitive also occur in 3,294; 5,616. 

Only two instances of the adjective used as a substantive 
modified by a pronoun were found : 3,285 ; quiddam unum; 
4,1119 malum id. Cf. Wiiseke op, cit. pp. 8 and 31 for the 
usage of Plautus and Terence. 

I pass now to that use in which the substantive takes an 
attributive adjective. This usage is common (Nagelsbach, 
p. 112). Aside from the technical term summum honum, 
I find also in Lucretius: 1,74 omne immensum; 2,1108 
magnum omne; 5,321 totwm, natvoum; 5,958 commune 

2. Neutee Plural. 

The adjective in the neuter plural is used as a substan- 
tive under three conditions: ^ (1) in a definitive sense, (2) 
expressing an abstract idea, (3) in cases in which there has 
been an ellipsis of a real substantive. The first two uses I 
will take up here. 

(1) The adjective-substantive used in a strictly definitive 
sense is found, for the most part, in the nominative and 
accusative cases. The forms occurring in Lucretius are: 
omnia {ex omnibus) 1,61, et passim; m/ulta 1,138 etc. etc.; 
cetera (passim); cuncta (12 cases); nulla 1,242; 2,680; 
pau/>a 2,20. 

(2) The abstract adjective substantives have been treated 
under the neuter singular. The neuter plurals occurring in 
De rerum natura are : 

a. Nominative and Accusative: 1,640 vera; 
1,700 vera ac falsa; 2,1042 vera; 4,481 veris, falsa; 2,793 
Candida; 2,867 munufesta; 3,464 delira; 3,734 mala; 4,1141 

^ Nftgelsbach does not make these divisions. 


Use op the Adjective as a Substantive 195 

mala; 3,957 praesentia; 4,324 splendida; 5,121 immortalia; 
6,1181 mira. 

b. Genitive: 3,1020 terminus malorum; cf. 
5,227 tantum m^orum, 

c. Ablative: 4,481 veris; 5,1114 validis et pul- 
chris; 6,1085 plenis. 

The last four examples and 1,172 and 661 ex omnibus 
(supra cit.) form a group in which the gender is not recog- 
nizable from the ending, but must be determined from the 
situation (Nagelsbach, p. 116). It must be said, however, 
that these are not common in the most careful writers. 
Repertis 5,2 (see below) and intolerabilihusque malis 
6,1158 should also be included in this group. 

Some adjective substantives which have a concrete ele- 
ment in their meaning are : 3,2 inlustrans commoda vitae; 
cf. 3,937 commoda; 4,1074 commoda; 4,504 manifesta; 
1,732 praeclara reperta; 5,2 hisque repertis^; cf. 5,13; 6,6 ; 
5,320 recipitque perempta; 6,1282 Multaque horrida. 

I give here two instances of the neuter plural compara- 
tive used alone : 1,828 plura; 6,245 plura. 

To illustrate the complete evolution of the adjective into 
the substantive the following examples are given, showing 
it in combination first with an adjective and then with a 
pronoun: 1,376 quamvis omnia sint plena; 1,1010 infimta 
omnia reddat; 3,734 mala mvlta; 3,937 omnia commoda; 
3,961 aliena omnia; 4,162 omnia plena (= 6,269; 1051) ; 
4,403 omnia tecta (=^ 6,575) ; 4,443 raraque nuhila por- 
ta/nt; 6,134 ramosa nuhila at que aspera; 5,13 divina an- 
tiqua reperta; 6,7 divina reperta; 5,94 tria talia texta; 
5,949 fluenta; luhrica; 6,527 cetera omnia; 731 nuhila 
omma; 1158 {supra cit.) ; 1282 {supra cit,). With a pro- 
noun: 3,945 eadem omnia (= 3,947) ; 4, mala haec. 

^ Conjecture of Ltambinus for MSS. maiestatis atque repertis. 

196 XJniveesity of Michigan Studies 

3. Masculine Plural. 

Nagelsbach (following Dietrich, op. cit.) states that in 
the masculine plural only those adjectives can be used as 
substantives which designate a class of individuals charac- 
terized by the quality suggested by the adjective ; e. g. Curt. 
8,17,4 militares = Skr. qatriyas, 

Wuseke op. dt. p. 45f. calls attention to the error of 
Dietrich and Nagelsbach in supposing that when the mas- 
culine and feminine adjective substantives are used in the 
plural, they necessarily refer to the whole class of individ- 
uals possessing the quality designated by the adjective. 
Wiiseke distinguishes three uses of these adjectives: (1) 
they refer to the entire class (" totum genus ") ; (2) They 
refer to two or more individuals, either (a) specific, definite 
individuals (certae ac definitae personae) or (b) indefinite 
(dubiae atque incertae) ; e. g. (1) ilium lauddbuni honi 
^die Guten,^ Plautus, Bacch. 397; (2) (a) ones nos uocant 
pessumae ' the wenches,' i. e. the two Bacchides, Bacch. 

It is clear that there is nothing in the adjective itself 
which restricts its application to any one of the three 
classes, the precise meaning being given by the context, 
when it exists at all. In the passage cited by Wuseke from 
Bacch. 397 there is nothing whatever to show that Plautus 
had in mind the " whole class '' of good men, rather than 
any good men. An author often added omnes when he re- 
ferred to the entire class. It should also be noted that ad- 
jectives, after coming to be associated with a class of in- 
dividuals sharing the quality designated by the adjective, 
may undergo a shift of meaning, inasmuch as other promi- 
nent qualities characteristic of that group may also pass 
into association with the given substantivized adjective. 
This is true for example of optimi, ' the aristocracy,' which 

Use op the Adjective as a Substantive 197 

becoming associated with this distinct political party, natu- 
rally connoted in specific instances other traits of the party 
than those usually implied in the adjective optinvus. M|ich 
depends also upon the previous experiences of the listener. 
For example periti militares (cited by Nagelsbach above) 
would mean a different thing to one who was already famil- 
iar with the Hindu caste system from what it would suggest 
to one whose only knowledge of it came from this particular 
passage. The absurdity of the view that only descriptive 
adjectives can be used substantively is disclosed by the ex- 
istence of such definitive adjective substantives as omnes 
and mvlti, which I have designedly omitted from the ex- 
amples cited above, because I considered the classification 
of Dietrich (cited above by Nagelsbach) too narrow. 

The treatment of Nagelsbach takes account only of de- 
scriptive adjectives. Various meanings are conveyed by 
these adjectives and consequently are expressed also in the 
substantives. Those found in Lucretius I have grouped 
under three heads: (1) Those that designate nationality, 
e. g. Orai, Chaldaei, Teucri, Troiani; (2) Those that desig- 
nate relations to other creatures, e. g. finitimi, similes, 
mares, consanguiruei, minor es; (3) Those that designate a 
quality of the object: (a) referring to some physical ap- 
pearance or state, e. g. squamigeri, arqiiati, lassi, sani, 
aegri, animantes, mortales; (b) referring to some mental 
characteristic or state: stolidi, crudeles, stvltorum, imbe- 
dllorum, indignos, puri, miseri; (c) referring to some 
activity: organici. 

The examples of such substantives used in Lucretius 
without a modifying adjective or pronoun are : 

a. Nominative : 1,641 omnia enim stolidi magis 
admirantur amantque. Other examples are animantes 
3,749 ; 1,808 ; 821 ; 4,645 ; 1101 ; 5,69 ; 80 ; arquati 4,333 ; 
caecigeni 2,741; extorres 3,50; Orai 3,100; 1,831; 2,629; 

* - * * #• 

** d 

198 University of Michigan Studies 

6,424; finitimi 4:,581; magni 1,741; mares 4,1224; miseri 
5,88; mortales 5,348; 6,51; nati 3,895; puri 4,1026; or- 
ganid 2,412; 5,334; similes 4,1211; 1218; sqiuimigeri 
1,378 ;vigiles 6,14:08. 

b. Genitive: 2,343 sqiunnigerum pecudes; also 
2,1083; 3,73 consanguineum mensas; 3,1023 stidtorum; 
4,1200 salientum; 5,727 Chaldaeum; 5,1023 ImhecUlorum 
esse aecum misererier omnis; 6,1245 lassorum vox; Graio- 
rum 1,136; Oraium veteres docti 2,600; Oraium poetae 
5,405; 6,754; maiorum 4,1226; gnatorum 6,13; animanr- 
tvm 1,194; 350; 1033; 1038; 2,78; 880; 921; 1063; 1071; 
3,266; 720; 5,431; 855; 919. 

c. Dative: 1,65 mortalibus; also 2,556; 1033; 
1158; 3,1074; 1078; 5,15; 1092; 1101; 1165; 6,10; 392; 
Teucris 1,469; Troianis 1,476; stoUdis 1,1068; %umams 
3,837; sanis 4,1075; miseris 4,1075; 5,983; maribus 
4,1198; immortalibus, beatis 5,165; marihus 5,853; aegris 
6,1152; animantibus 2,256;, 914, 3,417; 4,677; 6,773; 984. 

d. Accusative: 2,171 .... mortalis; also 
2,625; 3,778; 983; 5,1089; 1280; Graios 1,640; merentes 
2,1104; humanos 3,80; [organices] 3,132; vivos 4,38; qui- 
etos 5,168; agrestis 5,1383. 

e. Ablative: 2,919 mortalibus; also 5,205; 
gnatis 4,1256. 

Examples of definitive adjectives from Lucretius are: 
Omries 2,1029; 3,582; 1043; 4,708; 1049; 5,1023; omni- 
bus; 1,19; 2,836; 1,338; 3,971; 4,564; 708; 5,233; multi 
4,1015; 1018; 1020; 5,1158; 6,1174; multos 2,277; multo- 
rum 3,475. 

The masculine (and feminine) plural, like the neuter 
singular and plural of substantive adjective, are not infre- 
quently modified by participles or adjectives ; these are usu- 
ally numeral attributes. See Nagelsbach, pp. 121f. Aside 
from such adjectives (e. g. duo, omnes, multi, ceteri, pauci. 

to b b b 

_ / 

Use op the Adjective as a Substantive 199 

plurimi) Wiiseke pp. 49f. finds in Plautus and Terence 
only sontis reos Capt. 476, and inprobis vanidicis, Trin. 
273, the exact interpretation of both of which passages is 
doubtful. Lucretius shows a circumscribed, though some- 
what wider, rahge of usage, as can be seen from the follow- 
ing examples: 

With adjectives: 1,151 mortal is omms; 1,172 squrnni- 
geris mteniihus; 2,980 totis moHalihus; 4,1234 gnatis dul- 
cibvs; 5,944 miseris mortalibus (also 5,983) ; 6,1 mortdli' 
bus aegris. With pronouns: 6,1197 ndnoribu* nostris; 
6,1239 suos ad aegros; 6,1283 suos consangvineos. 

4. Masculine Singular, 

Lucretius used the masculine singular adjective as a sub- 
stantive very sparingly. Of those classes mentioned by 
Nagelsbach examples may be cited as follows : 

1. Where the substantive has a collective meaning: 
3,933 Quid tibi tanto operest, mortalis, quod nimis aegris 

Luctibus indulges? 4,1184 mortali; 3,775 immortali, 

2. Where the thought is directed to a single individual 
of a group, in which case one is opposed to the other or to 
others : 

5,1050 Cogere item pluris v/nus .... non poterat. 

3. Where the substantive designates an ideally conceived 
person : 

3,206 Qu>ae tibi cognita res vn muitis, o bone, rebus; cf. 
3,939 stuLte. 
The fourth and fifth classes of Nagelsbach, namely the 
substantivized adjective (4) in co-ordination with real sub- 
stantives, and (5) with the indefinite pronouns, do not seem 
to have been employed by Lucretius. There are, however, 
three more examples which belong in this general class, but 
which do not naturally fall under ^ any of the above head- 
ings: 3,10 aegri, 3,971 nulli; 5,173 tali. 

200 University of Michigan Studies 

5. Feminine. 

The feminine of the adjective is comparatively little used 
as a substantive in Latin. Neither Dietrich, Draeger nor 
Nagelsbach honor this gender with a separate classification. 
Its infrequency is, of course, due to the fact that outside of 
a few writers the bulk of Latin literature is of such a char- 
acter that words expressing the qualities of women would 
not naturally enter into it. If any proof of this were needed 
it would be afforded by the following list of substantive ad- 
jectives cited by Wiiseke from Plautus and Terence (the 
singular appears to be used somewhat more frequently than 
the plural) : alia (11 cases), hella, harda, bona (3 cases), 
ceterae, ignava, indocia, inepta, ingenua (2), inmemori, 
inproha (2), inp{r)udens, inopem, insipiens (2), insana 
(2), lepida (2), lepidissuma, lauta (2) libera, liberalis, 
mea (8), multae, mala (5), [misera"], nulla (2), neutram, 
nota, omnes (2), optuma, orba, paucae, pauper, peregrina, 
perita, pessuma (5), proba, pudica, sua scelesta (2), sicca, 
[su^ida^, stulia, tuu (3), timida (2), trivenefica, turpes, 
nostra (2), ulla, utraque (2), utravis, uuida. 

In Lucretius 4, 1151-1174 occur twenty-three such adjec- 
tive substantives: multae, pravas, turpis, nigra, immunda. 
fetida, caesia, nervosa, lignea, parvula, iota (?), magna, 
immanis, balba, muta, flagrans, odiosa, loquacula, [tumidd], 
mammosa, simula, labeosa, turpi, (compare the very similar 
passage in Horace, Serm. 1,3,43). Aside from these and 
the substantives classed in group B there occur diva and 
incluta (1,40-42), in which passage incluta may have ad- 
jectival force. 

With the exception of muta I find none of the above men- 
tioned substantives used as proper names. Muta was a god- 
dess, called also Lara or Tadta, whom Jupiter on account 
of her talkativeness struck dumb, Cf. Ov. Fasti. 2,583. 

Use of the Adjective as a Substantive 201 

6. Participles. 

Participles are used as substantives by Lucretius, as by 
other authors, and as such exhibit all the uses and appear 
in all the relations of other adjectives used as substantives. 

(1) Neuter {smgvlar and plural) : The neuter 
singular used as a substantive occurs from the earliest litera- 
ture on (see Wuseke, p. 31 ; Nagelsbach, p. 131), but I have 
found no instance of its use in Lucretius. In the neuter 
plural, however, there are several of the perfect participle.^ 

Nominative : Munita viai 3,498 ; saxorum structa 4,361 ; 
bene parta patrum 6,1129 ; teda 4,549 ; 575. Dative 

or ai)lative: dictis 1,28; 103; 126; 143; 267; 333; 401; 
418; 2,66; 987; 3,178; 902; 4,175; 592; 837; 5,50; 54; 
56 ; 99 ; 104 ; 113 ; 6,24 ; 42 ; factis 1,296 ; 3,897. Ac- 

cusative: 1,136 Oraiorum obscura reperta; reperta 1,732; 
aliorum antiqu4i reperta 5,13; divina reperta 6,7; dicta 
2,730; 3,12; 135; 4,461; 880; 914; 5,53; ahdita 6,809; 
clausa 1,354; deserta 1,164; 2,1102; culta 1,164; prompta 
6,817 ; Strata 4,415 ; structa 4,361 ; tecta 2,91 ; 1110 ; 4,403 ; 
430 ; 517 ; 5,984 ; 6,223 ; 261 ; 597 ; 1262 ; texta 4,743 ; 5,94 ; 
6,997; 1054. 

(2) Masculine Plural. 

In the plural of the present active participle all cases are 
freely used as substantives, even the nominative and 

(a) Nominative and accusative: medentes 1,936; reges 
rerumque potentis 2,50; 3,1027; saltantis 4,980. (b) 

Genitive: 1,318 dextras salutantum praeterque meantum; 
amantum 4,1077; 5,962; canentum 4,585; 5,1385; caren- 
turn 4,35; salientum 4,1200; venantum 4,991. (c) 

Dative: nascentibv^s 1,113; 3,671; opinantibus 5,1320; 
spirantibu^ 4,937; venientibus 5,1319; vigilantibus 5,1405. 

^ Fluenta (5,949) is an obscure formation. 

202 University of Michigan Studies 

(d) Accvsative: nocentes 2,1103; merentes 2,1104:; 
moventis 4,980 ; scUtantis 4,980 ; progredientis 5,1453. 

The perfect passive participle in Lucretius yields two ex- 
amples both in the accusative case (armatum 5,1297; 

The present active participle is used more than any other 

as a substantive in Latin. As stated above, however, this 

use is rare in the nominative and Nagelsbach even goes so 

far as to say that it is never so used in the classical writers, 

there being nothing, to hinder the participle referring to 

the preceding subject, and holds that Seneca was first to 

use it as a substantive employing it, e. g. in De Ira 1,4,1. 

Terence, Phormio 243 had already written Pericla, damna, 

ecdlia peregre rediens semper secum cogitet. Wiiseke, p. 44, 

cites also Amans from Plant. Pseud. 238; True. 26; 56; 

intellegens Eun. 232. There are at least two examples of 

the nominative in Lucretius : 

4,1024 Flumen item sitiens aut fontem propter amoenum 

4,1097 Ut bibere in somnis sitiens quom quaerit. 

In Cicero the ablative is not often used but the other cases 
are frequently. Other examples from Lucretius are : eunti 
3,524 ; cubanti 4,952 ; opvaanti 3,959. 

Lucretius affords no examples of the future participle 
with substantive force, since flexura (4,312), iunctura 
(4,1083), etc. are substantives belonging to the class nuiwra, 
cvltura, scriptwra, etc. formed with the sufl&x -teurra, 
-seiu-rd (cf. Hist. Gram. vol. 1 (Stolz), pp. 557f.). 


In this group the change in the meaning of the adjective 
is largely due not to any morphological elements in the 
word itself, but to the influence of a substantive which was 
subsequently omitted. On the nature of this change see 

Use op the Adjective as a Substantive 203 

above pp. 182ff. In some words the transfer of the meaning 
has been so completely effected that the substantive usage 
alone remains, the former adjective usage having become 
obsolete. Apropos the ellipsis involved in this usage, Ott 
(op. cit. pp. 2f.) says: "TJm die Sache an Beispielen klar 
zu machen, so steht es fiir mich fest, dass Substantive wie 
aerarium, apiarium, doUwrium, farinarium, farrarmm, fri- 
gidarium, {KaltTcammer) , granarium, mellarium, olearium, 
palearium, plumbarium, pomarium (Obstkammer), u. a. 
die Ellipse von horreum zu Grunde liegt, wenn ich auch 
nicht im Stande bin, ihre einstige Verbindung mit diesein 
Genusnamen nachzuweisen." ^Apropos of this statement 
three remarks may be made. First we should hesitate to 
trust our " feeling ^^ (implied in "es steht fiir mich fest'^) 
in dealing with any language except our own vernacular. 
Secondly the assumption of horreum may be too arbitrary. 
With aerarium for example, aedificvum or something simi- 
lar is more natural. Thirdly we must not overlook the fact 
that the suffix -arium was a productive suffix widely used to 
make substantives of this general character, even when a 
corresponding adjective ending in -arius did not exist.' 

When is an adjective to be regarded as a true substantive ? 
On this point Paul {op, cit. p. 298) says, " Sobald nun die 
Unterstutzung durch die Situation fiir das Verstandniss 
entbehrlich ist, so ist auch das Wort nicht mehr als ein 
Adjectiv zu betrachten, sondern als ein wirkliches Substan- 
tivum, und es kann dann von einer Ellipse in keinem Sinne 
mehr die Eede sein." To this Professor Rolfe ^ adds verv 
aptly, that even after an adjective has become a genuine 
substantive the original combination of adjective plus sub- 
stantive may nevertheless be used on stylistic or euphonic 
grounds just as in English we speak now of ' the Atlantic,^ 

* Trans. Amer. Phil. Assoc, vol. 31, p. 7. 

204 University of Michigan Studies 

now of Hhe Atlantic Ocean/ Other instances which I 
might mention are ^ to explore the Yellowstone ^ or ^ to ex- 
plore the Yellowstone Park/ * to ride on the Mississippi ^ or 
^ to ride on the Mississippi Eiver/ yet in such cases we must 
also raise the question as to whether in the fuller expres- 
sions ^Atlantic Ocean/ ^Yellowstone Park^ and the like 
a consciousness of the attributive character of the Atlantic, 
Yellowstone etc. is present or whether the words have so 
far coalesced as to form a single concept and thus become 
compound proper nouns, as for example * Lake Michigan/ 
To the Latinist the determination of such a question be- 
comes, of course extremely difficult in most cases, if not 
impossible. Thus in Lactantius, Inst. 5,1,^4 is found ex 
artis oratoriae professione although the word oratoria is 
used as a substantive by Quintil. 2,14,1 and even by Lac- 
tantius himself in Inst. 3,25,11 ne oratoria quidem igno- 
randa est^ In this connection attention should be called 
to the need of extraordinary caution in respect to the so- 
called nTTo Koivov construction, that is, when the substantive 
with reference to which the meaning of the so-called sub- 
stantive adjective is chiefly determined, is found not in the 
connection with such an adjective but in the more or less 
remote context. ,Thi8 case is clearly a special type of the 
general situation which Paul (p. 297) describes afi " Er- 
ganzung aus der Situation." ^ These cases must, of course, 
be carefully distinguished and excluded, 'atto koivov is also 
found in passages in which there is a general subject and 
words are used referring to the general subject at various 
places in the passage. In Pliny's Natural History whole 
chapters are devoted to special subjects, and when such a 
reference is made, the generic notion is suggested by the 
general subject of the chapter. It is very difficult and 

^ Cf . also Tac. Dial. 6,1 and 8,14 oratoria eloquentia. 

Use op the Adjective as a Substantive 205 

naturally it is often impossible to decide whether one has a 
case of true ellipsis or only the and koivov construction. 

A very remarkable instance of this is found in Lucre- 
tius 2,442-463. Beginning with line 60 the atoms have 
been under discussion. ,In lines 442-463 the expressions 
hamatis, ramosis, levibus, rotundis and perpleicis occur as 
designations of atoms. The fact that Lucretius in the course 
of book 2 uses a variety of general terms to designate his 
atoms (e. g. genitalia corpora (1. 63), exordia rerum (1. 
333), primordia (11. 379, 396), elementa (11. 393, 411, 
414), principia (1. 443), and uses in fact the last two at 
the beginning and at the close of this passage, would sug- 
gest that the poet had in mind throughout this passage no 
one particular word, with reference to which the under- 
scored words were used, but rather the picture or concept of 
the objects themselves.' 

I have divided these adjective substantives into three 
classes. In the first are included " true substantives '^ em- 
ployed without consciousness of ellipsis. The second and 
much larger class consists of adjectives substantivized in 
connection with an elided noun, yet sometimes used as a 
true adjective. This class is well illustrated by the word 
fera with bestia (?) omitted. This substantive has by long 
and popular use taken on all but exclusively the function 
of a substantive, yet we find in Cic. Lael. 21 Hoc apparet in 
bestiis, voJucribus, nantibus, agrestibus, dcwrribus, feris, 
the word feris has an adjectival force. 

The third division is that in which the omitted substan- 
tive is made evident only through the situation, i. e. the 
connection of the adjective used as a substantive with that 
particular portion of the context. A very good example is 
that of Lucretius 4,723 et unde quae veniunt veniant in 
mentem, percipe pauds. Here the omitted substantive is 

206 University of Michigan Studies 

plainly verbis yet nothing in the adjective itself shows this. 
It is, rather the whole situation which determines the 
omitted word. The word " positional " is used to describe 
this class. 

Group 1. True Substantives. 

Only one example of the first group is found in Lucre- 
tius and that is derived from a geographical proper 

3,382 nam neque pulveris interdum sentimus adhaesum 
corpore nee membris incussam sidere cretam 

Lucretius here uses creta ^ chalk ^ for creta terra. 

Professor Eolfe ^ says of this word ; " The original mean- 
ing of creta was so completely lost sight of that Pliny, Nat. 
hist. 33,163 speaks of cretam Eretriam exactly as we do of 
* Dresden china ^ — The word appears as a substantive in 
the earliest Latin, e. g. Plant. Aul. 709. The word seems 
not to occur in Greek, but Diosc. 5,171 has 'EfjfTfuiis (sc.y^ ). 
Plin. Nat. hist. 35,196 mentions Cimolia {Sarda, Umbrica, 
Thessalica) , etc.'^ The complete obliteration of the original 
idea in this adjective plus substantive is shown very clearly 
in Horace, Odes 1,36,10. Cre^sa ne careat pviohra dies nota. 
" The meaning is, * that the day may be a bright one in our 
memory,^ from the practice of recording especially happy 
days with a white mark and unhappy ones with black " ; * 
cf. Serm. 2,3,246 Sani ut creta, an carhone notanti? 

A word which comes very near this category is serpens, 
probably originally used with hestia, yet it is occasionally 
used as an adjective. 

The examples of serpens in Lucretius axe : 3,658 micanti 
serpentis cauda; 4,60 luhrica serpens; 4,638 serpens ipsa. 

* Op. cit. p. 8. 
" Smith, ad loc. 

Use of the Adjective as a Substantive 207 

Plautus has proserpens bestia in four instances: Asin. 695 ; 
Pers. 299 ; Stich. 722 ; Poen. 1034. Bestia is probably to 
be supplied with the examples given from Lucretius, al- 
though serpens is also used as a masculine in 5,38 where it 
refers to the dragon of the Hesperides, and it is used by 
other authors in the masculine as an appellative (see Virgil, 
Aen. 5,273). In such case perhaps draco is the omitted 
substantive (cf . Suetonius, Tiber 72 erat ei in delectamentis 
serpens draco. 

An adjective-substantive very freely used by Lucretius is 
inane with which I think spatium was originally used. 
That Lucretius himself felt the substantive is plainly evi- 
dent from the examples 1,527 Quae spatium pleno possint 
distingu^re inane; cf. 1,523. There is no instance in Lu- 
cretius of the use of inane in the purely abstract sense where 
the supplying of spatium would give an incorrect meaning 
to the passage. The instances of inane are so numerous 
that I cite them by figures only : 

(a) Nominative: 1,330; 420; 342; 480; 509; 511; 569; 
954; 1010; 1079; 2,236; 5,357; 365; 366. (b) Ac- 

cusative: 1,369; 382; 386; 399; 426; 439; 507; 514; 517; 
520; 536; 655; 658; 745; 843; 1074. (c) Genitive: 

1,365 pit/s manis, which shows an adjective-substantive of 
the third declension used with another of the same declen- 
sion, a very rare use, according to NSgelsbach. (d) 
Ablative: 1,526; 660; (v. 1. inane)-, 1,742; 1009; (v. 1. 
inane) ; 6,941. For inane in prepositional phrases cf. pp. 
210, 211. 

A word closely allied to inane is the technical term 
vacuum. This also became a substantive through the ellip- 
sis of spatium and the same phenomena are evident in the 
transformation. In a few places it retains its adjectival 
force as in 1,523. I give the examples from Lucretius ac- 
cording to case and use in the sentence. 

208 University op Michigan Studies 

Nominative: 1,393; 394. Oenitive: 1,367 vacui minus. 
Accusative: in prepositional phrases; in vacuum 6,1007; 
1014; 1017. 

Group 2. " Quasi-Substantives.^' 

The second division, intermediate between the extremes, 
includes those in which the omitted word is almost certainly 
known without the context. For all practical purposes these 
substantives, like the preceding, are true substantives, but 
as they are slightly over the boundary line I have thought 
best to make a separate division for them. Examples in 
point are our own words " right '^ and " left/' Latin dextera 
and laeva. With these two words m/inus is omitte'd, but 
they belong to the quasi-substantives because even without 
the context, the whole idea is conveyed with fairly reason- 
able certainty. To be sure laeva and dextra might, in an 
adjectival sense refer to something other than wunus, but 
in the ordinary usage the adjective plus the substantive idea 
is expressed with reasonable clearness by laeva and dextra 
alone. Examples: 3,649; 651; 5,1298. 

A substantive as frequently used by Lucretius as any 
other is fera with which hestia was originally used (cf. Cic. 
Lael. 21 cited above). 

The examples of fera in Lucretius are: Nomina- 

tive: 2,343 armentaferaeqwe; 2,922; 3,880; 4,1197; 5,228. 
Oenitive: 1,404 montivagae ferai; 163 genus omne 
ferarum; also 1,255; 2,539; 597; 598; 877; 995; 1076; 
1081; 1152; 3,753; 776; 872; 888; 4,413; 680; 686; 994; 
1264; 5,39; 201; 218; 932; 947; 967; 1059; 1338; 6,198; 
766. Dative: 5,991; Accusative: 2,604; 5,868. 

Words of this same type are volv^er, dies, quadrupes. 
With quadrupes I supply hestia or helua (cf. however, in 
Vergil quadrupedante used for horse) ; with ales, avis or 
hestia; with volucer, ales, avis or hestia. The examples in 

Use of the Adjective as a Substantive 209 

Lucretius are: 2,928 alituum; (also 5,801; 1039; 1078 
[6,818;] 821); 1,12 aeriae volucres; (also 1,162; 589 
2,145; 344; 3,880; 984; 993; 4,1007; 1197; 5,801; 825 
1078) ; 2,536 quadripedum in genere; also 4,1265; 5,1202 

An adjective-substantive made by the ellipsis of a noun, 
and frequently met in Lucretius, is summa. The word to 
be supplied here is, in my opinion, ratio (cf. Cic. De leg. 
1,18 lex est ratio summa, vnsita in natura, quae vubet ea, 
quae fadenda sunt, prohibetqu£, contraria; Ad Att. 8,11. 
D §5 ; and Lucretius 1,54 de summa caeli ratione, 

Summa is often used by Lucretius in a technical sense. 
I give first the examples in which this word is used alone ; 
second those in which another substantive modifies it; and 
third, those in which it is limited by an adjective either 
attributive or predicative. 

Summa alone. Nominative: 1,1045; 2,310; 5,194; 

330; 6,606. Genitive: 1,953; summae finis; 1,1053 

in medium summae. Accusative: 1,436; 636; 706; 

963; 1042; 2,513; 518; 527; 530; 5,368. Ablative: 

2,1054; 1077; 3,84; 514; 2,91 in mmm^, 

Summa with a modifying genitive: Rerum summa: 
1,756; 1008; 1028; 2,75; 5,237; 2,303; 649; 1,333. 
Summa loci: 2,1044. Sum^ma salutis: 2,863. 

Summarum summa: 5,361. The last example is particu- 
larly interesting. Immsnsi summum: 2,1095; 6,485. 
Summam materiai: 2,527. Ad maris s. 6,613. 

Summa with a modifying adjective: Summae totius 
1,988; incolumis summa 2,71; S. ulla 2,339; ad summam 
summai totius omnem 6,679 (cf. summarum s. above). 

Group 3. Positional Substantives. 

In this group the omitted substantive can be determined 
only by the situation, the environment of the substantive in 

210 University of Michigan Studies 

question. This class may be still further subdivided into 

(a) substantives which do not, in themselves, give a hint of 

the omitted word, and (b) those which barely suggest the i 

elided substantive, but do not afford sufficient evidence to 

determine it with certainty. 

Subdivision (a). | 

5,905 Qui fieri potuit, triplici cum corpora ut una, i 

Prima leo, postrema draco, media ipsa, chimaera \ 

Ore foras acrem flaret de oorpore flammam? 

Here the omitted substantive is plainly pars, but in other 
situations prima, postrema and media might refer to other \ 

objects or might have their adjectival force. 

3,522 medidna. On this word see Eolfe, Archiv. Lat. 
Lex. vol. 10 (1898), p. 235, who supplies ars rather than 
res; and compare C. I. L. VIII, 241 m^dica arte and Varro, 
De ling. lat. 1,593 ah arte medidna medicos dictus. 

The usage magni referre 2,894 seems to be phased on such 
expressions as parvi pretii, magni (pretii) aestimare, which 
are quite common. Other examples in Lucretius are 1,817 ; 
2,883; 894; 4,984; 1257; 1264; 5,545; always with referre; 
1,908 permagni referre. 

With pleraque in 1,1215-22 the omitted substantive ani- 
malia becomes evident only when we take the word pleraque 
in its environment; and even here it is quite unlikely that 
Lucretius had any verbal image of the word animalia in his 

Expressions denoting relations (chiefly local) are mostly 
prepositional phrases with the neuter singular adjective 
substantive in the ablative or accusative case. Lucretius 
follows the general Latin usage. He uses ad (in) imum, 
ah (ex) imo, in (per) medium, in (a) medio, in artum, 
ah (e) summo, ex {de) supero, in aperto, per inane, in 
inani, in arto. If there has been any ellipsis here, it was 


doubtless of locum, loco, though with inane the word spor 
Hum should rather be thought of. It is possible, however, 
that all these cases fall under class A above (p. 192). 

An interesting example is found in 6,62 : Rursus in anti- 
quas referuntvr religionis. 

The metaphorical meaning of the phrase becomes clear 
only when we have the context. I would supply here sen- 
tentias, the idea being of persons borne back into their old 

Expressions like dbrupta, ardua, summa are said (Drae- 
ger, op. cit. vol. 1, p. 50) to be due to the ellipsis of the word 
loca. If this view is correct the Lucretian representatives 
should find mention at this point. Unquestionably adjec- 
tives like the above occur as modifiers of loca, e. g. Liv. 
39,1,5 ardtui atque iniqua loca, Caes. Bell. Gall. 2,19,5 loca 
aperta, but that the substantive use of these plurdlia neuira 
developed out of the combination of adjective plus loca is 
not thereby proven. They may have developed like the 
adjective substantives in class A. Lucretius has the follow- 
ing examples: culia ac deserta 1>164; vitima naturae 
1,1116; summa atqvs ima 2,488; and the prepositional 
phrases : 1,223 per inania; 2,1102 in deserta recedens; 6,142 
per nubila (also 6,199) ; 4,74 de summis; 2,892 ex omnibus. 
In the last two cases the endings give no indication as to 
gender and we have to depend on the situation for the exact 

The examples of the substantive with a genitive to com- 
plete its meaning (see Nagelsbach, p. 114) : 1,354 clcuusa 
dom^rum; 4,612 clcuusa [domorum] ; 5,417 pontique pro- 
funda; 5,1374 olearum caeruia; 6,96 caerula caeli; 6,214 
nubila caeli (also 1,6; 278); 1,659 Ardua dum m&tuunt 
amittunt vera viai || MSS. ver, aula ||. In prepositional 
phrases: 1,340 per .... sublim/ique caeli; 1,1090 per 

212 . University of Michigan Studies 

caeli caerula; 2,115 per opaca domorum; 4,730 corporis 
.... per rara; 5,771 per caervJa mundi; 6,332 per rara 
viarum; 6,817 in apertum prom^ptaque caeli. 

The adjective-STibstaiitive modified by an attributive ad- 
jective or participle : 4,101 simili specie praedita rerum ex- 
tima 1 1 MSS. ex\\; 6,269 verdis atque igniius omnia plena 
sunt; 6,462 nubUa tenvia; 6,731 nubila omnia. 

To the above list of positional substantives we may ap- 
pend the following, which only provisionally reveal the 
omitted word through their inherent meaning: 2,369ff. 
Praeterea teneri trenvuiis cum vocihvs haedi comigeras no- 
runt matres agnique petvici halantum pecudes. If a sub- 
stantive haa been elided in connection with the substan- 
tive use of halantesj we should naturally think of oves (cf. 
Phaedrus 3,15,1 agno balanti) while avis or ales is sug- 
gested by 2,878 pennipatentum (cf. 4,1010 persectantes , 

With the expression ex infinite, while the idea of infinity 
is very evident, whether this infinity is of time or space can 
only be ascertained by a study of the connection of the 
phrase in its environment. The time element is in the as- 
cendency in 

2,265 Ex infinite (sc. tempore) ne causam causa sequatur; 
cf . 1,1025 ; 2,530. Lucretius also uses the full form of the 
expression, e. g. 2,574 ex infinite tempore; also in 1,550 ; 
578; 5,188; 316; 378; 423. 

On the other hand the spacial conception occurs with 
infinite in: 5,408; 1,1001; 1036; 5,367; 414. 

With salse I supply aequere in 5,1080 in salse \ \ salsis 
Lamb. 1 1^ on the basis of those passages in which Lucretius 
himself has used the fuller forms of expression; for exam- 
ple 3,493; 5,128; 6,634 all three of which show the ex^i 
pression aeqwore salse. 

Use of the Adjective as a Substantive 213 

Lucretius uses a few substantives in the neuter plural 
which have been formed from geographical names. For 
full discussion of such formation see Professor Eolf e^s ^^ The 
Formation of Latin Substantives from Geographical Ad- 
jectives by Ellipsis/^ already referred to. 

In each of the instances given below the word to be sup- 
plied is ascertained by the historical associations connected 
with the word itself and also by the situation in the text. 
For instance, we know that the Babylonians were famed for 
their textiles and that the Sicyonians were celebrated for 
the taste and skill displayed in the various articles of dress 
made by them, among which we find mention of a certain 
kind of shoe much prized in all parts of Greece. Cicero 
refers to such foot-wear in De oratore 1,231 with the words 
caiceos Sicyomos, The passages in Lucretius are: 

4,1125 Huic lenta et pulchra in pedibus Sicyonia rident. 
4,1029 Interdum in pallam atque AUdenaia Oiaque ver- 

Cum Babylonica (1. e. "coverlets") magnlfico 

splendore rigantur; 
4,1123 Labitur interea res et Babylonica fiunt. 


abdita (loca), Alidensia Ciaque (texta), altum (caelum, 
mare, locum), ales . (avis, bestia), amaracinum (unguen- 
tum), angustum (locum), antiquas (sententias), ardua 
(loca), apertum (locum, caelum), artum (locum). 
Babylonica (texta), balantum (ovium). Caerula 

(loca), cava (loca), Cia cf. Alidensia, clausa (loca), 
creta (terra), culta (loca). Deserta (loca), dextra 

(manus), [Extima] (loca). Fera (bestia). 

Infinitum (tempus, spatium), inane (spatium), inania 
(loca), ima (loca). Laeva (manus). Magni 

(pretii?), media (pars), medium (locum), medicina 

214 University of Michiqan Studies 

(ars), multa (verba). Nubila (loca). Omnia 

(loca), omnibus (elementis), opaca (loca). Parvum 

(argentnm), parvus (pner), paucnm (verbum), penni po- 
tens (ales, avis, bestia), persectans cf. pennipotens, 
planum (locum), pleraque (animalia), pluribus (verbis), 
postrema (pars), prima (pars), profunda (loca), profundi 
(spatii?), prompta (loca). Quadrupes (bestijt), 

quantum (pondus, spatium). Eara (loca). Sa- 

lientum (marum), salsum (aequor), serena (loca), serpens 
(bestia, draco), Sicyonia (calciamenta), sublima (loca), 
summa (loca), summa (ratio), superum (locum). 
Tantum (pondus, spatium). Ultima (loca), unum 

(locum). Vacans (spatium), vacuum (spatium), 

vera (loca), volans (ales, avis, bestia), volucer (ales, 


1. Lucretius did not use a neuter singular of the parti- 
ciple as a substantive, a use noted in other writers. 

2. He did not use the adjective as a substantive near or 
in the midst of real substantives or with indefinite pronouns. 

3. He used the nominative singular masculine present 
participle as a substantive, a use denied by Nagelsbach. 

4. He made sufficiently extensive use, at least in one sec- 
tion, of feminine adjective substantives to necessitate a 
special classification of them. 

5. His use of the future participle was limited.