Skip to main content

Full text of "The history of ethnological theory"

See other formats




. I^^' 

{ < 















To my Sister 
Bisa Lowie 


There are two histories of the entire science of man, 
but neither of them offers an adequate survey of eth- 
nological development. That justly esteemed veteran, 
Professor Alfred C. Haddon, has crowded into the 140 
small pages of his History of Anthropology (London, 
1934) an amazing number of names and dates connected 
with prehistory, physical anthropology, linguistics, and 
ethnology; but space limitations barred any attempt to 
trace the progress of thought. T. K. Penniman's A 
Hundred Years of Anthropology (London, 1935), on the 
other hand, gives a one-sided emphasis to biological 
problems and is highly capricious in its admission and 
rejection of matters pertaining to ethnology. 

The present treatise is explicitly devoted only to 
that part of anthropology (as the term is understood in 
English-speaking countries) which concerns culture. 
Within that sphere it attempts to indicate the course of 
theoretical progress ; but since theory must rest on fact, 
the growth of knowledge through the perfection of tech- 
niques for gathering information receives proper con- 

A glance at the Table of Contents suffices to show 
that there has been no undue emphasis on American 
theories; indeed, Morgan and Boas are the only writers 
to whom extended discussion has been granted. On the 
other hand, in illustrating points by special instances, 
the author has inevitably leaned heavily on American 
material, with which he happens to be most familiar. 



Bureau of American Ethnology — 




(Annual) Reports. 


Contributions to North American Ethnol- 


Columbia University, Contributions to An- 




Field Museum of Natural History — 






Publications, Anthropological Series. 


Internationales Archiv fiir Ethnographic. 


International Congress of Americanists 
(Comptes Rendus, Proceedings), 


International Journal of American Linguis- 


Journal of American Folk-Lore. 


Journal of the Royal Anthropological In- 


Museum of the American Indian, Heye 
Foundation — 




Indian Notes. 


Indian Notes and Monographs. 


Peabody Museum (of Harvard Univer- 








Public Museum (of the City) of Milwau- 
kee, Bulletin. 


Revue d 'Anthropologic. 


Societe des Americanistes de Paris, Journal. 


Smithsonian Institution — 


Annual Reports. 


Contributions to Knowledge. 


Miscellaneous Collections. 


University of California, Publications in 

American Archaeology and Ethnology. 


United States National Museum — 





NOTE xi 

UW-PA University of Washington, Publications in 

Ver Verhancllungen der Berliner Gesellschaft 

fur Anthropologic, Ethnologic und Urge- 

ZB Zeitschrift fiir Ethnologic. 

Z vgl R Zeitschrift fiir vcrgleichende Rechtswisscn- 




Preface vii 


I. Introduction' 3 

II. Pioneers 10 

Meiners — Klemm — Waitz 

III. Biology, Prehistory, and Evolution ... 19 

IV. Adolf Bastian 30 

V. Comparative Law 39 

Bachofen — McLennan — Maine 

VI. Lewis H. Morgan 54 

VII. Edward B. Tylor 68 

VIII. Progress 86 

Prehistory, Technology, Field Work — Diffusion — Com- 
parative Economics — Art — Society and Religion — Primi- 
tive Mentality — Hahn — Ratzel 

IX. Franz Boas 128 

Field Work — Views on Mental Processes in Relation to 
Culture — Functionalism — Geography and History — Sum- 

X. Historical Schools : British Diffusionists . 156 

Historical Schools — Elliot Smith — Rivers 

XI. Historical Schools : German Diffusionists . 177 
XII. French Sociology 196 

Durkheim — Durkheim 's Followers — L6vy-Bruhl — Radcliffe- 

XIII. Functionalism : Pure and Tempered . . . 228 

Malinowski — Thurnwald 

XIV. Retrospect and Prospect 248 

Geography — Psychology — Meta-Ethnography — The Out- 

Index 293 





Ethnograpliy is the science which deals with the 
''cultures" of human groups. By culture we understand 
the sum total of what an indiyidual acquires from his_ 
society — those beliefs, customs, artistic norms, food- 
habits, and crafts which come to him not by his own jere^ 
ative activity but as a legacy from the past, conveyed 
by formal or informal education.^ 

The relation of ethnography to sister disciplines is 
thus clear. It_is that part of antliropology (in the Eng- 
lish sense of the \vord, the whole science of man) which 
is not primarily concerned with races as biological di-.. 
visions oi Homo sapiens and does not interest itself in 
the psychology of individuals except insofar as it reflects 

or influ ences society. On the other hand, prehistoryJls, 
sunp ly the ethnograpliy of extinct social groups. 

At time s ethnography shares its subject matter with.. 
literature , but its attitude is distinct. An exotic milieu, 
say, Tahiti, impresses itself on the sensibilities of a 
Pierre Loti, whose talent may convey similar thrills to 
the reader. An ethnographer does less and more. He 
renounces aesthetic impressions except as a by-product; 



he does not select his facts for literary effectiveness since 
his duty lies in depicting the whole of cultural reality. 
As a luitnralist cannot confine himself to beautiful but- 
tediies, so the ethnographer must ignore nothing that 
belongs to social tradition. He records a boys' game on 
stilts as faithfully as he does the cosmogonies of Tahitian 
priests : both are part of his theme, and children at play 
may reveal as much of basic cultural process as does the 
metaphysical speculation of their elders. 

The ethnographer also parts company with the an- 
tiquarian who collects odd customs with a philatelist's 
zeal about his stamps. From raw facts a scientist pro- 
ceeds to orderly arrangement and interpretation. How 
have cultures come to be as they are? Why do remote 
peoples share similar ideas and usages? Why does a 
certain group fail to make an adequate adjustment to 
climate? Why does another perpetuate a custom no longer 
appropriate? These are among his problems; and in 
proportion as they engage him, the descriptive ethnog- 
rapher turns theoretical ethnologist. 

But theory can proceed sanely only on a w ide f oun - 
dation of fact. That is why all branches of anthropology 
necessarily lagged behind until geographical discovery 
enlarged their scope. A map of the ancient Greek or Ro- 
man world at once shows why its makers were precluded 
from sound conceptions of man as a species : they lacked 
elementary knowledge of his varieties. Eratosthenes (ca. 
200 B.C.) knew nothing of Australia or Oceania or Amer- 
ica; his view of Asia did not extend beyond India, and 
in that quaint triangle of his which stands for Africa 
the sources of the Nile are placed at the southernmost 
extremity. Ignorant of major races, the Greeks naturally 
were in no position to subdivide mankind on a rational 
basis. The Orient was no better off. When in 126 b.c. 
General Chang Kieng returned to China from his western 
travels, he brought *Ho his astonished countrymen a 


glowing account of the new world which he had discov- 
ered, and which was nothing less than the Hellenic- 
Iranian civilization inaugurated in those regions by the 
successors of Alexander the Great" (B. Laufer). 

Adequate knowledge of the globe is amazingly re- 
cent. New Zealand was not so much as sighted before 
1642, and a fuller acquaintance with Oceania sets in only 
with Bougainville and Cook. Little more than half a 
century ago our best maps showed a blank space for the 
Belgian Congo. How could anyone survey humanity even 
superficially until at least its location and its range of 
variation were determined? The discovery of Australia, 
for example, disclosed a new race and many distinctive 
social practices. 

What is more, even when the facts are established, 
it takes time for sound concepts to mature. The philoso- 
pher Meiners, writing towards the end of the eighteenth 
century, had read the accounts of the great explorers of 
his day. As a result of his studies, however, he classifies 
man into two main stocks — the Mongolic and the Cau- 
casian or Tartaric, the latter being subdivided into (a) 
the Gothic or Celtic; and (b) the Slavic. Meiners derives 
the Australians from ''the lower caste or the oldest 
inhabitants of Hindustan," which at first blush seems 
an anticipation of modern classification. But he mars his 
scheme by crediting these primeval natives of India with 
a Mongolic origin. His Mongolic stock is thus ancestral 
to the Eskimo, Chinese, Japanese, Indo-Australians, 
Papuans, Australians, and some of the African Negroes. 
As for the higher castes of India, they are indeed Cau- 
casian, but "Slavic" — a category that includes also 
Armenians, Arabs, Persians, European Slavs, and many 

If it took time to correct grosser misconceptions as 

1 C. Meiners, Grundriss der Geschiohte der Menschheit, 17 sq., 30 sq., 
Lemgo, 1785. 


to physical affinities, the situation was inevitably worse 
for an understanding of culture, especially in its less tan- 
gible aspects. Travelers who admirably elucidated ex- 
ternals failed to go deeply into native beliefs and custom. 
Captain Cook was accompanied by such scientists as 
Banks and Forster, whose observations remain inestima- 
ble. But the time spent on his voyages permitted no 
thorough study of religion or family life. On such points 
missionaries, fur traders, and others whose calling en- 
forces long residence are often superior even to modern 
specialists. The religion of Brazilian aborigines emerges 
more clearly from the reports of early Portuguese, 
French, and German visitors than from the works of 
such reputable ethnographers as Karl von den Steinen 
and Fritz Krause.^ And the meager or confused reports 
of Farabee and Koch-Griinberg on South American mar- 
riage customs do not approach the accounts of Andre 
Thevet (1575), Gabriel Soares de Souza (1587), and 
Father F. S. Gilij (1781).' 

The simple truth is that professional training, while 
important, cannot perform miracles. It can make an in- 
vestigator note what an equally good unschooled observer 
would neglect. The most brilliant amateur cannot divine 
that at a particular stage of science apparently trivial 
details, like a basketry technique or the number of tent 
poles, may assume crucial importance. On the other hand, 
the elusive facts of social life and religious belief cannot 
be ascertained by the best specialist without long and 
arduous inquiry. It is when a talent for observation ac- 
companies both protracted residence and contact with 
professional ethnography that we obtain such superb 
results as mark the work of Snr. Nimuendaju.* 

2 A. M6traux, La religion des Tupinamia, Paris, 1928. 

3 Paul Kirchhoff, ' * Die Verwandtschaftsorganisation der Urwald- 
stanune Siidamerikas, " ZE, 63:55-193, 1931. 

•* Curt Nimuendaju Unkel, "Die Sagen von der Erschaffung und 
Vernichtung der Welt als Grundlagen der Eeligion der Apapocuva- 
Guarani," ZE 46:284-403, 1914. 


From the preceding remarks it is clear why the 
very material for ethnographic interpretation was hard 
to obtain, and slow in coming in. Yet, in principle, sound 
theory presupposes ample information on every phase 
of life from every society in the world. However, even 
this demand is insufficient, for cultures vary in time as 
well as in space. Here we meet a second reason for the 
tardiness of ethnology. Until recently its time perspective 
was even more defective than its spatial vision. What 
vistas hitherto undreamed of were opened with the deci- 
pherment of the Eosetta stone and of the Behistun in- 
scription! Yet Champollion presented his discovery to 
the French Academy no earlier than 1822, and Rawlinson 
published on Old Persian cuneiform in 1847. And these 
discoveries, startling as they were, carried us only a 
few millennia before Christ. The real revolution came 
with the recognition of Boucher de Perthes. When at the 
joint meeting of the Austrian and German Anthropologi- 
cal Societies at Innsbruck Rudolf Virchow explained the 
organization of the great European associations devoted 
to our science, it was to this event that he rightly assigned 
the first place."* 

Jacques Boucher de Crevecoeur de Perthes (1783- 
1868) was an amateur antiquarian who as early as 1836 
argued that man was contemporaneous with extinct mam- 
mals, a thesis he defended before the local Societe Imperi- 
ale d 'Simulation at Abbeville. At first proceeding a priori, 
he soon (1838) submitted stone hatchets in proof of 
human craftsmanship in the Pleistocene period. His was 
the common fate of prophets : ' ' Practical men disdained 
to look ; they were afraid ; they were afraid of becoming 
accomplices in what they called a heresy, almost a mysti- 
fication: they did not suspect my good faith, but they 
doubted my common sense." ° 

^Kudolf Virchow, " Eroff nungsrede, " AGW-M, 24:70-77, 1894. 
^ Boucher de Perthes, De I 'homme antediluvienne et de ses oeuvres, 11, 
Paris, 1860. 


A treatise De V Industrie primitive (184G) made no 
impression on the learned until Dr. Rigollot (1854), a 
former antagonist, examined the sites from which the 
tools had been secured and announced his conversion. 
Still the guild of savants remained unconvinced. There 
were those who pronounced the strata of the hatchets as 
hardly older than the advent of the Romans. Some thought 
the tools had sunk to Pleistocene depths by their own 
weight. Others doubted the human origin of the flints, 
assigning their shape either to volcanic or to glacial 

At last, in 1858, the British paleontologist Hugh 
Falconer examined Boucher de Perthes' collection and 
expressed himself satisfied by the evidence. His com- 
patriots, Joseph Prestwich, John Evans, and Charles 
Lyell, followed suit ; and at the Aberdeen meeting of the 
British Association for the Advancement of Science in 
1859, Lyell announced his complete acceptance of the 
new views. The results harmonized \\ researches on 
British soil reported at the preceding meeting and led 
to the reinvestigation of stone finds previously registered 
but neglected. 

The recognition of Boucher de Perthes' thesis 
marked a new era because it implies that culture dates 
back to the Pleistocene : the flints were not only made by 
man, they were obviously more than random freaks and 
worked in conformity with a social tradition. What is 
more, Boucher de Perthes was, in modern parlance, 
something of a functionalist (see Chap. XIII). That is to 
say, he understood that the artifacts discovered could 
not be isolated products of some technological instinct, 
but fitted into a larger context. He maintained stoutly 
and convincingly that the ancient stone knappers must 
have had not merely knives and hatchets, but a language, 
an art, social customs — in short, an equivalent, however 


rude, of a complete culture/ In other words, culture dated 
back not a few paltry millennia, but to a period consid- 
erably prior to the geologically Recent era. The problem 
thus extended from the social existence of all living peo- 
ples to that of their cultures throughout all the thousands 
of years intervening between the Pleistocene and the 
present. The very question could not have been raised 
until about seventy-five years ago. Now it suddenly 
dawned upon students of civilization that what they 
had hitherto known was merely the final scene of a 
lengthy drama, that they had been trying to * ' reconstruct 
a book from its last chapter. ' ' ® 

^ Op. cit., 52-59. 

* Ch. Letourneau, La condition de la femme dans les diverses races et 
civilisations, 3, Paris, 1903. 



A point of departure is always arbitrary. We shall 
start with Meiners and Klemm because both had a tol- 
erably clear conception of the central core of ethnog- 
raphy. Yet each recognized predecessors with aims akin 
to his own. Among those cited by Meiners are Iselin, 
Falconer, Goguet, and Montesquieu; while Klemm con- 
cedes priority to Voltaire: *' Voltaire was the first to 
push aside dynasties, series of kings, and battles, and 
to seek what was essential, Culture, as it manifests it- 
self in customs, faith, and governmental forms." 

M E I N" E E s 

In Meiners' Grundriss (1785) ^ this concept is adum- 
brated, but not yet distinctly conceived. As already 
pointed out, this author treated man's bodily diversity 
along with his social characteristics, but — what was far 
worse — he arbitrarily selected for discussion certain 
phases of social tradition to the exclusion of others. 
Avowedly for mere convenience' sake, he eliminated re- 
ligion and the later stages of scientific development and 

1 See page 5 for precise reference. 



still less defensibly disregarded all ''but a few of the 
most remarkable customs — for to describe and explain 
all the practices of all peoples would be an enterprise 
equally foolish and thankless: foolish, because one can 
never include the infinite number of observances ; thank- 
less, because most of them can be no more satisfactorily 
explained than the formation and derivation of most 
words and because even the most probable explanations 
yield little for a knowledge of human nature." Meiners 
is thus very far from postulating a science that shall on 
principle deal with the whole social heritage of human 

Nevertheless, Meiners did sense the need of a new 
branch of learning to be set over against political history, 
a science to be dubbed ''the history of humanity." Nor 
can he be accused of vagueness as to its contents. He 
lists as topics "food and strong beverages, dwellings, 
dress, and adornment of all nations" . . . ; "the . . . 
opinions of wild and barbarous peoples about the most 
important phenomena and effects of Nature and finally 
the history of the beginnings of the most necessary 
sciences. ..." Elsewhere he adds "remarkable prac- 
tices, the education of children, treatment of women, 
forms of government and laws, customs, notions of wealth 
and decorum, of honor and shame." This is, indeed, a 
formidable roster; and when Meiners pleads for a view 
of man as he has been at all times and all places, he 
voices the aims of modern anthropology. "What we miss 
is a clear statement of what unites all these several dis- 
parate aspects of human life. 

Klb m m 

Clearer in his formulation and a more potent in- 
fluence on research was Gustav Klemm,^ to whose "in- 

2 Gustav Klemm, Allgemeine Cultur-GescMchte der Menschheit. 1. 
Die Evnleitung und die TJrzustdnde der Menschheit enthaltend (Leipzig, 


valuable collectiou of facts" E. B. Tylor paid his respects 
in his first general treatise (18G5). 

Gustav Klemm (1802-1867), a native of Chemnitz 
in Saxony, was from childhood a passionate collector of 
specimens and an eager student of museums, without, 
however, slighting religion and social structure. An om- 
nivorous reader, he offered what the accessible sources 
supplied, a veritable treasure-trove of facts, often sig- 
nificant, sometimes at least entertaining, from all known 
regions and periods. Even today one could not readily 
find a fuller compendium on the cuisine of all ages. He 
ransacked recondite chronicles for details of human 
interest, sometimes suggesting Berth old Laufer's writ- 
ings and perhaps still more Alfred Franklin's books on 
medieval France. Sometimes, to be sure, Klemm turns 
a mere gleaner of trivialities, as when his discussion of 
swords merges in a complete list of famous Toledo 
smiths of the sixteenth century. 

But Klemm did more than offer masses of raw 
material; we must credit him with anticipating Tylor 's 
classical definition of culture, which is virtually our own. 
Klemm makes it comprise ''customs, information, and 
skills, domestic and public life in peace and war, religion, 
science, and art." "It is manifest in the branch of a 
tree if deliberately shaped; in the rubbing of sticks to 
make fire; the cremation of a deceased father's corpse; 
the decorative painting of one's body; the transmission 
of past experience to the new generation." A tiny twist 
to the last phrase would have turned the trick of an ade- 
quate definition.^ 

The notion of progressive development was familiar 
to Klemm, even though his main works preceded Dar- 
win's Origin of Species. The idea itself is, of course, both 

1843; here quoted aa C-G. Allgemeine Culttt/rwissensohaft (quoted as Cw.), 
in two parts: WerTczeuge und Waff en, Leipzig, 1854; and Das Feuer, die 
Nahrung, Getr'dnke, NarTcoiika, Leipzig, 1855. 

3G. Klemm, Cw, 217, 1854; C-G, 1:21; Cw, 37, 1855. 


ancient and primitive. It occurred among the G-reeks 
and has been recorded from the lips of reflective sages 
among illiterate aborigines. The forms it took sometimes 
closely resembled the results of sober research, but we 
emphatically agree with Virchow that they must not be 
confounded with them. When a Chinese compilation of 
52 A.D. presents the sequence of a Stone, Bronze, and 
Iron Age, this is not a case of genius forestalling science 
by two thousand years; an alert intelligence is simply 
juggling possibilities without any basis of facts or any 
attempt to test them.* The Chinese Stone Age became a 
scientific problem when other civilizations turned out to 
have had premetallic periods, and a reality when, two 
decades ago, Andersson excavated sites of stone work- 

Klemm did not have to go far back for an evolu- 
tionary conception of man. Whether he was influenced 
by it or not, there was Condorcet's Esquisse d'un tableau 
historique des pr ogres de l' esprit humain (1795), out- 
lining how primitiveness rose through stages of animal 
husbandry and agriculture to alphabetic writing, and ul- 
timately enlightenment.^ Klemm recognized three stages 
— savagery (Wildheit) ; tameness (Zahmheit) ; and free- 
dom. He is clearest in defining the first of these: The 
savage roamed about, owning neither herds nor land 
and recognizing no paramount authority. On the second 
plane, families are consolidated into tribes with rulers 
by divine sanction. Here develop writing, pastoral life 
and farming, but with the limitations imposed by priestly 
domination. Freedom comes only when nations shake off 
this yoke, thereby gaining a chance to develop their men- 
tality in all directions. The Persians, Arabs, Greeks, Eo- 

4B. Laufer, "Jade," FMNH-PAS 10:71, Chicago, 1912. 
"J. Salwyn Schapiro, Condorcet and the Bise of Liberalism, 234-270 
New York, 1934. 


iiiaus, and, above all, the Germanic stock illustrate this 
highest step." 

However, it would be an anachronism to impute to 
Ivlemm either an anticlerical or a nationalistic attitude 
in the present sense of these terms. Christianity is for 
him the mainspring of progress toward freedom pre- 
cisely because it dissolves the national hierarchies. On 
the other hand, his views on race, though curious, bear 
no relation to those of Gunther or Hitler. His system 
does, indeed, involve a division of humanity into active 
and passive races, the latter being mere copyists who 
transmit what has been handed down from the past or 
imposed upon them by conquerors. However, the passive 
group comprises not only Mongoloids and Negroids, but 
even Egyptians, Finns, Hindus, and the lower strata of 
European society. Secondly, KHemm likens his two races 
to man and woman, respectively: as the sexes are mu- 
tually complementary, so the active race is incomplete 
without the passive, and vice versa. 

On this subject our author is somewhat obscure. 
He visualizes the earliest human beings as spreading 
over the globe, first as hunters and fishermen, later as 
herders, ''until increasing population necessitated agri- 
culture." At this point one might reasonably object that 
mere imitators do not inaugurate momentous economic 
changes. As for the active race, it somehow arose in- 
dependently near the Himalayas, traversing its own pe- 
riods of savagery and tameness, of stock raising and 
farming. Active peoples naturally subjected the passive 
ones, whence that fusion which Klemm considers es- 
sential for the ultimate ends of Nature. 

The foregoing principles — this dichotomy of our 
species and the three evolutionary stages — underlie the 
organization of the Cultur-Geschichte, which begins with 
"passive humanity" and passes on to the active race; 

8 Klemm, C-G, 21-23. 


while in each of these two sections the author proceeds 
from the lowest to the most advanced members according 
to the categories of savagery, tameness, and freedom. 
Klemm is prone to accept uncritically the psychological 
judgments of travelers and is equally naive in connecting 
mentality with geography. The tropical South American 
forest Indians lack ^'the finer sentiments of friendship, 
love, and modesty"; and though exceptional cases sug- 
gest that the Creator designed even them to progress, 
as a rule they are infinitely indolent and lethargic. As 
denizens of the woods they grow up with a limited hori- 
zon, while coast dwellers reflect the constant changes due 
to the sea, changes which foster their powers of con- 
centration. That is why the fishing tribes of Australia 
excel the South American forest Indian in alertness, 
reflectiveness, and intellectual independence. Fuegians 
are rated as peers of the Australians, thus assuming 
higher rank than the ''far cruder" forest tribes. In fair- 
ness to Klemm we must add that he is thinking rather of 
the hunting Botocudo than of the horticultural Arawak, 
so that the comparison, however unconvincing, is not 
quite so grotesque as it appears at first. He does refer 
to the use of bitter manioc, but without realizing the 
ingenuity required for eliminating its virulent poison.^ 
Klemm 's intellectual limitations appear when we 
compare his treatment of certain topics with Tylor's. 
He knew as well as the British ethnologist that the Kam- 
chadal cooked meat in wooden troughs filled with water 
into which they threw heated rocks. But it remained for 
Tylor to bring the custom into line with North American 
and Polynesian usages, to conceptualize the operation 
as ''stone-boiling," and to assign it a definite place in 
the history of cookery. Again, Klemm has excellent re- 
marks about fire — its universality, its being an exclu- 
sively human possession, the impossibility of investigat- 

TQ. Klemm, C-G., 1:196-200, 234, 280, 287 f., 327-332, Cw, 241, 1885. 


iiig its origin. Pie recognizes friction as an older technique 
than percussion and from a passage in Pliny infers its 
presence in pre-Christian Europe. But his typology 
marks no advance over Pliny's: the rubbing together of 
sticks, strike-a-lights, and the burning lens exhaust his 
inventory.'' Once more it was Tylor who defined the 
fire-plough as a distinct implement and determined its 
distribution ; Tylor who indicated drills managed with a 
thong, bow, or flywheel as diverse and superior forms of 

While his was hardly a great intellect, Klemm never- 
theless remains a noteworthy figure. His comprehensive 
and clear conception of what culture is, as well as the 
wealth and variety of his knowledge, assures him an hon- 
orable position among our pioneers. He also spread tech- 
nological information by the excellent and ample 
illustrations of his books at a time when museums were 
rare, inadequately equipped, and less accessible than 
they are today. 


Theodor Waitz (1821-1864) represents a wholly dif- 
ferent approach. As professor of philosophy at Mar- 
burg he was interested above all in psychological 
questions, and his AntJiropologie der Naturvolker (Leip- 
zig, 1858-1871), a six-volume work, in part published 
posthumously, is largely a treatise on primitive mentality. 
He explicitly refrained from technological detail on the 
plea that Klemm had amply dealt with that phase of the 
subject. On the other hand, he is strongest where his 
predecessor is weak — in the critical analysis of sources 
and the depth of his psychological insight. His position 
is defined in the first volume, which bears the sub-title 
Uber die Einheit des Menschengeschlechtes und den 
Naturzustand des Menschen. It was this part that was 

8G. Klemm, C-G 1:178 f.; Civ 66-70, 260. 


reissued in 1876 and in 1863 had been singled out for 
translation by the London Ethnological Society as the 
most representative continental treatise on man. The 
geographer Georg Gerland, who prepared the two final 
portions of the whole work for the press after the au- 
thor's death, rightly praises the *' strict, cautious, sober 
method" of this first volume. Considering its date (1858), 
we must pardon occasional errors and recognize it as a 
worthy forerunner of Boas' The Mind of Primitive Man 
(page 131), which closely parallels its argument. 

Waitz deprecates rash verdicts on racial disabilities, 
but his calm intellect remains undimmed by propagandist 
fervor. He is equally free from exaggerations of the geo- 
graphical order, explicitly refuting popular extrava- 
gances. Not all mountaineers love liberty, nor does grand 
scenery suffice to evoke aesthetic thrills. Environment 
does not automatically stimulate a specific adaptation, 
which is itself a function of a people's culture: the sea 
offers opportunities only to mariners, not to people who 
lack boats. We must thus distinguish between effective 
causes and mere occasions or minor factors of progress. 
Geography may inhibit but does not necessarily create. 
It is overshadowed by historical and social determinants 
— by migrations and the consequent diffusion of traits. 
A dense population is both a result and a cause of ad- 
vancement. Native capacity, Waitz undogmaticaily con- 
cludes, is at least roughly the same in all races. If some 
have shown a greater trend toward civilization, it is 
because of favorable circumstances. The degree of culture 
is thus far less an index of innate endowment than of 
the vicissitudes of history. Waitz is especially cogent in 
comparing civilized and uncivilized conditions. He 
stresses — though with characteristic moderation — the 
role of genius and insists that important discoveries were 
made on ruder as well as higher levels, but that in both 
cases they probably had to be repeated before people at 


large were able to profit from them. Essentially, Waltz's 
insight into the determinants of progress is as clear as 
that of today.® 

Waitz appears to advantage in the second volume 
(1860), which applies his principles to Negroes — Die 
Negervolker und ihre V erwandten. Its compact resume 
of accessible data, with its resolute exclusion of tourists' 
fancies, is surpassed only by the writer's incorruptible 
judgment on what was then a hotly disputed problem. 
He does not regard Negroes as merely excellent imitators 
who are thereby nearer to the ape; but neither does he 
accept at its face value the exaggerated estimate of w^ell- 
meaning humanitarians. Throughout there is discriminat- 
ing appraisal. Individual Negro geniuses are cited, such 
as the inventor of the Vei alphabet and the rulers of 
great Negro states. A race representing a lower species 
could not produce any individuals of such exceptional 
ability. On the other hand, among us, too, it is the out- 
standing genius who creates group progress. In answer 
to the charge of cruelty as an innate trait of Negroes, 
he adduces the mild treatment of African slaves. And 
considering how inadequate his sources were on such 
topics, he forms remarkably sound conclusions about 
Negro religion. Rejecting then current notions, he rec- 
ognizes the affinity of the African ''fetich" with the 
American "medicine"; and sees the natives hovering on 
the brink of monotheism.^" 

In 1860 a student who had mastered the writings of 
Klemm and Waitz would thus be in a somewhat better 
position than might at first seem conceivable. With a 
tolerable survey of the material equipment of human 
groups he might unite a fair perspective of the forces 
controlling progress. 

9Th. Waitz, op. cit., 1:408-424, 428 f., 447, 473 f., 482. 
lojfctd., 2:167, 175, 216, 222, 228-232. 



As we have seen, the idea of progressive develop- 
ment from savagery to civilization was much older than_ 
Darwin or even Lamarck. However, when evolution be- 
came not merely an approved biological principle but a 
magical catchword for the solution of all problems, it 
naturally assimilated the earlier speculations about cul- 
tural change as obviously congruous with its own phi- 
losophy. Similarly, the discoveries of prehistory neatly 
fitted into the evolutionary picture. Both biological theory 
and archaeological research powerfully stimulated the 
study of culture, but not without creating grave misun- 

Evolutionary doctrines implied that complex organ- 
isms had slowly developed from extremely simple forms. 
Pitt-Rivers,^ transferring the notion to the sphere of 
human arts, postulated Spencerian changes ''from the 
simple to the complex, and from the homogeneous to the 
heterogeneous." Moreover, not only evolutionary but 
Darwinian, he assumed continuous modification by minute 

1 A. Lane-Fox Pitt-Rivers, The Evolution of Culture and other Essays, 
Oxford, 1916. 



steps. Stonework, for instance, he traced from the crudest 
flint implements to the polished celts of modern savages 
by "numerous intermediate gradations." And in deriv- 
ing diverse Australian contrivances, such as the mush- 
room club, the shield, and the boomerang, from a single 
prototype — a simple cylindrical stick — he obviously pat- 
terns his procedure on the phylogenetic hypotheses of 
contemporary zoologists. 

One problem that seems to loom disproportionately 
large in the pages of Victorian anthropologists is that of 
degeneration. But if Lubbock, Tylor, and Pitt-Rivers de- 
voted page after page to that subject, there was an excel- 
lent reason. Influential writers, theological and otherwise, 
were contending that primitive peoples had retrogressed 
from a higher state. In rebuttal the evolutionists reiter- 
ated with unremitting emphasis that the dominant note 
in the history of the species was an upward movement, 
decline being exceptional.^ Their general point of view 
is concisely stated by two of its enthusiastic champions. 
Says Letourneau: "All the civilizations past or present 
had their barbarous or savage infancy, out of which they 
have slowly and painfully evolved . . . ; the rude con- 
temporary races, the lowest of which border on animality, 
picture for us, in general fashion, the slowly progressive 
phases which were traversed by the ancestors of civ- 
ilized peoples. ' ' ^ Similarly, Pitt-Rivers declares that 
"the existing races, in their respective stages of progres- 
sion, may be taken as the bona fide representatives of 
the races of antiquity. . . . They thus afford us living 
illustrations of the social customs, the forms of govern- 
ment, laws, and warlike practices, which belong to the 

2 See e.g. E. B. Tylor, Researches into the Early History of Mankind 
and the Development of Civilization, 150-190, London, 1865; Lord Aveburj 
(= Sir John Lubbock), The Origin of Civilization and the Primitive Con- 
dition of Man, 6th edition reissued with a new preface, 524-552, London, 

8 Ch. Letourneau, op. cit., 3. 


ancient races from which they remotely sprang, whose 
implements, resembling, with but little difference, their 
own, are now found low down in the soil. . . ." * 

Biologists had in part direct proof for their evolu- 
tionary schemes, paleontological findings ; the anthropolo-__ 
gist's counterpart was prehist ory. L et us, then, examine 
wherein its value lay. 

Boucher de Perthes had shown that stone work, 
hence culture, dates back to the Pleistocene, but he was 
not the first to propose a Stone Age on empirical grounds. 
That honor belongs to a Danish historian : in his Apergu 
sur les periodes les plus anciennes et les plus remarqua- 
bles de I'histoire nationale (1813) Vedel-Simonsen had 
already argued for three periods of Scandinavian an- 
tiquity — a Stone, a Copper or Bronze, and an Iron Age. 
Of his immediate successors. Christian Jurgensen Thom- 
sen accepted the idea, and Worsaae (1821-1885) extended 
it to other European countries.'^ This generalization was 
not more readily accepted than that of Boucher de 
Perthes. As late as 1875 German scholars categorically 
denied a Bronze Age, and bitter controversies were waged 
over this point, the Scandinavians — Worsaae and Sophus 
Miiller — opposing their German colleagues.* 

The Stone Age, however, had been definitely estab- 
lished by that time, and its local developments were be- 
ing brilliantly illuminated by new excavations and 
conceptualizations. The Danish zoologist Japetus Steen- 
strup demonstrated a premetallic age of hunters and 
fishermen living under climatic and arboreal conditions 
prior to those of written records. Others divided the 
Stone Age into two main epochs, that of chipped stone 
tools (paleolithic) and that of ground implements (neo- 
lithic). The former was again susceptible of division 

* Pitt-Eivera, op. oit., 53. 

^ IngAvald Undsftt, "Le prehistorique scandinave, ses origines et son 
dgveloppement, " Eev. 3" s6rie, 2:313-332, 1887. 
6L. Lindenschmit, Ar-A, 9:152, 1876. 


according to the artifacts produced, viz., into the Acheu- 
lean, Mousterian, Solutrean, and Magdalenian/ We must 
not omit the Swiss lake dwellings, which disclosed a 
complete neolithic culture with relics of farming, animal 
husbandry, earthenware, weaving, and ground stone 
axes, for the inability of scholars to link these finds with 
the products of any historically known peoples deeply 
stirred the minds of European scholars.* 

Undoubtedly, then, prehistory proved evolution by 
the rigorous technique of geological stratigraphy at a 
time when ethnographers were still groping for proper 
methods of investigating living aborigines. No wonder 
that ethnographers leaned heavily on the staff of archae- 
ology. People who raised cereals and ground stones were 
equated with the neolithic Swiss; a knife-chipping Aus- 
tralian horde was set down as Mousterian. 

What the evolutionists as a class failed to see was 
the limited range of cultural facts for which progress 
could be directly demonstrated. Prehistory reveals only 
material phenomena, and only part of them. Only under 
exceptionally favorable circumstances of preservation 
can it teach anything about even such tangible but per- 
ishable objects as bows or basketry. In short, it deter- 
mines accurately certain phases of technology and nothing 
else. A picture derived solely from archaeological sites 
is often grotesquely distorted: the most skillful wood- 
carvers and bark-cloth beaters, leaving little evidence of 
their craftsmanship, must suffer in comparison with 
potters and metallurgists. 

A mechanical transfer of the prehistoric categories 
in vogue fifty years ago is thus fatal even for sound ap- 
praisal of technical progress. It is true enough that all 
metal-working groups have passed through a premetallic 

^Gabriel de Mortillet, "Classification des diverses periodes de I'age de 
la pierre," Rev. 1:432-442, 1872. 
8 Cf . E. Virchow, op. oit. 


stage, but it does not follow that all peoples without 
metals make stone tools. As Von den Steinen insisted, 
many South American natives find this impossible for 
lack of stone.^ In order, then, to apply the concept * ' Stone 
Age ' ' here, we must divest it of its original meaning and 
make it include bone, shell, wood, and the like. Again, the 
separation into a Paleolithic and a Neolithic period can 
be kept useful only if we completely alter the primary 
sense of these terms. Some Australians merely chip stone, 
their neighbors not otherwise a whit more advanced grind 
axes because they have access to diorite; the difference 
implies only a difference in material resources, hence 
it cannot serve as a major line of demarcation. In order 
to infuse significance into the term, investigators have 
perforce redefined ''Neolithic" to indicate primarily the 
status of potters and farmers. Thus, prehistory, instead 
of being an infallible guide, required correction and con- 
ceptual purging by ethnographic treatment. 

But if prehistory left gaps and sometimes even mis- 
led scholars on points of technology, it had nothing what- 
soever to offer on the growth of supernaturalism and 
social organization. That was intolerable for the evolu- 
tionist mentality, which demanded the sequence of events 
for every phase of human activity. The biologist, simi- 
larly handicapped by the defectiveness of the geological 
record, had recourse to embryology. Anthropologists eked 
out results of excavation by falling back on a correspond- 
ing law of growth. As Homo sapiens was zoologically at 
the peak of the animal kingdom, so Western Europe in 
1870 marked the goal of civilization. As the single cell 
was the hypothetical starting point for evolution, so a 
savage hovering on the border of bestiality must serve 
as the point of origin for culture. Since, however, that 
primeval man could no longer be observed, modern sav- 

8 Karl von den Steinen, TJnter den Naturvolkern Zentral-Brasiliens, 
2nd ed., 196, Berlin, 1897. 


ages were lightly substituted insofar as they differed 
from Victorian Europe. On the other hand, usages of 
modern Europeans not in keeping with their advanced 
status were like those rudimentary organs of animals 
which Darwin had compared to the letters of a word 
that are no longer pronounced. 

A fatal fallacy of all this reasoning lay in its na'ive 
equation of modern primitive groups with the primeval 
savage, as in the sentences quoted from Letourneau. This 
led serious writers into absurd underestimation of re- 
cent tribes and uncritical acceptance of tourists' tales. 
Sir John Lubbock was one of the most versatile minds of 
Ms age, an eminent prehistorian, a writer who ranged 
over the whole field of anthropology and thought inde- 
pendently on all of its phases. Yet he is no better than 
Klemm at assessing poor evidence.^" The Andamanese 
have **no sense of shame"; "many of their habits are 
like those of beasts. ' ' The Greenlanders have no religion, 
no worship, no ceremonies. The Iroquois have no religion, 
no word for God, Fuegians not the least spark of re- 
ligion. * * . . . there can be no doubt that, as an almost uni- 
versal rule, savages are cruel." 

This last sentence illustrates another major error — 
the complete abandonment of objective criteria. What is 
cruelty? Is the cannibalism of Oceania worse than the 
wholesale massacres of modern warfare? Sir John's 
writings teem with subjective judgments, naively passed 
on the basis of resemblance to or deviation from Euro- 
pean standards. The Hottentots are ''disgusting," the 
Australians "miserable" savages. Occasionally he ex- 
hibits insight, as when he corrects Prescott for ascribing 
human sacrifice to "fiendish passions. "^^ But, generally 

10 John Lubbock, Prehistoric Times, 430, 437, 448, 509, 511, 536, 570, 
London, 1872. 

11 The Origin of Civilisation, 384. 


speaking, he is himself constantly mortified, shocked, 
horrified, by the savage scene. 

The modern scientific procedure is to refrain from 
all subjective pronouncements; to recognize that while 
material objects and rationalistic schemes are '* higher" 
or ''lower" — better or less suited to their purpose — this 
does not hold for art, religion, and morals, where no 
universally recognized standards exist. The anthropolo- 
gist as an individual cannot but respond to alien mani- 
festations in accordance with his national and individual 
norms; as a scientist, however, he merely registers can- 
nibalism or infanticide, understands, and if possible ex- 
plains such customs. 

Lubbock's egocentrism appears most oddly in a sub- 
ject that certainly does not bear directly on ethical at- 
titudes — the designation of relatives. He appreciatively 
taps the Eskimo on the shoulder for ** correctly" recog- 
nizing uncles and aunts, that is using true equivalents 
of the English terms, while the Hawaiians, who have no 
special word for these relatives, are credited with the 
most savage nomenclature known." 

The resemblance of modern savages to a primeval 
ape-man is so important a tenet that we must explicitly 
expose the error. It lies in failing to understand that 
even the simplest recent group has a prolonged past, 
during which it has progressed very far indeed from that 
hypothetical stage. To look for any living people with- 
out religion, for instance, is like trying to observe life 
evolving out of inorganic matter. As we shall see, the 
greatest of the evolutionists avoided this pitfall. 

Finally, as to "survivals," the rudimentary organs 
of social groups. Unquestionably civilization in all its 
stages teems with illustrations. Instances are offered by 
the most determined opponents of the evolutionist sys- 
tem. Thus, among some of the Eskimo Boas notes the 

^Ibid., 183, 202. Also Ib JEAI, 1:11, 1872. 


curious womon's stockings, which bulge out enormously 
just below the knee. Whence, he asks, such an odd fash- 
ion? An old source testifies that in 1750 the women wore 
huge boots kept open by whalebone hoops, children be- 
ing put inside these pouches. Thence comes Boas' con- 
clusion that "the wide ankle-pouch of the long stocking 
of the west coast of Hudson Bay may be a survival of 
this wide boot."" The question is not, then, one of ac- 
cepting or rejecting survivals, but whether an alleged 
survival is genuine or spurious. For in culture as well 
as in biology, there may be alternative explanations for 
a ''useless organ." More particularly, the utility may 
be merely masked, the feature fully functioning in some 
unexpected way. For example, a favorite argument of 
the period was to point to the avunculate, i.e., the special 
powers enjoyed by a maternal uncle, in societies that 
reckoned descent through the father. How, it was asked, 
could such a disharmonic trait be interpreted? Surely 
in only one way: as a legacy from an earlier period in 
which the tribe had traced descent through the mother, 
hence had emphasized the role of maternal kinsmen. 
This argument, however, no longer convinces because 
the maternal uncle may have become important in other 
ways. If, for instance, a society with or without patri- 
lineal clans made a husband do service for his bride, he 
would naturally settle in her home or village, and his 
offspring would automatically fall under the sway of 
her kin. There is thus no need at all to infer matrilineal 
descent from avuncular authority. In fact, the alterna- 
tive cited is only one of several plausible explanations. 
The survival argument would be uniformly trust- 
worthy only if there were a fixed law of sequence. Such 
laws were indeed repeatedly affirmed but hardly ever 
demonstrated outside of technology. The evolutionists, 

13 F. Boas, * ' The Eskimo of Baffin Land and Hudson Bay, ' ' AMNH-B, 
15:105, 356, pi. Ill, 1906. 


assuming that they knew the course of development as 
predestined, considered themselves able to interpolate 
missing links in the chain of events. In this they were 
far too optimistic, and their interpretation of survivals 
was correspondingly faulty. 

So far the discussion of the doctrine before us 
sounds like an arraigmnent. But it would be gross in- 
justice to underestimate its services. To the insight of 
our hypothetical student of Klemm and Waitz is added 
for the first time a synthesis of all cultural data, com- 
bining from a single point of view the results of ethnog- 
raphy and prehistory. If the lesser apostles of the 
theory rashly distorted the facts to eke out lacking evi- 
dence, saner spirits avoided such gross blunders. In- 
numerable new problems were broached, some of them 
were even solved under the impetus due to the luminous 
concept of progressive development. For, notwithstand- 
ing the qualifications cited, evolution is a positive fact in 
material culture and freely conceded by the most de- 
termined critics of its Victorian champions." To admit 
this, together with the possibility that material condi- 
tions may affect other phases of life, is to open the way 
for a fixed sequence of social and religious phenomena. 
Actually, these writers themselves postulate stages of 
development {Stufen der Entwicklung), in other words, 
evolution.^^ From entirely distinct starting points such 
contemporary anthropologists as Thurnwald and Rad- 
cliffe-Brown are also rehabilitating the concept. It is 
thus very far from dead, and our duty is merely to de- 
fine it with greater precision. 

Finally, another word of caution seems indicated. 
It has become customary to oppose cultural evolutionism 
to the principle of diffusion. This is by no means a fair 

^* W. Schmidt and W. Koppers, Volker und Kulturen, 45 f ., Eegensburg, 

^^Ibid., 264 sq. 


view of the matter. To be sure, given a fixed law of de- 
velopment, the same beginnings might lead to an inde- 
finite repetition of the identical stages. Crude stone- 
fracturing might thus be followed everywhere by the 
same series of techniques first recorded in France, so 
that Southern Spain and Africa and China would all 
exhibit a sequence of Chellean, Acheulean, Mousterian, 
Solutrean, and so on; and correspondingly with traits 
of other categories. Actually, the early evolutionists dif- 
fered individually in their attitude towards this problem. 
Some were inclined to explain resemblances by inde- 
pendent multiplication due to an identical law of gro\vth. 
But this was probably not wholly true of any writer, 
and demonstrably held for neither Tylor nor Morgan, 
the most influential thinkers of all. So orthodox an evolu- 
tionist as Pitt-Rivers was emphatically not a parallelist. 
He did not, to be sure, categorically deny that mankind 
might have ''independently designed the same forms of 
tools in various parts of the world." But he explicitly 
derives the boomerangs of Australia, the Deccan, and 
Egypt from a single center ; he holds that in the develop- 
ment of the bronze celt "each new improvement was 
communicated from tribe to tribe and from nation to 
nation" ; and his final words (in 1864) are that "by means 
of intercommunication, no less than by spontaneous de- 
velopment, have been formed those numerous combina- 
tions which so greatly puzzle the student of culture at 
the present time."^® A generation later, Pitt-Rivers' 
commentator is not less emphatic in vindicating the im- 
portance of dissemination: "Cases of independent in- 
vention of similar forms should be considered to have 
established their claim to be regarded as such only after 
exhaustive inquiry has been made into the possibilities 
of the resemblances being due to actual relationships." ^' 

16 Pitt-Rivers, op. cit., 145, 153, 183, 228. 

17 Henry Balfour, Introduction to Pitt-Rivers, op. cit., xii, xix. 


This is very nearly the principle of modern diffusionism 
(page 158). 

As a matter of fact, biological evolution would not 
suggest parallelism. Zoologists did not assert that man 
or the horse had again and again risen from humbler 
prototypes, but they believed in unique happenings cul- 
minating in so many end results. Parallelism was possi- 
ble only on the principle that the psychic unity of man- 
kind constantly impelled societies to duplicate one 
another's ideas. That, however, was not a corollary from 
evolution, but the doctrine of a staunch critic of Darwin- 
ism, Adolf Bastian. 



Adolf Bastian was born in Bremen in 1826. He 
studied law, natural science, and medicine in no less than 
five German universities, spending his last semester in 
Wiirzburg under the great pathologist Rudolf Virchow. 
Having graduated as a doctor of medicine in 1850, he 
promptly secured a position as ship's surgeon and spent 
the ensuing eight years voyaging to Australia, Peru, 
Mexico, California, as well as to various Asiatic and 
African countries. In 1859 he published Ein Besuch in 
San Salvador, and the following year Der Mensch in 
der Geschichte. In 1861 he was off again to Farther India 
and Eastern Asia, and henceforth his life was punctuated 
by lengthy travels to remote corners of the globe, with 
intermediate sojourns in Berlin, his chosen headquarters. 
There he became Curator of Ethnography (1868), sub- 
sequently founding what remained for decades the 
largest emporium of ethnographica in the world, the 
Konigliches Museum fiir Volkerkunde (1886). In 1869 
he helped Virchow organize the Berlin Society for An- 
thropology, Ethnology, and Prehistory, became co- 
editor of its journal, the Zeitschrift fiir Ethnologie, and 



took an active part in geographical activities. Bastian, 
paradoxically, was a successful promoter, a ''fisher of 
men," an astute buyer whenever a chance offered to 
enrich his beloved Berlin collections, and simultaneously 
a shy bookworm, if not a recluse. In 1903 von den Steinen 
asked him when he had last visited a theatre ; after some 
reflection came the answer, "In 1859." His meteoric ap- 
pearances and disappearances became proverbial in Ber- 
lin ; and he died in Trinidad, off the coast of Venezuela, 
in 1905, at the ripe age of seventy-nine. Karl von den 
Steinen epitomized his career in one sentence: "No 
German scholar has traveled more, none read more, none 
written more. ' ' ^ 

Two factors invested Bastian with a comic halo in 
the judgment of irreverent posterity — his determined 
opposition to Darwinism and his style, a combination 
that evoked Haeckel's vituperative title of "Geheimer 
Oberkonf usionsrat. ' ' Both manifestations, however, have 
been misunderstood. 

Bastian 's attitude towards biological problems re- 
flected not theological prejudice but a puritanical em- 
piricism. Let us not forget his association with Virchow. 
That great pathologist embodied, above all, the reaction 
of triumphant natural science against the speculations 
of the German Naturphilosophen. What could not be 
determined by direct observation or experiment savored 
of metaphysics. Like Virchow, Bastian regarded trans- 
f ormism as untenable so long as no one had ever seen one 
species changing into another. He spoiled his case when, 
decrying the homologies of the evolutionists as "scien- 
tifically undefinable similarities, ' ' he pretended to see no 
difference between them and the analogy between a tulip 
stalk and a swan's neck. But the basic objection that he 
leveled at Darwinism — and Darwin himself he held in 

iKarl von den Steinen, " Gedachtnisrede auf Adolf Bastian," ZE 
37:236-249, 1905. Also: Von Eichthofen, ibid,, 249 sq. 


high esteem — was that of the modern experimentalists, 
of Jacques Loeb and Thomas Hunt Morgan. His position 
may be narrowly unhistorical, barrenly skeptical, but it 
was not lacking in scientific respectability.^ 

About Bastian's style, it is also wiser to discriminate 
instead of joining the chorus of cheap gibes. At its worst 
it is surely inconceivably crabbed. To confront Bastian in 
some of his lucubrations is a never-to-be-forgotten expe- 
rience. The astounded reader runs into sentences twenty 
lines and more in length and hacks his way through 
bracketed quotations in Latin, Greek, or Polynesian, 
only to find that he has yet to extricate himself from the 
maze of some major parenthesis. Bewildered by recon- 
dite allusions and unheard-of authors, he is distracted by 
footnote after footnote that lend but mediocre illumina- 
tion until a full stop at last affords a breathing spell. Nor 
are bizarre figures of speech an aid to understanding. 
Geography is suddenly introduced as the "many-breasted 
mother of . . . ramifications spun over the globe"; she 
works "to level the soil for Ethnology," who in turn 
traces "the roots of Psychology embedded in Physiol- 
ogy"; as a result of all which, "Materialism is to see the 
hitherto amorphous torso of her world-view perfected 
. . . by her consecrated wedding to Idealism."® The 
following purports to explain why the Tree of Humanity 
may be glimpsed more clearly on primitive levels: "We 
meet it just and barely as light shoots sprout out of the 
earth, as the little stem puts forth its leaves, displays 
blossoms, is decked in floral splendor, at times perchance 
even affording little fruits . . .; and wherever we meet 
it, we can seize, grip and tousle, strip and pluck it on be- 

2 A. Bastian, Bie VolTcer des ostUchen Asien, 5 : xlf ., Leipzig, 1869. Cf . 
his review of Haeckel's Anthropo genie in ZE, 7:203, 1875; and his re- 
marks iUd., 3:133 sq., 349, 3551, 1871; 8:394, 1876. 

' Der VolTcer gedarike im Aufbau einer Wissenschaft vom Menschen und 
seine Begriindung auf ethnologische Sammlungen, 2, Berlin, 1881. 


half of science, even vivisect it."* Eestraint from pre- 
mature generalization is preached thus : ' ' Thereby would 
be tailored for us a beggar's cloak of mottled shreds and 
patches {Bettelmantel aus buntscheckigem Stiickwerk), 
whereas if we wait calmly for the facts to be gleaned for 
a definite survey, a magnificent peplos will be woven, as 
though spread by Zeus over a sacred oak, as a radiantly 
reflected image of reality. ' ' ^ One further example must 
suffice: "He [an individual] would perish without soci- 
ety, without that unifying community of spirits that, 
swelled by the billowing thoughts of the past, roars along 
in the current of history and in foaming spray surges 
around the barriers" {schdumend im GiscJit die Schran- 
ken umbrandet).^ 

All these illustrations are taken from a single work. 
The cumulative effect of such diction is easily imagined. 
Yet matters are far worse than described. For there is no 
intelligible organization: ideas turn up on the principle 
of free association, with favorite propositions recurring 
at irregular intervals like the leitmotifs of a music 
drama. This is not the biased verdict of a single critic, 
but the general consensus of opinion; Karl von den 
Steinen's obituary reference to the undammed stream 
of ideas (Uberqiiellen der Vorstellungen oJine jede not- 
wendige Hemmung) defines the identical impression. 
Add to this that Bastian rarely deigned to give an exact 
source reference, and the joys of consulting him will be 
fully appreciated. 

Bastian was not unaware of the obstacles he put in 
his readers' path. In rejoinder he argued that the 
primary duty of the day was to garner facts, that his 
travels often led him to places remote from libraries, 
that even from his meager indications experts would be 
able to verify his statements. 

*Z6ic7., 8. "Zfcid., 91. 8J6id., 135. 


Why, then, our insistence on qualifying the popular 
verdict! First of all, if this were the whole story, the 
nature of his influence would be an enigma. Yet he did 
enjoy the esteem of men like Virchow and Tylor; and 
for a generation his ideas loomed important in German 
ethnology. Let us, then, approach him from another 
angle, let us peep into his six-volume travel book on 
Die Volker des ostlichen Asien; at once a very different 
Bastian appears. The narrative is straightforward, often 
vivacious, nay, spiced with humor. Here is a traveler 
sympathetic with his hosts, avid of information where- 
ever he can get it. He watches processions of Burmese 
nobles with their henclmaen bearing betel boxes and hold- 
ing parasols, and listens to the itinerant astrologers in 
the market places. The same man who at home shunned 
the theater attends interminable native performances, 
persevering through their masquerading, conundrums, 
and ribaldry. He chuckles over the pious Buddhist fisher- 
men who never kill their catch but merely compensate 
them for their submergence by drying them in the sun. 
Equally entertaining is the story of the would-be cook: 
after Bastian had painfully persuaded his parents, 
uncles, aunts, and cousins not to restrain the young man 
from accompanying him, the chef himself developed 
scruples : ' * Squirming like a worm on the ground, he 
protested he could never murder an innocent fowl; he 
would cook as many chickens as I might order, would 
roast or chop them into ragout, but never would he con- 
sent to slaughter poultry himself.*' The theological de- 
bates with the king of Burma on the right of self-defense 
are likewise tinctured with a sense of the grotesque.^ 

While such items become rarer in subsequent vol- 
umes of the work, they are not lacking there. Bastian, 
trying to explain points of anatomy to a princely presi- 

^ Bastian, Die Volker des ostlichen Asien, 2:16, 48, 53, 71, 92, 157, 504, 
Leipzig, 1866. 


dent of the Siamese medical college, raises his arm above 
that noble's skull. At once this breach of etiquette is 
brought home to him "by the menacing growls swelling 
like muffled rumbling from the mouths of the crawling 
vassals, for in Siam there is no graver affront than to 
touch a superior's head." Again, there is the banquet 
with the minister of the exchequer at which a bigoted 
abbot glorifies Buddhism in stilted phraseology, taking a 
thinly disguised fling at the barbarian fellow guest who 
seeks wisdom without years of monastic seclusion. Fi- 
nally, in Japan the traveler views dramatic scenes "that 
we could not describe either in Latin or Greek script, but 
which the Japanese, in the company of their wives and 
daughters, viewed with dispassionate equanimity.'"* 

Stylistically, then, Bastian led a double life. It is 
true that a chronological factor enters, since the later 
writings are unquestionably more distorted than their 
predecessors. But this is not the whole story. In the last 
two volumes of the Reisen the clear, even, sprightly dic- 
tion of the narrative contrasts sharply with the trope- 
laden, labyrinthine manner of the prefaces, where the 
author expounds his creed, stressing the danger of 
system-mongering and the dependence of psychology on 
ethnology. Bastian, then, became confused in proportion 
as he discussed theory; and his progressive obscurity 
was probably due to his increasing concern with general 

What are these principles'? The one that most di- 
rectly affected contemporary thought was unquestion- 
ably his belief in independent development. He did not, 
indeed, deny diffusion, but he insisted that in each case 
it must be proved by detailed evidence. His position was 
thus diametrically opposed to both Graebner's (page 
158) and to Balfour's (page 28) since he contended that 
by a general law the psychic unity of mankind every- 

8 Op. cit., 3:71, 82, 1867; 5:325, 1869. 


where produced similar ''elementary ideas" (Elemen- 
targedankcn). Only different external stimuli would 
evoke different responses, whence the origin of geo- 
graphical provinces. At a higher stage, contact with 
other cultures may supersede such physical conditions as 
a stimulus, whence history and cultural development 
proper {Geschichte und eigentliche Kulturentwicklung) . 
But compared with the basic laws, these historical causes 
are of subordinate significance. 

Bastian's faith in supposedly uniform laws of 
growth, in a "genetic principle" through which lower 
and simpler phenomena become higher and complex, 
shows that cultural evolutionism may very well go hand 
in hand with a repudiation of biological transformism. 
Further, notwithstanding appearances, Bastian never 
preached the chaotic accumulation of raw data as an 
ultimate goal. He clamored for the harvesting of facts 
because they were in danger of being irretrievably de- 
stroyed by the leveling of modern civilization. And this 
was deplorable because for a proper perspective science 
needed samples of all cultures, past and present. 

It was Bastian's belief in a law of growth that em- 
boldened him to herald an "applied anthropology." 
Insight into normal processes would stave off pathologi- 
cal deviations, safeguard the national exchequer, avert 
"the formidable miscarriages" of Anglo-Indian admin- 
istration exposed by Maine, affect the weal and woe of 
millions. Bastian's idea thus foreshadowed the work of 
the Africa Institute today. 

Nor should we overlook Bastian's views on psychol- 
ogy. Decades before Rivers (page 169), he argued that 
a science of mental life must take cognizance of ethno- 
graphic data, because the "individual's thinking is made 
possible only by his functioning in a social group." ' 

■ Bastian, Der Volkergedanke. . . ., 1881. 


Thus, Bastian anticipated many of his successors. 
His gospel of saving vanishing data is Haddon's; his 
insistence on proof of assumed historical connections 
coincides with Boas'; like Thurnwald and Radcliffe- 
Brown, he postulates laws of sequence; like Malinowski 
he would apply anthropology to colonial government. 
And what are his geographical provinces but the culture 
areas of later research? Add his unchallenged achieve- 
ment of founding a great ethnographic museum, and it 
becomes intelligible that he loomed as a major figure of 
his time. 

At present we can hardly assign to him quite that 
distinction. Who now reads Bastian I It is one of the 
numerous paradoxes in his career that this untiring 
preacher on the need of complete data has left not a 
single standard monograph. Neither his early book on 
West Africa nor the learned tomes on Die Kulturldnder 
des alien Amerika (1878) can be considered indispen- 
sable to workers in these fields. As von den Steinen put 
it, *'he was not an ethnographer in the narrow sense of 
the term." He was too restless to settle down in one spot 
and immerse himself into the life of a particular people. 
For in practice his interests were not nearly so broad as 
in theory. At bottom what lured this scorner of meta- 
physics was the world-view of men in different ages and 
places, their conceptions of cosmology, cosmogony, and 
eschatology. On his Eastern Asiatic travels he observed 
keenly enough, but the one thing he studied systemati- 
cally was Buddhism. 

Theoretically, we find a host of sound and stimulat- 
ing ideas with an equal dearth of definitive results. Why 
did he not define the geographical provinces of the world? 
Why do we constantly hear of a law of development with- 
out ever seeing the proof for its existence? How are 
pathological modes of growth to be distinguished from 
normal ones? Bastian offers no answers. To formulate a 


problem, experiment in thought as to possible solutions, 
amass relevant material, and then logically to exclude 
one conclusion after another, was obviously beyond him. 
That is why he is a forerunner — a forerunner with a 
variously stirring message — but not a leader to salva- 



The decade following Boucher de Perthes' recogni- 
tion was a period of feverish activity; and the charac- 
teristic trends, comparative and evolutionist, are 
nowhere better exemplified than in the studies of early 
law. Naturally enough, they were pursued first by his- 
torians and jurists, but these influenced anthropological 
thought even when they themselves made little or no use 
of primitive data. For example, Numa-Denys Fustel de 
Coulanges (1830-1889) in La cite antique (1864) stressed 
the differences between classical and later forms of juris- 
prudence. Thus names, he pointed out, had developed 
quite differently in antiquity and in medieval times. In 
the Middle Ages the baptismal name was the true indi- 
vidual name and patronymics only evolved much later, 
while among the Romans the patronymic was earlier 
and more important. The early Teutons owned individu- 
ally what they harvested, but not the soil on which their 
crops were raised; in Greece, land was always private 
property. Whether these views are borne out by later 
specialist research or not, they gave a hint of the range 
of variation in legal institutions, enlarging the bounda- 



ries of juridical provincialism. Fustel de Coulanges' re- 
sults influenced Morgan's treatment of the clan. 
Moreover, we note that his position is militantly func- 
tionalist: institutions are declared unintelligible except 
in their context, law cannot be understood apart from 


A Swiss jurist, J. J. Bachofen (1815-1887), exerted 
a more powerful influence on ethnographic thought, for 
he was the first to throw into relief the existence of ma- 
trilineal descent. Moreover, while primarily concerned 
with classical antiquity, he also utilized material on 
primitive peoples. 

Though by profession a lawyer, Bachofen was, 
above all, a philologist who sought to penetrate to the es- 
sential core of classical life. ''Roman law," he writes, 
''always appeared to me as a part of ancient, especially 
of Roman philology, hence as a section of a larger whole, 
embracing the study of classical antiquity in its en- 
tirety. ' ' ^ Indeed, according to an admiring expositor, 
Bachofen 's chief claim to fame rests on his magnificent 
vision of ancient culture as a connected unit, while his 
contribution to sociology is secondary and incidental. 
However that be, Bachofen 's only work that concerns 
us. Das MutterrecM (1861), teems with references to 
classical mythology and is studded with Greek and Latin 

The essence of the argument is set forth in a lengthy 
introduction (pp. v-xxxiii). Starting from Herodotus' 
account of the Lycians as matrilineal, Bachofen deduces 
from it a coherent system of law antecedent and anti- 
thetical to the patriarchal principle of antiquity. Not 
only did children take their mother 's name in Lycia, but 

1 J. J. Bachofen, Selbsthiographie und Antrittsrede uber das Natur- 
recht (ed., Alfred Baeumler), 13, 1927. 


women ruled the household as well as the state. Apply- 
ing the principle of survivals, the author interprets 
mythological references to outstanding women as relics 
of a one-time gynaecocracy. Further, he argues, the very 
rigor of Roman patriarchy implies an inimical principle 
that had to be combated and ousted. 

Like Fustel de Coulanges, Bachofen is an aggressive 
functionalist: ''The hegemony of maternity in the 
family can not be conceived as an isolated phenomenon." 
Hence a rule for reckoning descent is necessarily only 
one link in a chain of ideas. Matrilineal descent, in con- 
trast to patrilineal, exalts the left above the right side; 
night above day; the moon above the sun; the youngest 
above the oldest child. Even the notion of general liberty 
and equality naturally flows from " chUd-bearing ma- 
ternity" {aus dem gebdrenden Mutterthum) . But how 
could the weaker sex attain ascendancy? Bachofen an- 
swers: Through woman's aptitude for religion. Specifi- 
cally, woman in the flesh represented the Earth goddess 
{tellurische Urmutter) ; where there is feminine domi- 
nance there is also a chthonic faith, connected with 
Demeter or an equivalent figure. It is a basic tenet of 
Bachofen 's that secular gynaecocracy merely reflects 
the primary phenomenon, viz., the cult of a female deity. 

In his chronology Bachofen is a typical evolutionist 
of the old school. Once more a belief in progressive 
stages appears independently of modern biological 
theory, for it is improbable that the Swiss jurist should 
have been affected by Darwinism when he wrote Das 
Mutterrecht. Gynaecocracy, we must note, was not the 
earliest social condition: it came as a reform supersed- 
ing promiscuity (Hetdrismus). Yet this antecedent stage 
sprang not from sheer lust but from an idea-system de- 
creeing that woman was not created to fade in the arms 
of a single mate. Exclusive possession by a male was an 
offense against God, to be expiated by periodic ceremo- 


nial prostitution. To Bachofen, then, the historic worship 
of Dionysus was a step backward : woman, naturally sen- 
sual, hence fascinated by a phallic system, relapsed from 
the severe gynaecocracy of the Amazons into the lowest 
form of tellurism, a fleshly venery. 

In principle, however, Bachofen postulates universal 
stages: Primeval promiscuity was followed by a revolt 
of woman, who craved delivery from such humiliation. 
The result was Amazonian assertiveness. Once in the 
saddle, however, women devoted themselves more and 
more to peaceable pursuits, among other things invent- 
ing agriculture. At this point, it is assumed, develop- 
ment diverged in detail; in some regions women lost 
their superiority in domestic affairs, elsewhere they 
yielded political power. Thus, paternity, a higher prin- 
ciple, came to triumph, ushering in not merely a social 
change but a revolutionary world- view: the celestial 
Apollonian idea gained the victory over tellurism; the 
right side came to outrank the left, day conquered night, 
spirit subjected matter. 

Though our author rests his case preponderantly on 
classical data, he culls parallels from primitive areas, 
such as Airican instances of feminine prestige (Ba- 
lunda) and Negro as well as American cases of inherit- 
ance by a sister's son. Here again is an illustration of 
his f unctionalism ; for nepotic inheritance can be inter- 
preted as a vestige of pristine gynaecocracy only on the 
assumption that a trait must fit into a logically linked 
setting. Bachofen also treats a phenomenon that looms 
large in subsequent discussion, the *'couvade," i.e., the 
custom of the father's taking to his bed when his wife is 
delivered of a child. Bachofen foreshadows some later 
interpreters by suggesting that the husband must be 
made to appear as a fictitious second mother in order to 
lay claim to his offspring.^ 

2 J. J. Bachofen, Das Mutterrecht, 105-110, 255 f., Stuttgart, 1861. 


At first blusli, a bald outline of Bachofen's scheme 
may provoke a wholly negative reaction. Much of it is 
fantastic, the whole of it shot through with more than a 
flavor of mysticism. Specifically, he thoroughly confused 
the phenomenon of matrilineal descent with a matriarch- 
ate, which has never been authentically demonstrated 
as its concomitant. Nevertheless, something can be said 
on the positive side. However specialists of today may 
appraise his classical studies, Bachofen applied to this 
field ideas which modern ethnography accepts as axioms 
for the investigation of alien societies. Like Fustel de 
Coulanges, he is emphatically a functionalist in reject- 
ing the study of a single aspect of civilization and espe- 
cially in connecting social structure with religious 
practice. Further, Bachofen, while full of emotional 
evaluations, was at least willing to contemplate all the 
facts of Greek culture — including its crudities — which 
the classicist guild of his day ignored from aesthetic 
snobbishness. Again, he repudiated interpretation of 
myths in terms of individual psychology, because they 
were the product of an unconscious social activity, the 
Volksgeist. Whatever we may think of this conception, 
its critical half forestalls error, and its positive side at 
least harmonizes with some modern trends.^ 

To sum up, Bachofen was the first to direct atten- 
tion to matrilineal descent as a problem; and he com- 
bined this service with some valid general principles for 
approaching alien cultures. 


Independently of Bachofen, though somewhat later, 
J. F. McLennan (1827-1881), discovered the importance 

3 Alfred Baeumler, in Der Mythus von Orient und Occident, eine 
MetaphysiTc der alten Welt; aus den WerJcen von J. J. Bachofen mit einer 
Einleitung von Alfred Baeumler herausgegeben von Manfred Schroeter, 
cclix-cclxxx sq., Miinchen, 1926. 


of miitrilineal descent; and though a chasm separated 
the combative, logical Scotch lawyer from the mystical 
Bale patrician, several of their speculations happen to 

Like so many of his contemporaries, McLennan was 
essentially a parallelist. ''All the races of men have had, 
to speak broadly, a development from savagery of the 
same general character."* Consistently with this, he 
stresses survivals, in fact, his interest in them prompted 
his first essay on Primitive Marriage (1865), subse- 
quently republished in the Studies in Ancient History: 
''wherever we observe symbolical forms, we are justi- 
fied in inferring that in the past life of the people em- 
ploying them, there were corresponding realities. ' ' * 

In particular it was the ritual of bride-capture that 
stimulated McLennan. In the same year Tylor briefly 
listed instances of sham bride-kidnapping, summarizing 
the facts in these words: "In these cases the abduction 
is a mere pretence, but it is kept up seemingly as a relic 
of a ruder time when, as among the modern Australians, 
it was done by no means as a matter of form, but in 
grim earnest. ' ' * Like McLennan, in other words, Tylor 
assumed systematic wife-capture; but what remained an 
incidental remark of Tylor 's became the very keystone 
of McLennan 's thinking. 

As a purely logical construct that scheme inspires 
respect. McLennan 's was a vigorous intellect, eager to 
read meaning into the oddments of available ethno- 
graphic literature and quite unimpressed by authority. 
Unfortunately his independence was coupled with a 
pugnacity that barred fairness to opponents. Worse than 
that, he loved to practice dialectics in vacuo, to speculate 
about conditions not only unknown but unknowable. 

* J. F. McLennan, Studies in Ancient History, 301, London, 1886. 
^Ibid., 5. 

8 E. B. Tylor, Eesearches into the Early History of ManMnd, 284, 
London, 1865. 


Two terms coined by McLennan have become com- 
mon scientific property. ''Endogamy" defines a condition 
in which ' ' members of a family or tribe are forbidden to 
intermarry with members of other families or tribes" 
(p. 24); ''exogamy" designates "prohibited marriage 
within the tribe" (p. 27). According to the scheme, early 
tribes were exogamous and chronically at war with one 
another. Thus, they could get wives only "by theft or 
force." Later, when intertribal relations grew more 
amicable, abduction ceased to be necessary and was pre- 
served as a mere symbol of past reality. 

At this point we must ask what is to be understood 
by a "tribe." In a spirited re joiner to McLennan 's attack 
on Systems of Consanguinity and Affinity, Morgan (page 
62) insisted that such a thing as tribal exogamy did not 
exist; it was merely the clan ("gens"), a tribal subdivi- 
sion, that was exogamous.^ This certainly holds for most 
American cases, but McLennan 's error is less crass than 
it at first sight appears. When he treats of Australians, 
for instance, he envisages their exogamous tribes as ' ' ex- 
ceedingly numerous and exceedingly small, being a 
species of family groups" (p. 41). In other words, he 
foreshadows Radcliffe-Brown's "hordes "even though he 
does not know that these units are patrilocal and patri- 
lineal. Likewise, when hypothetically picturing Scotch 
developments, McLennan conceives the primeval tribe as 
a localized clan on the Australian pattern, while the 
equivalent of Iroquois organization is considered a later 
type, due to "the interfusion of clans" but with exogamy 
persisting for persons sharing the horde name (pp. 56- 
58, 77 f.). McLennan fails to prove the alleged sequence 
and to discriminate clearly between the local exogamy of 
Australia and the clan exogamy of the Iroquois; but 
apart from this, his position is intelligible and even 

■^ Lewis H. Morgan, Ancient Society, Note to Part III, New York, 1877. 


Later research has supported McLennan 's claim 
that exogamy is a widely prevalent principle of law. On 
the other hand, there is nothing to show that it regularly 
accompanies bride-abduction. His own data do not bear 
out the correlation : for America he cites capture among 
natives of the Orinoco and Amazon, while his exogamous 
instances are from the Eastern United States and the 
Chaco. His logic errs in another argument. Granting uni- 
versal hostility and local exogamy, bride-capture would 
indeed become a necessity; but McLennan argues back- 
ward that wherever there is abduction, real or symbolic, 
exogamy also must have been present if it is no longer 
there. This does not follow, because conceivably bride- 
kidnapping might be due to an alternative cause. 

For McLennan exogamy was not an ultimate datum, 
but resulted from a general practice of female infanti- 
cide. Rejecting the idea that an instinct prevented early 
man from mating with his kin, he substitutes social 
causes for this biological factor. Tribes surrounded by 
enemies would find girls a source of weakness, whence 
wholesale destruction of female infants. The entire 
course of evolution, then, assumes the following shape. 
In the beginning there were promiscuous hordes, which 
killed off the majority of baby girls so that males came 
to preponderate. However, promiscuity came to be miti- 
gated by arrangements whereby a small set of men at- 
tached themselves to a particular woman. This is 
"archaic" polyandry, wherein the male partners of a 
woman bore no necessary relation to one another. It 
evolved into the ** fraternal" variant when only sons of 
the same mother shared a wife. This ushers in another 
cardinal feature of the system, viz., that under primeval 
conditions kinship could be reckoned only through the 
mother because paternity was uncertain. 

While not yet introducing the terms *'matrilocal" 
and "patrilocal," McLennan utilizes the concepts they 


represent. Under archaic polyandry the wife lives with 
her mother and brothers, her children being born in this 
matrilineal household. Later she has a house of her own, 
where her husbands visit her according to fixed rules, 
the offspring remaining affiliated with her kin. In the 
subsequent stage of fraternal polyandry, woman is de- 
tached from her own family and passes into her hus- 
bands', where her children are born and henceforth 
belong. To use modern parlance, patrilineal descent re- 
sults from patrilocal residence with fraternal polyandry. 

The levirate — that widespread custom by which a 
brother inherits a widow — is conceived by McLennan as 
a natural part of the polyandrous scheme and as a signifi- 
cant survival of it. Only, he avers, where there is or has 
been polyandry do brothers succeed in preference to 
sons. Similarly, the American practice of calling a 
paternal uncle by the same term as the father, as well as 
correlated ways of classing kindred ''bear the stamp of 
a polyandrous origin." Our author firmly believed in a 
strong bond uniting the several traits that especially 
interested him. Thus, he infers that where there is 
polyandry, there must once have been a rule of ''kinship 
through females only"; that all exogamous races must 
have been originally polyandrous, hence must once have 
"recognized blood- ties through mothers only." Here is 
another pioneer with a marked functionalist bias and, 
logically enough, with a proclivity for the survival argu- 

McLennan must also be noted as one of the very first 
to see totemism in broader perspective. In a series of 
papers he endeavored not to explain the origin of totem- 
ism but to prove that the civilized nations of antiquity 
had all passed through a totemic stage, which was thus 
conceived as a universal step in human civilization. Con- 
sistently with his general position he would not accept 
the view that totems were mere emblems but contended 


that here, too, a reality must have preceded the symbol- 
ism. In other words, he held that there was an ancient 
worship of animals and plants which subsequently 
evolved into higher cults but sporadically left relics in 
the form of emblems. The misconceptions in these essays 
are profound. Apart from the parallelist faith in uni- 
versal stages, we note the erroneous idea that totemism 
generally implies worship and the refusal to ascribe to 
early man any belief in benign beings apart from the 
totemic cults. The confusion of totemism with animal 
worship, while pardonable at the time, led many others 
astray and had to be dispelled three decades later in a 
masterly paper by Tylor.* 

Why was this vigorous intellect so frequently misled 
in its conclusions ? The explanation is simple : they lacked 
the support of facts ; and their author was too enamored 
with his reasoning to entertain rival hypotheses. Empiri- 
cal data do not bear out either the universal enmity of 
primitive groups or the prevalence of female infanticide. 
Nor, as Morgan rightly contended in the rejoinder already 
cited, is there any reason to suppose that polyandry — 
decidedly rare as a fixed institution — ever characterized 
an epoch of human development. This critic likewise 
correctly repudiated McLennan 's notion of "kinship 
through females only," seeing that primitive tribes 
regularly designate relatives on both sides by specific 
terms irrespective of whether the rule of descent is 
matrilineal or the reverse. As for bride-capture, Thurn- 
wald admirably summarizes the essential facts about this 
''scientific myth." A new status in the life cycle evokes 
a social demand for some change in conduct, whence a 
ritual expression of the stage reached. A transitional 
state brings with it mental inhibitions, which may be 

8 J. F. McLennan, "The Worship of Animals and Plants," The Fort- 
nightly Review, 6: 407-427, 562-582, 1869; Ihid., 7:194-216, 1870. E. B. 
Tylor, "Eemarks on Totemism with Especial Eeference to Some Modern 
Theories Respecting it," JEAI 1:138-148, 1899. 


dramatized, as a result of which we find McLennan 's 
symbol of capture. As E. C. Parsons has shown, there 
are instances of reluctance on the part of the groom, not 
the bride ; and more recently Thurnwald himself records 
groom-abduction in New Guinea.® To concentrate on 
bride-kidnapping as the phenomenon to be studied was 
to emphasize unduly one extreme variant of the natural 
context requiring interpretation. 

Notwithstanding these strictures, McLennan 's place 
in the history of ethnology remains important. He added 
to the concepts and nomenclature of nascent comparative 
jurisprudence; his speculations, even when sterile from 
want of material, stimulated his contemporaries by the 
vigor with which they were set forth; and his trenchant 
criticisms, however unfair in part, sometimes hit the 


In 1861 — the year of Das Mutterrecht — an English 
author published the book that above all others laid the 
foundation of comparative jurisprudence, Ancient Law. 
Henry James Sumner Maine (1822-1888) was Regius 
Professor of Law at Cambridge, and went to India as 
Legal Member of the Supreme Council of the Governor- 
General in 1862, afterwards becoming Vice-Chancellor 
of the University of Calcutta. Returning to England in 
1869, he became Corpus Professor of Jurisprudence at 
Oxford and later (1887) Whewell Professor of Interna- 
tional Law at Cambridge. His principal works, besides 
that already cited, are : Village Communities in the East 
and West (1871) ; Early History of Institutions (1875) ; 
and On Early Law and Custom (1883).^° 

» Richard Thurnwald, Die menschliche Gesellschaft, 2:104 sq., Berlin u. 
Leipzig, 1932. Elsie Clews Parsons, "Holding Back in Crisis Ceremonial- 
ism," AA 18:44 f., 1916. 

i^'M, E. Grant Duff, Sir Henry Maine: A Brief Memoir of His Life, 
London, 1892. 


Maine enlarged the boundaries of traditional juris- 
prudence by comparing Roman law and modern Western 
systems with those of India and Eastern Europe. Un- 
like McLennan and Morgan, he dealt only incidentally 
with primitive usage so that his predominantly legal 
training puts much of his discussion beyond the etlmol- 
ogist's competence. In Ancient Law the reader constantly 
faces the unfamiliar problems of traditional jurispru- 
dence. The very terminology is forbidding, the facts lie 
beyond his scope. He is unable to judge conclusions on 
the process of feudalization, Salic law, or Brehon usage. 
Nevertheless, what he does carry away is part of his 
most valuable intellectual equipment. 

Wliile McLennan impresses us by the bumptious 
vigor of his intellect, Maine is the embodiment of serene 
wisdom coupled with unusual subtlety. He thus achieved 
a series of concepts which turned out to hold valid for 
primitive law no less than for the Indo-European peoples 
whose legal development was his primary field of study. 
Among the most important of his ideas is the antithesis 
of the blood-tie and the territorial bond, a distinction 
adopted by Morgan and of great significance for sub- 
sequent ethnological thinking. Maine — and Morgan after 
him — erred in denying that early and primitive peoples 
ever united on the basis of territorial grouping, but cer- 
tainly kinship bonds are much more potent on that level 
than those of mere contiguity. What is more, the con- 
ceptual distinction of the two types of social solidarity 
remains valid and illuminating. Maine further brought 
out the contrast of tort and crime; of status and con- 
tract; of the inalienable real estate of early times and 
the merchantable land of higher civilizations ; of archaic 
formalism and modern emphasis on substance. Every 
one of these distinctions is definitely applicable to ethno- 
graphic data and brings enlightenment. The same holds 


for the principle of legal fiction, which Maine also treated 
at length. 

What astonishes us, above all, in Maine is his 
thoroughgoing historical-mindedness. At a time when 
Lubbock was forever undergoing emotional spasms over 
"revolting" primitive customs, Maine wrote: "It is not 
the business of the scientific historical enquirer to assert 
good or evil of any particular institution. He deals with 
existence and development, not with its expediency." 
Neither a mystic like Bachofen nor a romanticist like 
Rousseau, Maine, through the incorruptible medium of his 
common sense, sees things as they are and were: "Al- 
though there is much in common between the Present and 
the Past, there is never so much in common as to make 
life tolerable to the men of the Present, if they could step 
back into the Past. ' ' " 

Maine was a true historian in another sense. Mc- 
Lennan speculated about the unknowable social condi- 
tions of a remote past and traced a generalized scheme of 
development; and we shall find in Morgan's work an 
even more elaborate system of this category. Maine may 
occasionally drop a word of homage to "continuous 
sequence, inflexible order, and eternal law in history," 
but this sop to regnant fashion agrees neither with his 
practice nor with his philosophy. "So far as I am 
aware," he writes elsewhere, "there is nothing in the 
recorded history of society to justify the belief, that, 
during that vast chapter of its growth which is wholly 
unwritten, the same transformations of social constitu- 
tion succeeded one another everywhere, uniformly if not 
simultaneously." He thus explicitly rejects the idea 
"that human society went everywhere through the same 
series of changes." He is avowedly interested in "the 
real, as opposed to the imaginary, or the arbitrarily as- 

" Village Communities, 3d cd., 230, 289 f., London, 1876. 


sumed, history of the institutions of civilised man."*" 
In a period of evolutionary schematism such a posi- 
tion proved unbelievable. Because Maine dealt with early 
Indo-Germanic law he naturally stressed patriarchal 
features; as a result he was attacked for championing 
the theory of universal development from a patriarchal 
stage. Yet he explicitly repudiated this view along with 
McLennan's and Morgan's because both are "open to 
considerable objection as universal theories of the gene- 
sis of society. ' ' This is what he really said : ' ' There are 
unquestionably many assemblages of savage men so de- 
void of some of the characteristic features of Patriar- 
chalism that it seems a gratuitous hypothesis to assume 
that they had passed through it." Also : "There has been 
room . . . for many courses of modification and develop- 
ment, each proceeding within its ow^n area." Finally, 
since he knew actual history, Maine recognized the force 
of diffusion: the Belgian constitution, he pointed out, is 
not an independent parallel of the British but was copied 
from one of its copies.^^ 

As already explained, Maine made only sparing use 
of ethnographic data and was doubtless imperfectly ac- 
quainted even with those available in his day. He exag- 
gerated the power of sexual jealousy as a psycho- 
sociological constant — a factor real enough in savagery 
but often quite differently manifested there. But as a 
rule Maine's superb insight enabled him to see clearly 
even where he was not particularly conversant with the 
material. He showed that exogamy and endogamy were 
not at all mutually exclusive except with reference to the 
same unit ; and he saw through the flimsiness of much so- 
called evidence from primitive societies. Thus, he re- 
jected the cock-and-bull stories about the immorality of 
the Andaman Islanders and was triumphantly supported 

^Ibid., 266; On Early Law and Custom, 219, 192, 201, London, 1890. 
13 Ibid., pp. 204, 218, 285. 


by later research. His closing words on that subject 
should still be the field investigator's motto: ''There is 
no subject on which it is harder to obtain trustworthy 
information than the relations of the sexes in com- 
munities very unlike that to which the inquirer belongs. 
The statements made to him are apt to be affected by 
two very powerful feelings — the sense of shame and the 
sense of the ludicrous — and he himself nearly always 
sees the facts stated in a wrong perspective. Almost in- 
numerable delusions are current in England as to the 
social condition in regard to this subject of a country so 
near to us in situation and civilisation as France. ' ' " 

Finally, like Fustel de Coulanges, Bachofen, and Mc- 
Lennan, Maine was a functionalist in treating phenomena 
not as discrete but as interrelated. He recognized the 
contacts of archaic law with religion and morality, and 
pointed out the military weakness inherent in a system of 
matrilocal residence. Among the Southern Slavs, he sug- 
gests, the assertion of individual property rights, at first 
exceptional, sapped the whole scheme of the house com- 
munity. In India, again, the establishment of local courts 
unintentionally altered Indian law because legal rights, 
obligations, penalties, and political superiority are all 

A picture of the pioneers of comparative law would 
be incomplete without Lewis H. Morgan; but his in- 
fluence has been so potent that it warrants separate treat- 

^^lUd., 215 sq., 222, 278, 231. 

^^ IMd., 288, 253; Village Communities, 67-76. 



By a freak of fortune Lewis H. Morgan, whose 
main original contribution to ethnology lies in its 
most aridly technical field — kinship terminologies — has 
achieved the widest international celebrity of all anthro- 
pologists. Naturally this is not due to his solid achieve- 
ments, but to a historical accident: his Ancient Society 
(1877) attracted the notice of Marx and Engels, who ac- 
cepted and popularized its evolutionary doctrines as be- 
ing in harmony with their own philosophy. As a result it 
was promptly translated into various European tongues, 
and German workingmen would sometimes reveal an un- 
canny familiarity with the Hawaiian and Iroquois mode 
of designating kin, matters not obviously connected with 
a proletarian revolution. Even in America Morgan's 
book has long been most readily accessible in the in- 
expensive reprint issued by a Socialist firm in Chicago. 
In the immediate past the bourgeois lawyer, who never 
severed his connections with Christian orthodoxy, has 
been officially recanonized by the present Russian regime. 
Its spokesmen declare his work "of paramount impor- 
tance for the materialistic analysis of primitive com- 



munism"; its Academy of Sciences has published a 
translation of Ancient Society for its "Classics of Scien- 
tific Thought"; and there are rumors of veritable pil- 
grimages to Rochester, New York, as the focus of 
Morgan's new dispensation. Neither falling under the 
hypnosis of such hero-worship nor recoiling from its 
quaint manifestations into the opposite extreme, we shall 
endeavor to see Morgan against the background of his 

Like McLennan and Maine, Lewis H. Morgan (1818- 
1881 ) was a lawyer, but unlike them he was not merely a 
philosopher of institutions but also a first-hand observer 
of aboriginal custom. His League of the Ho-de-no-sau- 
nee or Iroquois (1851) is one of the best earlier descrip- 
tive reports of Indians ; and his brief visits to other tribes 
produced much valuable material on social organization. 
This, indeed, was the topic that especially attracted him 
in the field or library, while he unaccountably neglected 
other aspects of culture. In the opening chapter of 
Ancient Society, for instance, he dismisses supernatural- 
ism with the dictum that ''all primitive religions are 
grotesque and to some extent unintelligible." However, 
he did not ignore Iroquois technology, to which indeed 
he paid considerable attention. We must also credit him 
with a sympathetic attitude towards the Indian, such as 
was not by any means general at the time. Theoretically, 
Morgan's direct contacts with tribes east of the Rocky 
Mountains were not an unmixed blessing, for they in- 
clined him to see all American Indians in the image of 
those he knew, i.e., with uniformly democratic govern- 
ment and organization. Had he begun his studies among 
the Eskimo or Paiute, his general views might have been 

1 Bernhard J. Stern, Lewis Henry Morgan, Social Evolutionist, Chicago, 
1931; also, idem, "Selections from the Letters of Lorimer Fisoii and A. W. 
Hewitt to Lewis Henry Morgan," A A, 32:257-279, 419-453, 1930. 



What, then, were these views? His aim was to fur- 
nish a coiiii)lote scheme of institutional prog-ress, with 
special attention to marriage, kinship, government, and 
property. It resembled the systems of his contemporaries 
in being evolutionary and avowedly opposed to the 
theory of degeneration as applied to savages. But it was 
laid out on a vaster scale, being much more inclusive than 
Maine's in dealing with humanity at all periods and 
places, while the facts, in no small measure first noted by 
Morgan himself, were better authenticated than Mc- 
Lennan 's. What is more, though Morgan intensively 
studied only relatively few aspects of group life, he de- 
fined these in relation to the rest. He divided all history 
into three main stages — Savagery, Barbarism, and Civi- 
lization — and correlated each with economic and intel- 
lectual achievements. Savagery was the period before 
pottery; Barbarism, the ceramic age; Ci^dlization began 
with writing. The first two major periods were subdivided 
each into a lower, middle, and upper status, each provided 
with its signpost. Thus, the upper status of Savagery is 
heralded by the use of bow and arrow ; the upper status of 
Barbarism, by iron tools. So far the periods and sub- 
periods are defined by single traits. But actually Morgan 
characterized them by a whole series of features, so that 
major inventions appeared as the correlates of such and 
such economic activities, social customs, and political in- 
stitutions. Thus he offers a comprehensive scheme of 
cultural wholes far beyond anything attempted even by 
the systematic McLennan. Morgan's Ancient Society was 
a synthesis of sociological material that for the first time 
brought together material on Australian and American 
natives, on ancient Greece and Eome; and all this in an 
orderly arrangement prescribed by an evolutionary 
doctrine. No wonder the book was hailed in many 
quarters as a classic. 

In judging this landmark we must recollect that in 


1877 immense areas of the globe were imperfectly de- 
scribed. Yet the gaps in Morgan's knowledge about facts 
thoroughly determined in his time are amazing; and, as 
usual, it is largely ignorance that accounts for glaring 
errors in theory. Along certain chosen lines he was in- 
comparably ahead of his period; in others he neglected 
data that should have been at his fingertips. His treat- 
ment of the Polynesians is inconceivable. Though Cap- 
tain Cook's observations must have been accessible, 
Morgan puts this horticultural, sophisticated people in 
the same class with the Australian hunters. Lacking bow 
and arrow, both races are degraded to the middle status 
of Savagery — below the level of the rudest North 
Americans. This is taking a classificatory device far too 
seriously ! Even so, it is hard to understand how Morgan 
could have missed the social strata of the caste-ridden 
Oceanians. Similarly, enough about Negro societies was 
available to suggest the ever-recurring establishment of 
autocracy on the Dark Continent. Nevertheless, Morgan, 
seeing primitive mankind in Iroquois terms, denied that 
monarchy could appear prior to Civilization, i.e., before 
literacy: monarchy and a clan system, he averred, were 
incompatible. Aristocracy, again, could not evolve before 
the later period of Barbarism, i.e. the Iron Age. This 
dogmatism happened to yield a valuable by-product — 
the scrutiny of Spanish chronicles with their extravagant 
descriptions of an Aztec empire. But the total picture of 
ancient society that resulted was curiously distorted. It 
not only did violence to African and Polynesian facts, 
but misrepresented in part even the American aborigines 
since slavery and class stratification were thoroughly 
established for British Columbia. 

Morgan's discussion of Polynesia at last reveals one 
pioneer who was not a functionalist at heart, who was 
willing to ferret out the minutiae of Oceanian kinship 


systems although he apparently remained in utter dark- 
ness as to otlier facts of their society. 

One other important factual deficiency must be 
noted. ^lorgan, who had himself described fraternities 
among the Iroquois, found no place in his system for 
clubs or any other primitive associations based on 
voluntary affiliation. Yet on the American Plains the 
"Dog Soldiers" were widespread in his day and several 
decades before had been accurately described in a famous 
travel book. While according to Morgan the only impor- 
tant unit in the career of a Mandan or Hidatsa would be 
the clan, he could have learned from Prince Maximilian 
of AYied-Neuwied that the military organizations of these 
Indians largely dominated social life. Their police 
activities, indeed, bore directly on that ''Idea of Govern- 
ment" to which fifteen chapters are devoted in Ancient 
Society. That Morgan omitted Melanesian and African 
secret societies is more intelligible, but his neglect of 
well-known and by no means obsolete phenomena in the 
United States is hard to understand. The skew vision of 
primitive life imposed on ethnologists by Morgan's dis- 
regard of associations was not corrected until Heinrich 
Schurtz's Altersklassen und MdnnerbUnde (p. 99). 

In his general approach to culture history, Morgan 
was a typical evolutionist firmly convinced that the high- 
est known, i.e. mid- Victorian, condition of society had 
been gradually achieved through a series of stages start- 
ing with the very antithesis of the glorious present. In 
other words, believing in a law of progress, he failed to 
maintain the detachment prescribed by Maine and gave 
ethical appraisal to the facts he encountered. Essentially, 
Morgan held conditions of society to be "substantially 
similar" at any particular stage of development. Inso- 
far as he recognized differences he traced them to diver- 
sity of physical environment, a point thus shared with 
Bastian. Specifically, he often mentioned "the unequal 


endowments of the two hemispheres" to account for the 
lack of Old World elements, such as livestock in America. 
Yet even in the Old World the very same domestic 
beasts were quite differently utilized by Egyptians, 
Chinese, and Turks, so that ''equal endowments" may 
evoke unlike responses. 

Occasional qualms were not lacking. "The phrase 
'similar conditions of society,' which has become techni- 
cal," he writes, "is at least extremely vague. It is by no 
means easy to conceive of two peoples in disconnected 
areas, living in conditions precisely similar. ' ' ^ Never- 
theless, as a rule he adhered to the principle that "the 
experience of mankind has run in nearly uniform chan- 
nels." And, so believing, he freely extrapolated where 
authentic information was wanting ; although admittedly 
he had no documentary proof of female descent in Greek 
and Latin societies, the supposedly universal law that 
any patrilineal organization necessarily grows out of a 
matrilineal one made him infer that Hellas and Rome 
must once have been matrilineal. In other words, Morgan 
credited himself with possessing a generally valid 
scheme of sequence by which unknown events could be 
safely deduced. This is the point at which the historians 
of culture level their critical batteries : culture, they de- 
clare, is far too complex to be reduced to chronological 
formulae; its development is mainly divergent, not 

Morgan was not unduly disturbed by cultural loans, 
though he freely admits them. The ancient Britons had 
iron tools, hence should be assigned to the upper status 
of Barbarism. But because their social organization is 
rude, they are put into the middle status: "The vicinity 
of more advanced continental tribes had advanced the 

2 Systems of Consanguinity and Affinity, 472 f., Washington, 1871. 

3 B. Lauf er, Dokumente der indischen Kunst ; erstes Heft : Malerei, das 
Citralalshana, nach dem tibetischen Tanjur herausgegehen u. iiiersetzt, 32, 
192, Leipzig, 1913. 


arts of litV amoiii;" them far beyond the state of de- 
velopment of their domestic institutions." As Stern has 
shown, til is exposes the weakness of the whole scheme. 
"We can logically grade either by chosen "inventions or 
discoveries" or by sociological criteria, but we cannot 
propound a system on one basis and at our convenience 
shift its basis later on. If iron can be borrowed, so can 
other "tests of progress." The proximity of archers 
would account for the shooting of arrows by Northern 
Canadians, who on general grounds might well be de- 
graded to the middle status of Savagery; and contacts 
with potters explain earthenware among the various 
simple peoples. In short, diffusion plays havoc with any 
universal law of sequence. This difficulty, however, Mor- 
gan does not face. 

However, Pitt-Rivers' case has show^n us that deri- 
vation from a single focus and a theory of evolution 
are not mutually exclusive. Actually, in two major prob- 
lems Morgan's general parallelism was superseded by 
its exact antithesis. He regarded the clan as practically 
universal at a given stage, but the principle of reckoning 
certain blood-relatives as kin and excluding others 
seemed so abstruse that he postulated a single origin. 
To explain the spread of the concept he fell back on 
natural selection: the rule of marrying outside one's 
clan would be so advantageous as to "propagate itself 
over immense areas through the superior powers of an 
improved stock thus created."* 

On the subject of relationship terminology Morgan 
was still more extreme. A system of designating kindred, 
he argued, cannot be borrowed ; it can merely be spread 
by migration. The Hawaiians and Zulu classified certain 
relatives in the same way, hence Polynesians and Kaffir 
must spring from the same stock. The Tamil of India and 
the Iroquois of New York share the same system, hence : 

* Ancient Society, Part II, Chap. 15. 


''When the discoverers of the New World bestowed upon 
its inhabitants the name of Indians^ under the impres- 
sion that they had reached the Indies, they little sus- 
pected that children of the same original family, al- 
though upon a different continent, stood before them. 
By a singular coincidence error was truth. ' ' ^ 

This was, of course, mystical nonsense that could 
be readily refuted from Morgan's own schedules. Lub- 
bock at once demolished it. Not only the Tamil but also 
the Fijians and Australians resemble the Iroquois in 
their kinship categories. Are the last, then, specifically 
allied in race with all these biological groups! Worse 
still, the several Iroquois systems given by Morgan are 
far from uniform, one of them being closer to the 
Polynesian than to the Seneca type. In other words, some 
Iroquois, on Morgan's contention, would have to be 
classed as racially Polynesian ! ® 

The instance is enlightening because here Morgan 
was dealing with material he knew better than did any 
of his contemporaries. His amazing inferences illustrate 
his deficiency in historical sensitiveness. 

Notwithstanding such lapses, Morgan's influence on 
comparative sociology was not only tremendous but in 
many respects beneficial. We must simply discriminate 
between his services and his blunders, his original 
achievements and what he took from others. Various 
views commonly associated with him were in no sense 
peculiarly his. It was Bachofen who first proclaimed 
the priority of matrilineal descent, and Morgan simply 
joined the chorus of McLennan, Lubbock, and Tylor. 
Similarly, the idea that rude savages could not live in 
individual wedlock was contemporary patter. Again, 
primitive communism was part of the scientific credo of 

s Systems, 508. 

«Lord Avebury, The Origin of Civilisaiion, 6th ed., 179, 1911. 


the period.' But Morgan rendered a real service in giv- 
ing early currency to Maine's distinction between terri- 
torial and kinship units; he helped clarify the concept 
of exotianiy (pai;e 45); and he repelled McLennan 's no- 
tkm of polyandry as a generally significant social phe- 
nomenon. AYhat is more, he was the first and, unfor- 
tunately, the last to summarize in scholarly manner the 
North American data on clan organization, material in 
large measure secured in the field by himself or through 
personal correspondence. 

However, Morgan's unique distinction lies in liter- 
ally creating the study of kinship systems as a branch 
of comparative sociology. His fame, we may confidently 
predict, will ultimately rest on his Systems of Con- 
sanguinity and Affinity (1871). As the most active re- 
viver of interest in the subject has written, *4t was he 
who collected the vast mass of material by which the 
essential characters of the [classificatory] system were 
demonstrated, and it was he who was the first to recog- 
nize the great theoretical importance of his new discov- 
ery."^ Indeed, Morgan did more, for he gathered not 
merely samples of the type Rivers here mentions, but 
described as adequately as he could all the kinship no- 
menclatures of the world, considerably over a hundred 
in all. Any ethnographer who has tried to secure even 
a single complete terminology of relationship from na- 
tives can appreciate what this means. Morgan collected 
a large number of systems by direct inquiry in the field ; 
for others he enlisted the aid of missionaries, traders, 
and consular agents, to whom he gave the requisite in- 
structions. In this part of his work the factual 
deficiencies are wholly excusable. South American infor- 
mation along this line was buried in recondite sources 
only quite recently made available ; African relationship 

7 See Lord Avebury, op. cit., 103, 478. 

8 W. H. R. Elvers, Kinship and Social Organisation, 4 f ., London, 1914. 


terms had never been methodically recorded. Morgan 
spared no pains, and his achievement was colossal. 

A mere compilation of raw data, however, can never 
confer the title of greatness; and of course Morgan 
would never have undertaken the task unless spurred on 
by the hope of deeper insight. Here discrimination is 
again necessary. We have seen that his interpretation of 
the schedules in racial terms was not merely wrong but 
absurd. However, immersing himself for years in the 
welter of fact, he came to sort out his data so as to fur- 
nish a solid basis for important historical conclusions 
even though they were quite different from those he 
primarily had in view. For example, he recognized the 
criteria of what is now known as the "Omaha" system 
and indicated its occurrence among the Algonkian as 
well as the Siouan family. He was equally successful in 
defining the criteria of the "Crow" type and in tracing 
them among at least three distinct linguistic stocks — a 
determination fraught with significance irrespective of 
his failure to make the most of it. Still more suggestive 
was the similarity of American with Old World features, 
though a racial explanation never was so much as de- 
batable. When a long series of relationship classes found 
in New York State turns out to be almost exactly du- 
plicated in Southern India, such coincidence clamors for 
an explanation. By bringing under a common category 
all comparable phenomena the world over, Morgan ad- 
vanced the typological and the distributional aspects of 
the subject; and in exploiting the resemblances for his 
own purposes, he at least drew attention to a very real 

His two basic categories, the Classificatory and the 
Descriptive, have long since proved inadequate, but here 
again the fault was partly due to the unavailability of 
information from certain crucial areas. Broadly speak- 
ing, Morgan regarded most European systems as "de- 


scriptive" and those of primitive peoples as ''classifi- 
catory," i.e., Ihoy uroiiped together many relatives of 
a certain type under a common class term. Typically, 
a primitive man would apply the same word to his father, 
his father's brothers, and at least certain of his father's 
cousins; and the term for **son," "mother," and so on, 
had correspondingly wide extension. We now know that 
in America by itself many tribes unknown to Morgan 
are as definitely nonclassificatory as the English or 
French.'' With the pertinent anomalies among the Eskimo 
Morgan wrestled manfully, though in some bewilder- 
ment, recognizing the differences from his standard 
American types, yet unable, for want of enough com- 
parable material, to see the variants in proper perspec- 
tive. On the other hand, Kroeber has rightly urged that 
Indo-European languages are not lacking in classifica- 
tory terms, such as "cousin." As to "descriptive," Riv- 
ers pointed out that while the Scandinavian farhror for 
paternal uncle describes the relationship, this does not 
hold for the French oncle, English uncle, and so on. 
Such terms merely denote the relationships in question 
without describing them ; and even if Morgan meant his 
rubric for the hypothetical proto-Indo-European ter- 
minologies, this would leave virtually all modern Euro- 
pean nomenclatures without a place in his scheme. 

A still more vital objection may be made. "Descrip- 
tive" relates to a technique for defining relationship, 
"classificatory" to a mode of grouping. A descriptive 
term like farhror might very well be extended to a large 
class of the father's kinsmen. The two basic concepts of 
Morgan's scheme are thus not complementary, but be- 
long to different logical universes. 

In justice to Morgan we must add that even today a 

9 Leslie Spier, "The Distribution of Kinship Systems in North 
America " University of Washington Publications in Anthropology, 1:69- 
88, Seattle, 1925. Konald L. Olson, "The Quinault Indians," tfetd., 6:91, 


satisfactory all-embracing classification of kinship ter- 
minologies has not yet been achieved. It does seem cer- 
tain, however, that a particular terminology is rarely, 
if ever, a unified logical system. As Kroeber has shown, 
it is rather the result of several mutually intersecting 
principles; and it is these that must be determined be- 
fore a complete typology is possible.^" 

Besides the historical explanation in racial terms, 
Morgan advanced a sociological interpretation of far 
greater significance. In an original manner he used his 
systems to prop up the evolutionary series, by no means 
peculiar to him in essence, that led from promiscuity to 
compulsory monogamy. His simplest set of terms came 
from Polynesia; hence he assumed it to be the most 
ancient, a conclusion generally held untenable now since 
in classification simplicity often comes late. He further 
noted that these Oceanians used the same word for a 
father and a mother's brother. This, Morgan argued, 
implied that at one time a man mated with his own sister, 
since on that assumption the children would have no 
reason to distinguish their maternal uncle from their 
father. In other words, the classification of kin survives 
from a period in which the closest blood-kin regularly 

The fatal error here lies in reading an unwarranted 
meaning into the facts reported. For what the schedules 
tell us is not that the uncle is called ' ' begetter, ' ' but that 
the uncles and the begetter are all designated by a single 
term which, strictly speaking, has no European equiva- 
lent. The custom postulated by Morgan would indeed 
logically produce the given terminology; but there are 
other possible explanations. For example, the Polyne- 

10 A. L. Kroeber, " Classifieatory Systems of Relationship," JRAI, 39: 
81, 1909. On the problem of classifying the systems, cf. Bernard W. 
Aginsky, "Kinship Systems and the Forms of Marriage," AAA-M, 45:1- 
102, 1935; also Kingsley Davis and W. Lloyd Warner, "Structural Analy- 
sis of Kinship," A A 39:291 sq., 1937. 


sians simply may liave used one term for all kin of 
equal status ^vith reference to the speaker. 

Nevertheless, a splendid and fruitful idea remains. 
Though the special interpretations of terminological fea- 
tures offered by Morgan are erroneous, his principle that 
they are correlated with social factors of some sort, with 
forms of marriage and rules of descent, is largely true. 
It has already led to important discoveries by Tylor, 
Rivers, Radcliffe-Brown, and others ; it holds out promise 
of the most exact interpretative findings in the range of 
cultural anthropology. 

Morgan's Systems, then, blazed the way for two 
avenues of research: distributional studies of kinship, 
which automatically merge in important historical prob- 
lems; and inquiries into the organic nexus between ter- 
minologies and associated usages. 

A total judgment of Morgan hinges on one's atti- 
tude towards all-embracing systems. Men who are will- 
ing to hail Herbert Spencer as a greater Newton in- 
evitably gaze upon Ancient Society as an unrivaled 
synthesis. Minds that discriminate what is achieved 
from what is attempted must dissent. Recognizing the 
value of the book at the time of its appearance, they 
cannot overlook its strange neglect of vital sociological 
phenomena and the consequent distortion of the picture 
offered. They see many of its views as the product not 
of the author but of his generation. Their abiding faith 
in Morgan's greatness rests on the Systems of Consan- 
guinity and Affinity. Imperfect in form and matter, like 
everything from Morgan's pen, it remains a tower- 
ing monument. The case invites comparison with an- 
other evolutionist. Lubbock's versatile mind ranged over 
the whole field of civilization; he thought much and in- 
dependently, but rarely with concentration. Outside of 
prehistory he has left little mark; though he gives con- 
siderable space to the growth of supernaturalism, for 


example, a historian of comparative religion might 
easily leave him unread. Morgan expressed no enlighten- 
ing ideas on art, language, or religion; but he can never 
be ignored by the student of kinship. His was not a 
flashy intellect, but one of unusual honesty, depth, and 
tenacity; and prolonged groping rewarded his real, if 
drab, intelligence with glimpses of true insight. There 
is no better illustration of Darwin's saying, ''It's dogged 
does it." 



In the heroic period of modern British civilization 
no one represented ethnology more worthily before fel- 
low scientists and the laity than did Edward Burnett 
Tylor (1832-1917). He was the peer and comrade in 
arms of Huxley, Galton, Spencer, and "Wallace; he was 
cited by psychologist and historian, biologist and philoso- 
pher, by everyone interested in primitive thought or 
behavior; and the lapse of time has merely confirmed 
the earlier judgment of his greatness.^ 

Tylor had been privately educated, and after a brief 
business career, in 1856 he visited Mexico in the company 
of Henry Christy, the prehistorian associated with Lartet 
in the exploration of the Perigord caves. This journey 
resulted in Tylor 's first book, Anahuac; or Mexico and 
the Mexicans (London, 1861). Several years later came 
a major work, Researches into the Early History of Man- 
kind and the Development of Civilization (London, 1865), 
which was followed by the two volumes on Primitive 

1 For biographical details, see Andrew Lang, in Anthropological Essays 
Presented to Edward Burnett Tylor in Honor of His Seventy-fifth Birth- 
day, 1912; also A. C. Haddon's obituary notice in Nature, Jan. 11, 1917; p. 
373; and R. R. Marett, Tylor, New York, 1936. 



Culture: Researches into the Development of Mythology, 
Philosophy, Religion, Language, Art, and Custom (Lon- 
don, 1871). A popular textbook on Anthropology: An 
Introduction to the Study of Man and Civilization (1881) 
rounded out the number of Tylor's books, many articles 
being previously or later contributed to the Journal of 
the Anthropological Institute. Though not a university 
graduate, Tylor was associated with Oxford as keeper 
of the University Museum and advanced from a reader- 
ship in 1884 to professorial status in 1896. By then he 
had long enjoyed an international reputation, his princi- 
pal books having been translated into several languages. 

Tylor was not technically a field worker, yet he was 
the very opposite of an armchair anthropologist. That 
he saw Mexican natives in his early manhood and later 
(1884) paid a brief visit to Pueblo villages counts for 
something, but more important is his unremitting tend- 
ency to study culture in the very heart of a metropolis. 
He receives a Tasmanian skin-scraper and forthwith 
has it tested by his butcher ; he peers into shop windows 
for a parallel of the Oceanian pump-drill; in Somerset- 
shire he watches a weaver throw her shuttle from hand 
to hand; and discerning a problem in aboriginal gesture 
languages, he learns hundreds of signs in the Berlin Deaf 
and Dumb Institution. 

This inclusiveness of his interests is one of the con- 
spicuous facts about Tylor. Most of his compatriots, both 
contemporaries and successors, limited their inquiries to 
particular topics. Lang, Frazer, and Marett cultivated 
religion and folklore; Rivers, social organization; Pitt- 
Rivers, Balfour, and Haddon worked on technology and 
art. Tylor embraced the whole field, even though he could 
not contribute equally to all its divisions. He kept abreast 
of prehistory, in which his interest had been stimulated 
by Christy; and he steadfastly concerned himself with 
linguistics, as the references to Steinthal and Lazarus 


indicate. It is merely necessary to compare Tylor's re- 
marks on lanprnage willi the pitiable treatment of the 
subject in Rivers' The Hist or u of Melanesian Society in 
order to appreciate the greater seriousness of Tylor's 
approach to the problems of civilization. 

Equally noteworthy is Tylor's uncanny sense of fit- 
ness in dealing with sources and extricating from them 
a solid core of knowledge. When he began to write, much 
of the globe was unknown ethnographically. Otherwise 
excellent reports skimmed over matters now — largely 
thanks to Tylor — methodically investigated by the veri- 
est novice. Able travelers mingled fancy with observa- 
tion, indulged in the superficial psychologizing that 
duped Klemm, and otherwise twisted the facts from 
initial bias. 

Here, again, comparison is illuminating. Lubbock 
was not a man of mean caliber. He was scientifically 
trained; he made original contributions to prehistory, 
including the distinction between the Paleolithic and 
Neolithic periods; in breadth of interests he rivaled 
Tylor ; and his critical acuity is attested by his strictures 
on Morgan. Yet, in approaching savage man, he lacked 
two requisites, detachment and a sense of probability. 
Himself forever emotionally affected by exotic custom, 
he timidly apologizes for describing what may be "very 
repugnant to our feelings," fearful "lest I should be 
supposed to approve that which I do not expressly 
condemn."^ And his eagerness to fill in the gaps in an 
evolutionary system constantly deflects his judgment. On 
both counts Tylor's superiority is clear. It is interesting 
to see Tylor spurning the very evidence which Lubbock 
readily swallows. Lubbock accepts the word of "sailors, 
traders, and philosophers, Roman Catholic priests and 
Protestant missionaries" that there are races of men 
"altogether devoid of religion"; and among these he 

2 Avebury, The Origin of Civilization, xviii, 212. 


includes some Eskimo and some Canadian Indian tribes, 
Brazilian aborigines, Andamanese, Hottentots, East 
Africans! Tylor in masterly fashion exposes these fan- 
tastic stories, revealing their psychological causes and 
concluding that *'the asserted existence of the non-reli- 
gious tribes in question rests ... on evidence often 
mistaken and never conclusive. ' ' ^ Again, Lubbock does 
not, indeed, hold that savages use gestures because they 
lack words, but he quotes the famous tale about the 
Arapaho Indians unable to converse in the dark as though 
he half believed it. Tylor explicitly rejects it: ''Captain 
Burton only paid a flying visit to the Western Indians, 
and his interpreters could hardly have given him scien- 
tific information on such a subject."* 

It is impossible to exaggerate Tylor 's services in 
separating the dross from the gold of early chronicles 
and thus rescuing a substantial body of authentic fact 
on every phase and period of civilization. 

Tylor, however, went far beyond a mere assemblage 
of trustworthy material; he derived new concepts from 
these data and formulated problems which bear both on 
the history of civilization and on the interrelation of its 
traits. To quote Clark Wissler, it was he who ''blocked 
out the essential processes of primitive fire-making"; 
Dr. Walter Hough has indeed added much detail, but he 
adhered to Tylor 's categories, which marked a definite 
advance on Klemm's. We have already pointed out the 
fruitful definition of "stone-boiling" as a stage in culi- 
nary development. This, too, was not a mere matter of 
coining new words, but of creating a new concept by 
segregating one set of data from the rest. Similarly, all 
subsequent discussion gained definiteness by Tylor 's 
minimum definition of religion as animism, ' ' the belief in 

3 Avebury, ibid., 219 f., Tylor, Primitive Culture, 1:425. 
* Avebury, op. cii., 432-434 ; Tylor, Eesearches, 78 f. 


Spiritual Beings. '"* Again, facts of social organization 
hitherto iin-co-ordinated or inadequately classed were 
ticketed ''local exogamy," " teknonymy, " ''cross-cousin 
marriage. ' ' 

Tylor's epigoni have been reproached with amassing 
facts under a common head witliout special regard either 
to their provenience or to local differences. The substance 
of their material might thus properly be condensed into 
tabular form, indicating that such beliefs or customs 
were found here, and others elsewhere. Tylor's own 
treatment, however, is often explicitly geographical, as 
when he traces the distribution of the Stone Age from 
continent to continent or passes from North American 
to Kamchadal stone-boiling.® Such a survey logically 
merges in some historical interpretation, for resem- 
blances are naturally interpreted either as evidence of 
contact or of some inherent law. Tylor wrestled mth this 
problem throughout his career; indeed, his Researches 
revolves largely about this question ; and he applied both 
principles of explanation — historical connection and 
Bastian's concept of psychic unity. 

There is a strange legend about the development of 
ethnological theory. It represents ethnology as sunk in 
the slough of Bastian's elementary ideas until rescued 
in 1887 by the geographer Eatzel (page 119). He, it 
seems, denounced the independent evolution of culture 
as equivalent to spontaneous generation in biology and 
for the first time demonstrated the complexity of culture 
due to the migration of its elements.^ A still more pictur- 
esque account is offered by the eminent anatomist G. 
Elliot Smith (page 160). After vehemently ridiculing 
biological analogies in culture, he ecstatically hails 

^Primitive Culture, 1:424. 
^Eesearches, 203 sq., 263 f. 

^ W. Schmidt and W. Koppers, Volker und Kulturen, 32 f ., Eegens- 
burg. 1924. 


Ratzel's attack on ''spontaneous generation"* and as- 
signs to the Leipzig geographer much the same part as 
do Fathers Schmidt and Koppers. What is, however, still 
more interesting, he portrays Tylor as a double person- 
ality, now guided by the benign patron saint of Diffusion, 
now tempted by the sinister demon of Bastianism until 
he finally succumbs to the lure of the Evil One.^ A few 
lines in an extremely brief Encyclopaedia article written 
in 1910 towards the close of Tylor 's life and probably 
never seen by most of Tylor 's admirers are taken to blot 
out his influence on the side of historical connection, and 
this despite the admission that throughout his life he had 
steadfastly championed diffusion. Absurdity can go no 

The actual facts bear no resemblance to this clap- 
trap. Bastian himself never denied the dissemination of 
culture, he merely demanded specific proof of it. Had 
Professor Elliot Smith been more conversant with an- 
thropological literature, he might have recalled the two 
collections of essays issued in celebration of Bastian 's 
seventieth birthday (1896). One contains an essay by 
Tylor, the other by Boas, two men intimately associated 
with Bastian. Both articles avow a faith in diffusion — 
Tylor 's in the most uncompromising manner. Did these 
two friends of Bastian deliberately attempt to insult the 
man they were ostensibly honoring by flaunting views in 
mockery of his own? 

In a widely read work, Richard Andree's Ethnogra- 
phische Parallelen und Vergleiche (1878), Bastian 's 
position is clearly set off in the abstract and illustrated 
by concrete examples. Andree is quite willing to entertain 
the view that Sudanese, Somali, and Bantu derive the 

* Incidentally he errs in asserting that the comparison originated with 
Eatzel. It occurs in the good old parallelist Eichard Andree's Ethnogra- 
phische Parallelen und Vergleiche, p. iv, Stuttgart, 1878. 

8 G. Elliot Smith, The Diffusion of Culture, 66, 116-183, London, 1933. 

10 Cf. R. E. Marett, Psychology and Folk-Lore, 81, 1920. 


rule of mother-in-law avoidance from a single source ; it 
is the American and the Australian occurrence that he 
feels compelled to treat as separate instances. A rejection 
of ditfusion on principle is out of the question. Indeed, 
the same author subsequently traced to a single center 
of origin all cases of shoulder-blade divination, ranging 
from Great Britain and Morocco to Bering Strait.^^ Un- 
doubtedly there were writers who took Bastian's elemen- 
tary ideas very seriously, but it is safe to say that from 
1860 to 1887 there never was a time when any responsible 
writer denied contact as a factor in culture history. 

But whatever may have been the situation in Ger- 
many, ethnologists who could read English required no 
Eatzel to teach them about historical connection. As early 
as 1864, Pitt-Rivers — an intimate associate of Tylor — 
held strong views on that subject (page 28). As Profes- 
sor Elliot Smith is compelled to admit, Tylor himself 
was constantly producing proof of contact. He was, in 
fact, the very antithesis of a strict parallelist, even if 
he viewed the facts with scientific poise instead of falling 
prey to a cheap diffusionist dogmatism. In short, he was 
thoroughly convinced of the force of borrowing in human 
history and expressed this faith both abstractly and with 
respect to special cases. ''Civilisation," we read, ''is a 
plant much more often propagated than developed." 
Again, "Most of its phenomena have grown into shape 
out of such a complication of events, that the laborious 
piecing together of their previous history is the only 
safe way of studying them. It is easy to see how far a 
theologian or lawyer would go wrong who should throw 
history aside and attempt to explain, on abstract prin- 
ciples, the existence of the Protestant Church or the Code 
Napoleon." And in the introduction to a translation of 
Ratzel's History of Mankind he contrasts "the small 

11 Richard Andree, " Scapulimantia, " in Boas Anniversary Volume, 
143-165, 1906. 


part of art and custom which any people may have in- 
vented or adapted for themselves" with "the large part 
which has been acquired by adopting from foreigners 
whatever was seen to suit their own circumstances. ' ' " 

Let us now consider some of Tylor's specific ideas 
on the subject. First we note his introducing diffusion 
into a discussion of prehistoric technology. After point- 
ing out the amazing similarity of stone tools in different 
parts of the world, he argues that with all due allowance 
for psychic unity "it is very doubtful whether it can be 
stretched far enough to account for even the greater 
proportion of the facts in question." While not ruling 
out independent origin, he concludes that the observed 
uniformity "may some day be successfully brought in 
with other lines of argument . . . which tend to central- 
ize the early history of races of very unlike appearance, 
and living in widely distant ages and countries. ' ' ^^ With 
regard to various inventions, Tylor urges the same point. 
He derives the piston bellows of Madagascar from In- 
donesia ; argues from the distribution of North American 
pottery to a single source; and at least favors a single 
world focus for the bow and arrow." Of nonmaterial 
elements, the Australian, African, and American theory 
of disease as due to an extraneous object which the 
physician must suck out is treated as having a common 
origin; and the same explanation is offered for various 
mythological parallels between America and the Old 

What distinguishes Tylor from the extreme dif- 
fusionists is simply his serene willingness to weigh 
evidence. Refusing to assume a priori that all resem- 
blances result from dispersal, he applies definite criteria 
for settling the question. They have never been improved 

12 Tylor, Primitive Culture, 1:53, Researches, 4. 

13 Researches, 203. 

^^Ihid., 167-169, 366; Primitive Culture, 1:64. 
^^ Researches, 277, 335, 360. 


upon. Ill the case of American pottery he applies the 
principle of continiums distribution: earthenware vessels 
occur not sporadically "as if a tribe here and a tribe 
there had wanted it and invented it" but "in a compact 
field" from Mexico northwards. In comparing the Tor- 
toise myth of India with its New World counterparts, he 
cites the combination of three specific features as evi- 
dence of connection." He thus forestalls by decades 
Graebner's "qualitative" and "quantitative" criteria 
(page 158). Quite formally, in the discussion of American 
lot-games, he marshals the reasons for a common center. 
The Hindu game of pachisi is connected with the Aztec 
patolli because both share not a single trait but a series 
of independent features — divining by lot, a sportive 
wager, an appreciation of the law of chances, transfer of 
the result to a counting board, and rules of moving and 
taking. Tylor concurs in the opinion that "highly special 
or complex phenomena" are less likely to be duplicated 
than "the obvious and simple" and thus concludes that 
the games are related, i.e., he infers communication 
across the Pacific from Eastern Asia.^'^ 

This statement implies conditions in which resem- 
blance is not held sufficient evidence of connection. Else- 
where Tylor is more explicit: whenever we see "like 
grounds ' ' from which the similarities could have grown, 
it is permissible to suggest independent development.^® 
So far Tylor is wholly right. A priori a resemblance 
may just as well be due to the same antecedent as to 
historical contact; and the logical value of the two ex- 
planations is identical. But Tylor does at times lapse into 
the vagueness of Bastian when he speaks of similarity 
as assignable to "the like working of men's minds under 

18 Ibid., 335. 

"E. B. Tylor, "On American Lot-Games, as Evidence of Asiatic 
Intercourse before the Time of Columbus," in Ethnographische Beitrdge, 
Supplement zu lAE 9:55-67, 1896. 

^^ Researches, 296. 


like conditions," of ''general laws" that explain phe- 
nomena as "direct products of the human mind." ^® We 
must, of course, recall that Tyler's prime preceded the 
epistemological purging of natural science that is as- 
sociated with Poincare and Mach. The laws of physics 
were still "eternal and inexorable," and any new branch 
of knowledge aspiring to recognition simulated the ritual- 
ism of its elders. At times, to be sure, Tyler's reference 
to "laws" in civilization seems to imply no more than a 
recognition that its phenomena are causally determined ; 
but he obviously has more in mind when he compares 
these principles with the law of magnetic attraction.^" 

Now, this position requires much modification in 
order to retain any validity. The attraction of iron filings 
by a magnet is a predictable phenomenon, but the sav- 
age tendency to explain fossil remains by myths of giants 
is not similarly predictable.^^ We are merely wise after 
the fact; Tyler's caption "Myths of Observation" marks 
a useful descriptive category, but we can neither be sure 
that a particular people will have developed such tales 
nor what may be the specific plot if they have. Also, 
granting the specific unity of the human mind, when can 
we be sure of "like conditions"? Unless "the uniform 
action of uniform causes" can actually be traced, paral- 
lelism of culture traits is an empty allegation. 

That is why, as a rule, historical connection accounts 
so much more satisfactorily for resemblances than the 
rival hypothesis. Whether probable or not, former con- 
tact does explain how remote peoples come to share 
customs and beliefs. The champion of parallelism scores 
only if he can demonstrate the same specific determinants 
in both areas, and this he usually fails even to attempt. 
For general "psychic unity" will not do : on that assump- 

19 Ibid., 3, 5, 325. 

^^ Primitive Culture, 1:1-4, Besearches, 3. 

21 Besearches, 3, 299. 


tion all the societies of the world should share the 
features in question; at least, the parallelist must point 
out the particular circumstances tliat militate against the 
invariable reproduction of the same result by a common 

Let us illustrate by reverting to Andree. This typical 
parallelist cites many instances of African, American, 
Australian, and Asiatic parent-in-law avoidance, treating 
them as so many spontaneous effects of the same mental 
disposition. There is, however, no attempt to account for 
the absence of the phenomenon elsewhere; and insofar 
as the author extricates from his sources any cause for 
the feature compared, it is not uniform. The Kaffir are 
said to refrain from intercourse with a mother-in-law for 
fear of committing incest even in thought; but in the 
Argentine the rule springs from the practice of sacrific- 
ing old women to a deity, and this offering is facilitated 
if son-in-law and mother-in-law remain strangers.^^ Here, 
then, the logical treatment of the problem is not at all 
satisfactory from the point of view of parallelism. Not 
only does it remain unexplained why psychic unity is so 
capricious in creating the rule in some places while not 
in others ; but the specific determinants are not the same, 
so that actually the same mentality produces like results 
under unlike conditions ! Granting the reality of such 
convergence, the logic remains deficient within the paral- 
lelist scheme. 

It is here once more that Tylor's superiority ap- 
pears. While sometimes he employs the vague phrase- 
ology of his period, he repeatedly brings forward specific 
determinants of specific effects. Why, for example, does 
he not use diffusion to explain the general belief in a 
hereafter? Because of the general occurrence of dreams 
apparently showing the continued existence of deceased 

22 E. Andree, op. oit., 159-164, 1878. 


kin.^^ We may doubt the facts — though in my judgment 
Tylor correctly represents them — but the logic of tracing 
a ubiquitous effect to a ubiquitous antecedent is unas- 

Tylor, however, deserves still greater credit for his 
treatment of the more difficult instances of restricted 
distribution that baffle the parallelist. Again we can illus- 
trate by the parent-in-law observance. In his earlier 
discussion Tylor inclines to a single focus, for "it is hard 
to suppose that the curiously similar restrictions . . . 
can be of independent growth in each of the remote dis- 
tricts where they prevail."^* This we have recognized 
as a logically valid interpretation. When Tylor subse- 
quently produced a substitute theory, he did not fall back 
upon the inadmissibly vague phrase "psychic unity," but 
requisitioned a definite sociological determinant, matri- 
local residence, to explain the mother-in-law taboo ; while 
by patrilocal residence he explained the reverse rule for 
a woman and her father-in-law.^^ In the same paper he 
similarly treats as causally linked other customs, such as 
exogamy and one of Morgan's Classificatory types of 
nomenclature. Moreover, he tried to support his conclu- 
sions by the theory of probabilities, comparing the actual 
associations with those to be expected on chance. 

These investigations, merely summarized in the 
paper cited, were unfortunately never presented in full 
and seem to be irretrievable, so that we cannot satis- 
factorily check the conclusions by the evidence on which 
they rest. The statistical technique has been proved 
inadequate in several respects. To concentrate on a single 
point, it is difficult to define the basic group to be selected 
as the unit in such an inquiry. Tylor does not seem to 

^Researches, 5-8. Primitive Culture, 1:450. 

2* Researches, 296. 

25 E. B. Tylor, " On a Method of Investigating the Development of 
Institutions; Applied to Laws of Marriage and Descent," JRAI, 18:245- 
269, 1889. 


distinguish between cases reducible to a single place of 
origin and otliers where several or many foci are proba- 
ble. In its published form the paper fails to strike the 
balance, then, between diffusion and independent origin. 
Apparently Tylor here inclines to the latter position, for 
he writes: ''The institutions of men are as distinctly 
stratified as the earth on which he lives. They succeed 
each other in series substantially uniform over the globe, 
independent of what seem the comparatively superficial 
differences of race and language, but shaped by similar 
human nature acting through successively changed con- 
ditions in savage, barbaric, and civilized life." If trans- 
mission is as potent as Tylor elsewhere recognizes it, 
would it not seriously disturb the uniform action of any 
law of sequence, or at least render impossible its demon- 
stration? Thus, if twelve patrilocal Siberian tribes re- 
strict speech between a woman and her husband's father 
in complete independence of one another, the case for an 
organic nexus between the taboo and the residence rule 
is much stronger than if all of these peoples borrowed 
the custom from a single source. On this latter assump- 
tion the accidental contact of two groups would seem far 
more influential than any immanent law of development. 
In any case, how could a law be inferred from a single 
occurrence ? 

Tylor 's achievement, however, remains unaffected 
by such statistical foibles, for it rests on a number of 
absolutely sound logical principles. In the first place, he 
eliminates vague psychologizing in favor of specific social 
factors. The objection to Andree's type of explanation 
is thus overcome ; human nature produces such and such 
a restriction on social intercourse, hut only in specified 
circumstances. Further, Tylor recognizes the complexity 
of cultural phenomena when, in his explanation of tek- 
nonymy, he accepts more than one determinant of the 
same effect. He thus substitutes for the antiquated 


metaphysical concept of cause the mathematical concept 
of function : teknonymy is no longer the inevitable effect 
of matrilocal residence or of an avoidance rule; it is 
simply more probable with these concomitants than with- 
out them. The notion of ''law" as thus purified is thus 
nothing more than what it has become in the definition 
of philosophical physicists — a limitation of our expecta- 
tion in the light of experience. 

The essential thing is this: ethnologists have con- 
stantly, consciously or unconsciously, affirmed causal 
connections ; but they have rarely stooped to justify their 
assertions. Tylor's scientific conscience prompted him to 
offer proof; and while from the published fragment of 
his schedules the validity of his demonstration cannot be 
established, at least one major conclusion seems to be 
borne out by later research : some correlation, though not 
so high as he thought, between exogamy and the Iroquois 
type of kinship terminology. Apart from specific results, 
we must insist that Tylor remains one of the few scholars 
whose championship of independent evolution is not a 
sterile, however warranted, denial of diffusion. For our 
sense of causality is satisfied only when the conditions 
of the problem are met in his spirit — by the demonstra- 
tion of specific determinants that tend to produce like 
results in historically independent centers. 

One point of Tylor 's logic, however, must be exposed 
as vulnerable. Any statistical treatment can prove only 
correlation, not a time sequence; it yields propositions 
of the order that the side of a triangle varies with the 
opposite angle, but can never establish the primacy of 
either angle or side. But in his eagerness to prove 
maternal societies earlier than paternal, Tylor ignores 
this limitation. He begins by defining three social sys- 
tems. In the maternal type matrilineal descent is coupled 
with avuncular authority and nepotic inheritance ; at the 
opposite pole the paternal society has patrilineal descent. 


paternal authority, filial succession; in the intermediate 
condition the features of tlie two extremes are variously 
blended. Tylor inquires how certain selected usages, such 
as the couvade, are correlated with these three types. 
The couvade is absent from the maternal ''stage," occurs 
twenty times in the intervening, and only eight times in 
the pure paternal system. This, he argues, shows the 
priority of the maternal stage, for had it come later it — 
and not the paternal — would show survivals of the 

But observation does not reveal "stages," only 
certain combinations of traits. The facts show merely 
negative correlation of the couvade wdth features a, h, c, 
of the matrilineal system, and positive correlation with 
e, f, of the patrilineal and the intermediate system. A 
sequence is smuggled in only by assuming from the start 
that the classificatory differences have chronological 
meaning, viz., by assuming what is to be proved. 

What doubtless misled Tylor here was the contem- 
porary preconception that what differs from modern 
civilization is ipso facto inferior and earlier. Swanton 
has shown (page 145) that in North America the rudest 
peoples are either patrilineal or without fixed rule of 
descent, while many higher tribes, like the Hopi, are 
strictly matrilineal; and comparable evidence has ac- 
crued from other continents. The priority of ''mother- 
right" as a general principle has thus been pretty 
generally abandoned. 

The same bias produced a corresponding error in 
Tyler's Primitive Culture, when he denied high gods to 
the simpler peoples. In other w^ords, he assumed that 
such conceptions could arise in civilization only as the 
result of slow evolution out of primitive soul beliefs. On 
this subject Tylor was challenged by Andrew Lang, a 
versatile man of letters who had shown far more than a 
mere dilettante's interest in comparative religion and 


folklore.^® Lang insisted that deities did not necessarily 
improve in dignity with advancing civilization and that 
extremely rude peoples, including some Australians, 
shared with Christianity the conception of a primeval 
and morally pure Creator. This view has been vigorously 
championed and elaborated by Father Schmidts The 
truth certainly seems to be that sundry unequivocally 
simple tribes, — certain Negritoes, Californians, and Fue- 
gians — have the conception of a Supreme Being free 
from the undignified traits characteristic of many myth- 
ological figures; and in most of these cases the notion 
is clearly not inspired by Christian missionaries. 

Here, as in the treatment of rules of descent, Tylor's 
mistake sprang from the difficulty of applying the evolu- 
tionary scale to elements of nonmaterial culture associ- 
ated with values. As pointed out above, he kept himself 
on the whole remarkably free from this type of error; 
yet, like other evolutionists, he sometimes felt he had a 
key to the law of progress, confidently speaking of an- 
thropology as ''essentially a reformer's science," as 
"active at once in aiding progress and in removing 
hindrance." But such subjectivism is not at all obtrusive 
in his writings. 

Like so many of his distinguished compatriots, 
Tylor was averse to systematization. His most ambitious 
scheme, that of animism, has nothing like the illusively 
clear-cut classification of Morgan's Ancient Society. As 
Issaurat said in reviewing the French translation of 
Primitive Culture: "What one notes above all is the 
abundance of documents. One finds them by piles, by 
heaps, by mountains, and when these are cleared there 
are still others. ' ' "^ Swamped with facts, careless readers 

28 A. Lang, Magic and Beligion, 15-45, 224 sq., London, 1901; The 
Making of Beligion, 160-190, 3d ed., London, 1909. 

2^^ Wm. Schmidt, Per Ursprung der Gottesidee, 6 vols., Munster in 
Westfalcn, 1926-1935. 

28 Issaurat, in Eev., 6 : 133 sq., 1877. 


wiio must have bold outlines for complex phenomena 
inject into the treatment generalizations that do not exist. 
Thus, Elliot Smith imputes to Tylor the view that "all 
people instinctively regard the universe as alive, and 
regard the objects of it — the mountains, the trees, the 
rivers, objects of wood and stone, — as animate beings 
possessing souls which make the whole world akin."^ 
Undoubtedly that is how Professor Elliot Smith would 
have simplified matters if he had constructed a theory of 
animism. Tylor 's intellect, however, was of a different 
category. He did not confuse the belief in spirits with 
the universal animation of nature. He definitely ascribes 
the latter view only to Algonkians, Fijians, and Karens, 
merely contending that ''many other peoples, though 
they may never have stated the theory of object-souls 
in the same explicit way . . . . , have recognized it with 
more or less distinctness." ^° Even in his survey of re- 
ligion, then, Tylor 's parallelism is limited by his knowl- 
edge of tribal variation, precisely as his general belief 
in progress does not preclude explicit statements that 
advancement fails to be uniform in all branches and that 
degeneration is a reality, even though overshadowed by 

Tylor, we noted, did not escape the imprint of his 
time. He lapses into comparisons of the savage with the 
child of civilized countries, yet he remains free from the 
extravagances of Letourneau and Lubbock with their 
constant blurring of the line that divides man from other 
species. His position in practice approaches Waitz's, for 
he recognizes that, whether real or not, racial differences 
are negligible in the study of civilization.^^ If his psy- 
chology erred, it was in another direction; he did slight 
the emotional in favor of rational factors, and here 

29 G. Elliot Smith, The Diffusion of Culture, 172, London, 1933. 
^'^ Primitive Culture, 1:476-484. 
si/fttd., 7. 


correction has been imperative. With Waitz, Tylor 
further shared a critical sense in evaluating testimony. 
The student of Tylor in 1890 could thus profit from a 
vast mass of thoroughly sifted and authenticated mate- 
rial, interpreted from a unifying, evolutionary point of 
view, tempered with sanity. Problems were set forth in 
definite terms, as was the logic that must underlie their 
solution. There were no capricious escapades of fancy; 
a clear intelligence was seeking order in a vast and 
largely virgin field. The words which Virchow pro- 
nounced on the death of the physiologist Johannes Miiller 
might well be applied to Tylor : what evokes admiration 
is ''the methodical rigor of the investigator, his temper- 
ate judgment, his secure serenity, the ample perfection 
of his knowledge." ^^ 

88 * ' In dem Physiologen Miiller bewunderte man nicht so sehr das Genie 
des Entdeckers, nicht so sehr den bahnbrechenden Flug des Sobers, sondern 
vielmehr die methodische Strenge des Forsehers, das maassvoUe Urteil, die 
sichere Euhe, die reiche Vollendung des Wissens. " Eudolf Virchow, 
Johannes Miiller; eine Geddchtnisrede, 15, 1858. 



Prehistory, Technology, Field "Work 

Notwithstanding the colossal achievement of Tylor 
in his two main works, a great deal remained to be done. 
First of all, the inadequacy of factual knowledge became 
patent. For example, the universality of a Stone Age 
throughout the world had been made probable on general 
grounds, but science required positive proof, and for 
Africa as well as Southern Asia the evidence was avow- 
edly meager. Apart from this broad problem there were 
innumerable special questions as to the precise sequence 
of cultures in every region of the globe. Thus there 
developed a feverish activity among the prehistorians of 
Europe and America, of which the learned journals of 
the period give ample evidence. In part the findings were 
only of local antiquarian interest, but in part they were 
basic. Stone implements continued to be reported from 
Swaziland and the East Horn, establishing with increas- 
ing cogency a premetallic age in all of Africa. On the 
other hand, a critical scrutiny of American data led to 
the conclusion that man was not nearly so ancient in the 



New World as in the Old. With respect to later periods, 
Scandinavian scholars, snch as Oscar Montelius and 
Sophus Miiller, traced the relations of Northern Europe 
with the South and even the Near Orient/ 

Technology developed both from archaeological and 
ethnographic investigations. A prehistorian, in order to 
discuss intelligently the origin of bronze work in a given 
region, inevitably went into refinements far beyond a 
layman's competence. He had to ascertain where tin 
could have been procured at all and what percentages of 
it entered the alloy; also the impurities of particular 
bronzes demonstrated a specific relationship between 
the countries that shared them. For all such determina- 
tions the anthropologist was forced to requisition the 
aid of chemists and other specialists. Corresponding de- 
tail was found necessary for other branches of early or 
primitive industry, as a sound basis both for connecting 
peoples with one another and for appraising their skill. 
It was clearly not enough to say that such a tribe had 
arrows, and even their use of compound bows only 
vaguely placed them as archers. A detailed structural 
study was required for establishing the essential facts.^ 
The United States, where Indians could still be seen 
fashioning arrow points and tanning hides, formed an 
unusually favorable area for testing archaeological in- 
terpretation by the practice of living tribes. Here, ac- 
cordingly, technological studies flourished, especially 
with the development of museum collections. Frequently 
they were combined with experimentation. Walter 
Hough, not content to enlarge on Tyler's assemblage of 
data on fire-making, determined by actual trial how 
quickly a native might produce fire with a simple drill. 
A mechanically gifted inquirer like F. H. Gushing would 

1 E.g., Oscar Montelius, * ' On the Earliest Communications between 
Italy and Scandinavia," JRAI, 3:89 sq., 1900. 

2 Henry Balfour, "On the Structure and Affinities of the Composite 
Bow," JRAI 19:220-246, 1890. 


himself manufacture arrow points in order to gain an 
insight into the inwardness of the stone-knapper's tech- 

The accumulation of facts cannot be dissociated 
from the progress of theory. It was only by new material 
that the generalizations of Tylor and others could be 
contirmed, and in turn it prompted novel interpreta- 
tions. Tylor had declared the gesture-language to be 
'* essentially one and the same in all times and all coun- 
tries," but only a much wider range of observations, 
such as Garrick Mallery began to adduce in the early 
Reports of the Bureau of American Ethnology, could 
settle the point. Morgan had suggested a stage of group- 
marriage, which Fison and Howitt thought they could 
still observe in full swing among Australian blackfel- 
lows.* Miss M. H. Kingsley's observations in West Af- 
rica were guided by her reading of Tylor; on the other 
hand, R. H. Codrington's findings in Melanesia turned 
out to give a new direction to speculations on the origin 
of religion.^ 

Gradually there rose the demand for regional stud- 
ies, undertaken not incidentally to a naturalist's or mis- 
sionary's main interests, but as complete investigations 
of particular peoples by professional anthropologists. 
In 1884, the British Association for the Advancement of 
Science appointed a committee, of which Tylor was a 
prominent member, for investigating the Northwestern 
tribes of Canada ; and from 1888 until 1898, Franz Boas 
was connected with the relevant reports. These investi- 
gations doubtless stimulated the Jesup North Pacific 
Expedition (1897-1902), organized by Boas for deter- 

3W. Hough, "Aboriginal Fire-Makiiig, " AA 3:359 sq., 1890. F. H. 
Gushing, "The Arrow," AA 8:307 sq., 1895. 

*See e.g., A. W. Howitt, "The Dieri and other Kindred Tribes of 
Central Australia," JRAI 20:30 sq., 1891. 

!> R. H. Codrington, The Melanesians, Studies in their Anthropology 
and Folk-Lore, Oxford, 1891. 


mining Siberian-American connections. Comparable in 
intensiveness and roughly contemporary was the Cam- 
bridge Expedition to Torres Straits, led by Dr. A. C. 
Haddon, assisted, among others, by Dr. W. H. R. Rivers 
and C. G. Seligman, each collaborator devoting himself 
to a special topic. Here may also be cited the description 
of the Arunta in Baldwin Spencer and F. J. Gillen 's The 
Native Tribes of Central Australia (London, 1899), 
which precipitated infinite discussion, mainly because of 
the aberrant type of totemism discovered by the au- 
thors. Equally stimulating were Karl von den Steinen's 
two expeditions to the upper Xingu in 1884 and 1887, 
which culminated in the discovery of unsuspected Ara- 
wak and Carib tribes in the interior of Brazil. This 
gifted observer clarified many phases of the daily life 
of the natives, though his residence was too brief for 
intensive investigation. Nevertheless his brilliant recon- 
naissance, set forth in a dashing style saturated with 
humor, exerted a deep influence and illustrates the pos- 
sibilities as well as the limitations of pioneer work. 

Such detailed studies naturally fostered regional 
comparison and delimitation, which implied a rigorous 
typology.^ Field work also brought home the inadequacy 
of the techniques hitherto employed. Thus, it proved far 
from easy to secure a kinship nomenclature by direct 
questioning; and Rivers' experiences in the Torres 
Straits led to a method employed ever since — that of 
first ascertaining a subject's genealogy and then asking 
how he addressed such and such an individual whose re- 
lationship to him had been independently determined by 
the pedigree.^ In the United States linguistic studies led 
Gatschet and others to transcribe tales phonetically in 
the aboriginal tongue ; and it soon became clear that the 

8 E.g., A. C. Haddon, "A Classification of the Stone Clubs of British 
New Guinea," JRAI 3:221-250, 1900. 

'' W. H. R. Rivers, * ' A Genealogical Method of Collecting Social and 
Vital Statistics," JRAI 3:74 sq., 1000. 


philological approach was the only safe method for re- 
cording native praj^ers, literary forms, and kinship 


Theoretical interpretation was by no means lacking 
in this period. And once more we must repudiate the 
hoax that in this era diffusion was taboo. In France 
Hamy was ready to trace the Copan monuments of Hon- 
duras to China; in Germany Schultz-Sellack favored a 
connection of Toltec ideas with China and Japan.^ 
Nephrite objects led to prolonged debates as to the pos- 
sible provenience of the raw material.^ The great Scandi- 
navian archaeologists held strong views as to Near 
Oriental influences on the w^iole of Europe. In England, 
the supposed hotbed of evolutionism. Miss Buckland 
from 1878 until the end of the century presented paper 
after paper in defense of as radical a diffusionist doc- 
trine as has ever been broached. According to her, civili- 
zation was never independently acquired; foreshadowing 
Elliot Smith, she believed that sun- and serpent-worship- 
ers had spread agriculture, weaving, pottery, and metals 
over the earth; she adduced ceremonial haircutting and 
sweating among the Navaho as evidence of intercourse 
with Japan." Whatever we may think of her evidence, 
she was certainly not lynched by her audience. 

Why, indeed, should she be? Evolution, as we have 
seen, lay down amicably beside Diffusion in the An- 

8 E. T. Hamy, "An Interpretation of one of the Copan Monuments," 
JEAI 16:242-247, 1887. Carl Schultze-Sellack, "Die amerikanischen 
Gotter der vier Weltrichtungen und ilire Tempel in Palenque, " ZE 11:209 
sq., 1879. 

9 E.g., Ver 15:211, 478, 1883. 

10 A. W. Buckland, "Primitive Agriculture," JEAI 7:2-18, 1878; 
"Prehistoric Intercourse between East and West," ibid., 14:222-232, 1885; 
"Points of Contact between Old World Myths and Customs and the 
Navaho Myth entitled 'The Mountain Chant,' " ibid., 22:346-355, 1893; 
"Four as a Sacred Number," ibid., 25-96 sq., 1896. 


thropological Institute. Pitt-Rivers, during this period, 
reaffirmed with emphasis his faith in a connection of 
Australian and Egyptian boomerangs. Yule brought out 
cumulative evidence — head-hunting, aversion to milk, 
bachelor dormitories, pile dwellings, piston bellows, 
water ordeals — on behalf of intercourse between Indo- 
China and Indonesia. In a paper read in 1878 and pub- 
lished the following year, Tylor himself suggested the 
Asiatic origin of a Mexican game. He followed this up 
with evidence for the origin of cat's cradle in South- 
eastern Asia, and traced Polynesian kites to the same 
region. Using the identical argument of modern ditfu- 
sionists, — the uninventiveness of rude peoples — he even 
pleaded for a Scandinavian origin of Eskimo clothing 
and such games as cup-and-ball. In America, to be sure, 
Brinton inclined to an intransigent parallelism, but his 
views were being vigorously combated by F. W. Putnam 
and 0. T. Mason. Mason was quite willing to swallow a 
moderate dose of diffusionism ; even to interpret the spe- 
cific features shared by canoes on the Amur and the 
Columbia as proof of contact.^^ 

Comparative Economics 

To turn to other developments, there were whole de- 
partments of culture perforce neglected by earlier writ- 
ers that Vs^ere at last being seriously attacked. Prominent 
among these was primitive economics. Prehistorians, of 
course, taught that agriculture and animal husbandry 
were preceded by hunting and gathering, but the history 

11 Pitt-Eivers, * ' On the Egyptian Boomerang and Its Affinities, * ' 
JRAI 12:454-463, 1883. Colonel Yule, "Notes on Analogies of Manners 
between the Indo-Chinese Races and the Races of the Indian Archipelago," 
Ibid., 9:290-304, 1880. E. B. Tylor, "On the Game of Patolli in Ancient 
Mexico and Its Probably Asiatic Origin," Ibid., 8:116-129, 1879; "Re- 
marks on the Geographical Distribution of Games," ibid., 9:23-29, 1880; 
' ' Old Scandinavian Civilisation among the Modern Esquimaux, ' ' ibid., 
13:348-356, 1884. Report on World's Fair in A A 6:425, 1893. O. T. 
Mason, ibid., 8:113, 1895. 


of cultivated plants and domesticated animals was very 
inadequately known. A. L. P. P. de Candolle's Origine 
dcs plant es cultirees did not appear until 1883, Eduard 
Halm's Die Haustiere und Hire Bezieliungen zur Wirt- 
schaft des Menschen until 1896. It was the generation 
before the turn of the century, then, that brought major 
enlightenment. Miss Buckland specifically concerned her- 
self with primitive agriculture and, along with some 
more questionable views, advanced the sound idea that 
cereals were not necessarily the first species to be 
brought under cultivation but might well have been ante- 
dated by roots and fruits. She also stressed the promi- 
nence of women ''exclusively in agricultural pursuits 
among the low^er races" — an idea that attained great 
prominence in subsequent discussion. Eoth, who pursued 
the same subject at somewhat greater length, likewise re- 
garded women as the first cultivators.^" Illuminating 
ideas on the psychology of animal domestication were 
thrown out by Francis Galton, whose researches tended 
to show that all species domesticable had actually been 
domesticated.^^ Scholars were also recognizing the im- 
portance of cultivated plants and domestic animals as 
evidence of historical relationship. Obviously a plant 
that was not a part of a regional fauna must have been 
imported in its cultivated form, as was argued for the 
species found in Swiss lake dwellings. Thus, botanists 
and zoologists came to be impressed into the service of 
anthropology." There appeared monographic studies of 
the uses to which domestic species were put in different 
regions, such as Von Tschudi's paper on the llama and 

"A. W. Buckland, "Primitive Agriculture," JRAI 1:3, 17, 1878. 
H. Ling Eoth, "On the Origin of Agriculture," ihid., 16:102-136, 1887. 

13 Francis Galton, Inquiries into Human Faculty, 243 sq., London, 
1883 (Originally in Trans. Ethnol. Soc. 1865). 

1* See, e.g., A. Braun, ' * tJber die im Kgl. Museum zu Berlin 
aufbewahrten Pflanzenreste aus altagyptischen Grabern," ZE 9:289 sq., 


Radloff's contributions to early volumes of the Zeit- 
schrift fur Ethnologie on Turkic beasts, later reissued 
in his book on Siberia." 


Within our present period also fall the first system- 
atic investigations of primitive art. In the early eighties 
Hjalmar Stolpe of Stockholm began copying the orna- 
mentation on objects in the principal European museums 
and later added to his materials during a voyage around 
the world. His general conclusions, first issued in the 
Swedish journal Ymer in 1890, were made accessible in 
English a year or two later, under the title On Evolution 
in the Ornamental Art of Savage Peoples. To these he 
added in 1896 Studies in American Ornamentation — a 
Contribution to the Biology of Ornament}^ 

Stolpe rendered an unquestionable service by his 
resolute attempt to define regional styles in Polynesia, 
which incidentally brought out the amazing diversity of 
decoration within this circumscribed area inhabited by 
closely related peoples. Of less value, though at least 
equally influential, was his attempt to trace the develop- 
ment of ornament from realistic figures to pure geomet- 
rical designs ''through a series of intermediate forms." 
In this he had been anticipated by Pitt-Rivers and F. W. 
Putnam, but Stolpe applied this biological conception on 
a larger scale. According to him, the savage hunter, on 
attaining periods of leisure, would begin to decorate his 
implements, carving into a more realistic image a piece 
of wood offering a chance resemblance to the beast he 
pursued. ''These animal figures, at first realistic, would 
become in time such invariable adjuncts that a mere in- 

15 J. J. Von Tschudi, "Das Lama," ZE 17:92, 1885. W. Radloff, Aus 
Sibirien, Leipzig, 1893. 

1^ The English translations have been reprinted as Collected Essays in 
Ornamental Art, Stockholm, 1927. 


dication would be suflBcient to satisfy the craving for 
tlieir prosonco, and tliis chaii.Gfo, through the awakened 
desire for synnnolry and for covering the entire surface 
may have been so great that the figures pass into mere 
lines, or what used to be called geometric ornament." 
Practically all primitive ornament was thus derived 
from zoomorphs, plant originals being assigned to a 
higher level. In the most emphatic way Stolpe denied 
that primitive man knew any geometric figures as such: 
'*"\Ylien at any time he employs similar figures they have 
for him an entirely concrete significance. He sees in them 
either a sign of the object from which they evolved, or 
else a picture of the thing itself. ' ' 

It is important to note a psychological assumption 
that underlies Stolpe 's reasoning. Denying to primitive 
man a purely aesthetic or play impulse, he sought some 
deep — especially a religious — meaning behind the deco- 
ration and applied this idea to tattooing no less than to 
the designs or artifacts. Stolpe did not dispute the in- 
fluence of textile techniques on ornamentation, especially 
in producing rectangularity ; but he assigned to this cause 
merely transforming, not creative, potency. 

Whether through Stolpe 's influence or independent 
thinking along the same lines, this basic view was widely 
accepted, dominating von den Steinen's study of Bra- 
zilian designs and A. C. Haddon's Papuan researches, 
the latter, however, also embodying a noteworthy at- 
tempt to define regional styles.^^ A quite distinct position 
was taken in America by W. H. Holmes, who stressed the 
effect of technical processes. Basketry techniques, more 
particularly, automatically produce decorative patterns 
which would extend their sway when copied on pottery 

"A. C. Haddon, The Decorative Art of British New Guin-ea, in Cun- 
ningham Memoirs, Royal Irish Academy, Dublin, 1894 ; Idem, Evolution in, 
Art, London, 1895. 


or in carving." This fruitful idea evidently had impor- 
tant implications. If the very act of making a basket 
yielded geometrical forms, Stolpe was no longer war- 
ranted in tracing such designs to a representative 
motive. Further, an important historical conclusion re- 
sulted. For, on this hypothesis, remote peoples could in- 
dependently evolve similar decoration from similar 
techniques; there would be parallelism of art develop- 
ment even though the techniques might conceivably have 
come from a single source. 

Art, of course, is much broader than ornamentation ; 
and actually there was one attempt — Ernst Grosse's Die 
Anfdnge der Kunst (1894) — to cover the entire field. But 
the time was not yet ripe for a significant discussion of 
either literature or music because the basic facts were 
as yet unavailable. In order to discuss primitive literary 
effort a mass of linguistically secured texts was prereq- 
uisite, and these were only beginning to accumulate in 
appreciable number. As for comparative music, the even 
greater dearth of trustworthy raw material was fatal. 
What Dr. von Hornbostel says of the Fuegians largely 
holds true of all primitive music before the systematic 
use of the phonograph: ''Of Fuegian native music 
scarcely anything was known until recently, except a few 
poor notes scattered in ethnological literature, and half 
a dozen musical examples recorded by ear and hence of 
doubtful reliability."'" 

Society and Religion 

On social organization the period hardly reveals any 
advance comparable to that achieved by Maine, Morgan, 
or Tylor. Edward Westermarck (1862- ), a Swedish 
Finn who has taught at Helsingfors, London, and Abo, 

'^W. H. Holmes, "Origin and Development of Form in Ceramic Art," 
BAE-R 4:443-465, Washington, 1886. 

"Erich M. von Hornbostel, "Fuegian Songs," A A 38:357, 1936, 


aroused great notice by his treatise on The History of 
Human Marriage (London, 1891; 5tli edition, 1921), 
which was translated into several hinguages. Its main 
propositions are concisely set forth in A Short History 
of Marriage (London, 1926). Heavily documented. The 
History of Human Marriage on its first appearance im- 
pressed many as a definitive counterblast to the more 
questionable aspects of Morgan's teachings. It rejected 
primitive promiscuity, describing the family, with a pa- 
ternal protector, as the earliest form of social unit, one 
already prefigured among anthropoid apes. Wester- 
marck suggested reasonable alternative explanations for 
phenomena often explained as relics of promiscuity or 
group marriage ; and he repudiated the dogma that mat- 
rilineal had uniformly preceded patrilineal descent. 
However, Maine had previously voiced similar points of 

More original is Westermarck's theory of incest. In 
his later formulation he epitomizes it as follows : ' ' Gen- 
erally speaking, there is a remarkable absence of erotic 
feelings between persons living very closely together 
from childhood. Nay more, in this, as in many other 
cases, sexual indifference is combined with the positive 
feeling of aversion when the act is thought of. This I 
take to be the fundamental cause of the exogamous pro- 
hibitions. Persons who have been living closely together 
from childhood are as a rule near relatives. Hence their 
aversion to sexual relations with one another displays 
itself in custom and law as a prohibition of intercourse 
between near kin. ' ' ^° This idea, while plausible, is not 
easily tested, and certainly the ultimate verdict on its 
value rests with psychologists. 

Reverting to Westermarck's sociological ideas, we 
unenthusiastically note his essential agreement with 

20 Edward Westermarck, A Short History of Marriage, 80, London, 


present views. Westermarck is very widely read, and his 
original researches in Morocco,^' though only appraisa- 
ble by Islamists, bear the earmarks of scholarship. His 
use of aboriginal data, however, is unsatisfactory. In- 
deed, his own account of the procedure when preparing 
one of his later books is not apt to arouse confidence: 
'*I made use of the same method as I had employed in 
my book on marriage. I made my excerpts on slips of 
paper, which I numbered according to subject-matter, so 
that afterwards I should be able so much the more easily 
to group together all data bearing upon the same ques- 
tion: homicide, theft, love of truth and falsehood, adul- 
tery, cannibalism, and so on. . . ."^"^ This approach is 
doubly suspect: first of all, the classification does not 
grow naturally out of the material but is imposed on it ; 
secondly, the collector is likely to concentrate only on 
what seems to fall under his rubrics, omitting correlated 
phenomena of the utmost significance. 

There is an obvious reason for these deficiencies. 
Westermarck is not primarily interested in culture; he 
is a philosopher who uses its phenomena to illustrate his 
points. Unlike Tylor, therefore, he makes no effort to 
assimilate all the relevant data. When Tylor cites a 
North American fact, there is in the back of his con- 
sciousness a picture, accurate in the light of what was 
then known, of all pertinent aspects of life. With Wester- 
marck the reader has the uncomfortable feeling that 
nothing interests him less than to comprehend primitive 
tribes as a culture historian would like to understand 
them. Sweeping generalizations of his, chosen here and 
there, will illustrate his ethnographic inadequacy. In di- 
vorce, we learn, ''among a large number of peoples all 
the children generally follow the mother. This is espe- 
cially the case where descent is matrilineal, and among 

21 E.g., E. A. Westermarck, Bitual and Belief in Morocco, London, 1926. 

22 Edward Westermarck, Memories of my Life, 101, London, 1929. 


the native tribes of North America it seems to be the 
general rnle."""^ But though descent through the mother 
is common enough, there are numerous instances of 
North American tribes with the reverse law or no defi- 
nite rule at all. In North America, again, ' ' the wife must 
carefully keep away from all that belongs to her hus- 
band's sphere of action,"'* — a statement so vague as to 
mean nothing and certainly not correct in the absolute 
way indicated. Indeed, three or four pages further, we 
read that Pawnee women figured at tribal councils ; that 
Puget Sound wives were always consulted before a bar- 
gain was closed ; and that an Omaha husband would not 
give away anything without his spouse's consent. How 
these specific utterances are to be harmonized with the 
preceding general statement is not easily seen. 

Throughout we painfully miss Tylor's careful sift- 
ing of evidence. The profusion of Westermarck's docu- 
ments has blinded some critics to his amazingly 
uncritical use of them. Not only are there inconsistencies 
of the kind just cited, but bad, good, and indifferent 
sources are cited indiscriminately. Even good sources 
are abused: ''Lewis and Clarke," writes Westermarck, 
"affirm that the status of woman in a savage tribe has 
no necessary relation even to its moral qualities in gen- 
eral." And he goes on quoting his authorities to the ef- 
fect that "the tribes among whom the women are very 
much debased, possess the loftiest sense of honor, the 
greatest liberality, and all the good qualities of which 
their situation demands the exercise." ^^ Tylor would 
have inquired into the opportunities the writers had for 
such comparative judgments and into their educational 
equipment for forming, let us not say an objective, but 

23 Idem, A Short History of Marriage, 279. 

^* The Origin and Development of the Moral Ideas, 1:636, London, 

^^Ibid., 1:647. 


at least a reasonably convincing, view in so delicate a 

In short, Westermarck neither appraises his evi- 
dence discriminatingly nor becomes absorbed in his cul- 
tural phenomena ; and while his views on early family life 
largely coincide with current doctrines, we are not able 
to discover any signal advancement of ethnology due to 
his writings. After all, the study of culture is not likely 
to be greatly promoted by those uninterested in its data. 
Nevertheless, Westermarck, by boldly challenging what 
he supposed to be the general view of anthropologists, 
at least helped to concentrate attention on vital issues. 
Tylor's contemporary review of The History of Human 
Marriage acknowledges its industry and independence 
and, as was then natural, praises its blending of biologi- 
cal and sociological points of view. On the other hand, it 
exposes the one-sided underestimation of matrilineal in- 
stitutions and makes it clear that Westermarck was not 
the first to reject primitive promiscuity, which, indeed, 
Tylor himself had never sponsored.^** 

While we cannot attach outstanding significance to 
Westermarck 's bulky tomes, another author appearing 
shortly after the turn of the century marks a real epoch 
in the study of social organization. Heinrich Schurtz 
(1863-1903) was a pupil of Eatzel (page 119), and ab- 
sorbed his master's ideas on diffusion without, however, 
relinquishing the older form of parallelism. But his im- 
portance lies in an entirely different direction. His Al- 
tersklassen und MdnnerbUnde (Berlin, 1902), while 
replete with adolescent subjectivism on a variety of ir- 
relevant topics, for the first time summarized those 
associational activities independent of blood-ties which 
previous treatises had ignored. The picture of primitive 
society was thus radically altered. A person was no 
longer to be conceived as merely a member of a family 

2«r/ie Academy, 40:288f., 1891. 


or clan ; he belonged simnltaneously to an age-class, a 
club, a secret fraternity. It is true that Schurtz commit- 
ted a pioneer's errors. He underestimated the associa- 
tional capacity of women, he offered an artificial genetic 
scheme, he committed quaint errors of detail, as when 
he connected the age-societies of the Plains Indians with 
totemism. But his faults are more than compensated by 
the new insights he afforded, the new problems he 
broached or suggested. Among these we note the influ- 
ence of men's organizations on political structure and 
their relation to the kinship groups. The timeliness of 
Schurtz 's achievement is indicated by two facts. Inde- 
pendently of Schurtz, though several years later, the 
American sociologist Hutton "Webster similarly united 
relevant material in his meritorious book on Primitive 
Secret Societies (New York, 1908). On the other hand, 
the need for more descriptive data led to the first inten- 
sive professional investigation in the field — Kroeber's 
researches among the Arapaho of Oklahoma and Wy- 
oming.^^ This, in turn, prompted a whole series of com- 
parable studies by the same institution, largely with the 
view of testing Schurtz 's theory of the age factor as a 
determinant of social solidarity. 

In a widely read work^ Van Gennep (1873- ) 
hardly does justice to the enlargement of perspective 
due to Schurtz. But he suggestively supplements it, as 
well as Tylor's somewhat summary treatment of ritual- 
ism, by concentrating on the rites of initiation connected 
with age-classes and secret organizations. These he 
aligns with ceremonies of admission into castes and pro- 
fessions and even with coronation ritual. Initiation, in- 
deed, figures in Van Gennep 's system as only one of a 
large series of ceremonies linked with the life cycle and 

2^^ A. L. Kroeber, ' ' The Arapaho : Ceremonial Organizations, ' ' 
AMXH-B 18:151-229, 1904. 

28 Arnold Van Gennep, Les rites de passage, Paris, 1909. 


other critical periods. He is concerned with disengaging 
from the multiplicity of detail a common sequence, ''le 
schema des rites de passage." Essentially there are 
three steps: rites de separation, de marge, d'agregation. 
The purpose of this category of ceremonies is to lead the 
individual from one status to another. Sometimes, as in 
mortuary observances, the preliminary separation is em- 
phasized, sometimes, as in initiation, the intermediate 
step {de marge) of the novitiate looms important. But 
quite generally the ceremonial performer is first segre- 
gated from his or her normal social setting; then re- 
mains for a while in a neutral state; and is at last 
formally reintroduced to a recognized social position. 
This would be typically illustrated by the seclusion of a 
prospective primipara; her continued separation with 
definite taboos; and her final promotion to the status of 
a ** mother." 

Van Gennep himself seems aware of the schematism 
into which such a classification might degenerate and 
explicitly recognizes that the rites in question have as- 
pects beyond their transitional elements. With these 
qualifications his book is a welcome contribution, for it 
defines a large set of phenomena and assists in their de- 
scriptive analysis. The relative validity of his basic con- 
cept is certified by its repeated application to field data 
by such observers as E. C. Parsons, H. Junod, and A. W. 

Within the latter part of the Tylorian period also 
falls the rise of Sir James George Frazer (1854- ). 
His little book on Totemism, published in 1887, was the 
precursor of the gigantic Totemism and Exogamy (4 
vols., London, 1910), itself ultimately dwarfed by the 
final edition of The Golden Bough (1st edition, 1890; 3d 
edition, 12 vols., London, 1911-1915). Among his other 
works may be cited Folk-Lore in the Old Testament 
(London, 1918). 


Frazer's reputation is probably unparalleled by that 
of any contemporary writer on anthropological subjects. 
If our treatment seems unduly curt, there is a reason. 
Undoubtedly Frazer's assiduous compilation of data, 
often from recondite sources, is very helpful; but after 
Maine and Tylor and McLennan, he disappoints by his 
failure to grapple with problems in a thoroughgoing 
fashion. To put it briefly, he is a scholar, not a thinker 
— and a scholar, moreover, who in his eagerness to as- 
similate descriptive data has somewhat perversely ig- 
nored the strides of theory. Thus, he has hardly kept 
abreast of the attempts to supplant the older parallelist 
schemes with a more critical insight into the effects of 
tribal intercourse; and his interpretations suffer from 
an a priori use of vulgar psychology, with constant 
lapses into a false rationalism. With his intellectual 
limitations we could hardly expect novel insight into so- 
cial organization nor in any department the creation of 
many significant new concepts. Some of his confusions 
were, indeed, already pointed out by Tylor.^^ 

But in the study of comparative religion, which is 
evidently most congenial to Frazer, we must concede to 
him more than a mere accumulation of raw fact. To be 
sure, he champions a number of indefensible proposi- 
tions. He contrasts magic with religion, following 
Tylor 's rationalistic classification of the former as a 
pseudo science on the ground that it attempts to coerce 
nature in conformity with an immutable law of causality. 
This is psychologically objectionable, because magical 
beliefs are commonly saturated with the same reveren- 
tial attitude towards the associated rites which Frazer 
restricts to religion. As Marett (page 109) and Golden- 
weiser suggest, both are properly classed as forms of 

29 E. B. Tylor, "Remarks on Totemism, with Especial Reference to 
Some Modern Theories Respecting It," JRAI 1:138-148, 1899. 


"supernaturalism." Chronologically, Frazer also errs: 
he supposes magic to precede religion because it is psy- 
chologically simpler than the conception of personal 
agents — a wholly a priori contention; and, further, be- 
cause its uniformity as contrasted with the multiple 
forms of cult implies priority. However, there is no such 
homogeneity of magical practice: magical formulae, 
prominent in Oceania, Siberia, and Arctic America, are 
lacking over large areas of North America; contagious 
magic is not found in Central Australia; and so forth. 
Besides, homogeneity does not logically involve antiq- 
uity; one would rather suppose that the more ancient of 
two systems of thought would be liable to more manifold 

Nevertheless, Frazer 's discussion has solid merit. It 
contrasts with the utmost clarity two antithetic attitudes. 
On the one hand, a worshiper supplicates superior pow- 
ers ; on the other, the possessor of extraordinary knowl- 
edge uses it to effect desired results, with or without the 
aid of spirits he controls. It is true that in practice the 
two antagonistic principles may be joined. Thus, the oc- 
cult information may itself be granted by a god in an- 
swer to a humble prayer. Nevertheless, there are these 
two extremes about which supernaturalism revolves, and 
Frazer 's formulation has been of great value in classi- 
fying relevant phenomena and defining them in particu- 
lar cases. For example, Frazer himself has correctly 
pointed out the '* conspicuous predominance of magic 
over religion" among the Melanesians of New Guinea; 
and according to Bunzel what among the Pueblo Indians 
formally represents a prayer *'is never the outpouring 
of the overburdened soul," but "more nearly a repeti- 
tion of magical formulae" with no trace of humiliation 

30T7ie Golden Bough, 11-60, New York, 1922. Cf. W. Schmidt, Der 
Vrsprung der Gottesidee, 1:510-514. 


before divine beings.''* On the other hand, the Plains In- 
dian typically invokes the commiseration of the super- 
natural powers by drawing their attention to his 
miserable plight. Similarly, Frazer's explicit subdivision 
of magic into "imitative" and "contagious," while 
hardly exhaustive, helpfully ranged masses of pertinent 
material. Finally, his conception of taboo as negative 
magic, staving off misfortune as positive occult tech- 
niques ensure good luck, is both novel and stimulating 
if not wholly convincing.^" 

Frazer, then, can certainly not be ignored in the 
study of the development of thought on comparative re- 
ligion. But in my opinion his proper place is in the his- 
tory of English literature and of the intellectual classes 
of Europe. His style, overornate for some tastes, is un- 
questionably a thing of beauty; and this gift, coupled 
with remarkable erudition, has enabled him to imbue his 
readers with that sense of perspective in envisaging the 
phenomena of civilization which ethnology conveys to its 
votaries. The anthropologist assumes this vision as a 
foregone conclusion and asks for an illumination of 
special problems; and in that respect Frazer's services, 
while not negligible, shrink to moderate proportions. 

Primitive Mentality 

Racial psychology, which Waitz had examined in its 
broader aspects, required a more technical investigation. 
The first adequate systematic research on any one primi- 
tive group was conducted by Rivers, who subjected his 
Torres Straits Islanders to the tests perfected in the 
psychological laboratories of Europe (see page 170). But 
rather earlier there was a partial step in this direction, 

31 Preface to Br. Malinowski, Argonauts of the Western Pacific, xiv, 
London, 1922. Euth L. Bunzol, "Zufii Eitual Poetry," BAE-E 47:615, 
618, Washington, 1932. 

32 See B. R. Marett, Psychology and Folk-Lore, 192 f ., London, 1920. 


oddly enough stimulated by the British statesman Glad- 
stone, who piqued himself on his classical scholarship. 
From the dearth of Homeric color designations he in- 
ferred an inferior color sense, which precipitated a very 
lively discussion as to possible racial differences.^^ The 
upshot was to rule out deficiencies in vocabulary as in- 
dicative of racial inferiority — a minor contribution to 
the general belief in psychic unity. 

More significant for ethnology was the revolution 
of psychology inaugurated by Francis Galton's concept 
of individual differences in his Inquiries into Human 
Faculty (London, 1883). Boas was profoundly impressed 
by Galton's personality and soon introduced his idea into 
ethnology, with consequences both for theory and field 
investigation (see page 134). 

Another reform of anthropological psychologizing 
came from sociological considerations. As pointed out, 
Bastian had already clearly seen that individual psy- 
chology could not cope with the problems of thought and 
behavior because men's mentality is largely determined 
by their cultural setting. But the explanations of primi- 
tive mentality long continued to ignore this factor, to 
treat belief held by a particular savage, for instance, as 
though it had sprung from his individual psyche in re- 
sponse to certain experiences. 

Still more important was the type of mental re- 
sponse that was almost invariably offered. Tylor, him- 
self the child of a rationalistic period, tended to 
represent the savage primarily as a reasoner, as bas- 
ically moved by intellectual promptings and merely mis- 
led by ignorance. ' ' Human custom, ' ' we read, * ' is hardly 
ever wilfully absurd, its unreasonableness usually aris- 
ing from loss or confusion of old sense." In identically 
the same spirit, Elie Reclus defines Australian supersti- 

38 E.g., A. S. Gatschet, * * Farbenbenennungen in nordamerikanischen 
Sprachen," ZE 11:293 sq., 1879. Eabl-Eiickhard, "Zur historischen 
Entwicklung des Farbensinnes, " ibid., 12:210, 1880. 


tions: "They are the consequences reasoned out and 
logically dodiicod from premises, which are admittedly 
false, but justified by appearances: mere optical illusions 
due to an as yet imperfect camera."^* This intellectual- 
ism, which minimized tlie emotional and generally irra- 
tional determinants of civilization, required revision, 
and one of the influences in administering the corrective 
was the French thinker Tarde. 

Gabriel de Tarde (1843-1904), magistrate, chef du 
service de la statist ique in the ministry of justice, and 
professor of modern philosophy at the College de 
France, was a voluminous writer on criminology and the 
philosophy of law as well as on sociology, but we are 
concerned with only one of his books, Les lots de V imita- 
tion (1890, 2d ed. 1895), which profoundly impressed 
Boas and, through him, dozens of anthropologists in the 
United States. It was also translated into English by the 
American ethnographer Elsie Clews Parsons in her 
earlier, sociological phase. 

In judging Tarde we must remember that he was a 
sociologist, and that the essence of his book largely dates 
back to the eighties, when several chapters appeared in 
the form of articles. In other words, while he knew 
Tylor and Lubbock, he was not saturated with ethno- 
graphical information; like Maine, whom he repeatedly 
cites, he largely relies on historical and contemporane- 
ous data. Accordingly, prepared for lapses in the treat- 
ment of special anthropological problems, we merely 
smile when he proclaims the pristine universality of 
bloodthirsty gods and sketches religious evolution 
through human sacrifice, animal sacrifice, and vegetable 
offerings up to spiritual symbolism.^' 

Where Tarde sees more clearly than the contempo- 

3^E. B. Tylor, in JAI 23:236 f., 1880. Elie Beelus, " Contributions H 
la sociologie des Australiens, " in Eev. 3' serie, 1:240 sq., 1886. 
35 Les lots de I 'imitation, 2d ed., 296, Paris, 1895. 


rary evolutionary anthropologists is in his objective at- 
titude towards the civilization of his period. Here there 
is no trace of smugness, no suggestion that in 1885 man 
had reached a peak from which he might look down pity- 
ingly, if not scornfully, upon his predecessors. Tarde 
does not accept the traditional fetiches of modern life, 
such as the jury system, but aligns it with defunct ju- 
ridical methods of procedure: ''It is stupefying to see 
how quickly at certain periods there are diffused crimi- 
nal procedures equally odious and absurd, such as tor- 
ture, or equally inefficient and unintelligent, such as the 
jury system." His picture of revolutionary movements 
tending to sink into as rigid a dogmatism as that against 
which they rebelled is unexcelled: ''The most profound 
revolutions tend to become traditionalized, as it were 
. . ." "The molieristes . . . with their devout attach- 
ment to minor traditions of the French theater, ought 
not to make us forget that their idol, Moliere, was in his 
century an artistic innovator, a man most openminded 
toward innovations, most antagonistic to f etichism. ' ' ^° 
This sane position reacts on the judgment of savagery. 
Unlike Lubbock, who minimizes moral sentiments among 
primitive peoples, Tarde convincingly shows that they 
are identical on their and our level, being simply more 
narrowly applied at the earlier stage.^^ 

The basic concept of the book, however, is the force 
of imitation: "Society is imitation, and imitation is a 
kind of somnambulism." That is to say, imitation pro- 
ceeds irrationally, through prestige suggestion, the in- 
ferior individuals or classes aping their betters — and 
that, irrespective of practical considerations. Men follow 
tradition and sometimes flout it in favor of contempo- 
rary innovations from without; but whether dominated 
by la coutume or la mode, to use Tarde 's terms for this 

^Ilid., 266, 343, 320, 370. 
" Ihid., 376. 


antithesis, in neither case do they submit their views to 
intellectual scrutiny. Nearly half of Tarde's book is de- 
voted to these extralogical influences. Contrary to ap- 
pearances he announces that imitation proceeds ''du 
dedans au dehors," i.e., ideas are imitated first, behavior 
later. This questionable aspect of his theory will engage 
our attention in another context.^* 

Consistently with his major postulate, Tarde em- 
phasizes diffusion in all periods of history. Like other 
adherents of this principle, he assumes sterility of 
imagination and spreading of ideas even without vast 
migrations or conquests. Even in the Stone Age tools 
passed from country to country, and the same holds for 
pottery. If early flint implements are strangely uniform, 
it does not follow that "this similarity was due to the 
spontaneous appearance of like ideas and like wants 
among these primitive men. Nothing could be more arbi- 
trary than this conclusion. ..." The facts merely in- 
dicate wide dissemination. If even the Incas were unable 
to invent the wheel, how can we credit ruder peoples 
with an innate tendency to evolve ceramics? "Thus it 
seems to me fallacious to see in the almost universal dis- 
tribution of this art, the proof of the necessity, the in- 
nateness of certain discoveries." However, he accepts a 
number of separate culture centers in the light of con- 
temporary know^ledge, "des foyers encore indecomposa- 
bles de civilisation. ' ' ^^ 

Imitation, however, presupposes at some time an in- 
vention that serves as a model. Tarde by no means 
neglects this aspect of the matter. Invention is the fusion 
of two or more pre-existing ideas into a new synthesis; 
its laws belong essentially to individual logic, while the 
laws of imitation are in part social, largely extralogical. 
Since such creative synthesis is not calculable, Tarde as- 

38 Ibid., 95, 205-394. 

39 Ibid., 50 f., 53, 105, 109, 352. 


signs a role to historic accident. He acutely recognizes, 
however, the logical interrelation of ideas, whence the 
irreversibility of intellectual progress.*' 

Many of Tarde's generalizations are not capable of 
rigorous proof, and certainly none such was attempted 
by their author. He presented, however, fresh principles 
that could be tested by ethnographic material precisely 
because they were avowedly timeless in their applica- 
bility, holding for societies as such. To sum up, we recog- 
nize two principal contributions emanating from him: 
a detached view of modern civilization, and a psychology 
of social man that did justice to nonintellectual motives 
of behavior. The latter influence presently became effec- 
tive when the young ethnographer Boas investigated the 
growth of secret societies on the coast of British Colum- 
bia. The multiplication of ritual was traced to the promi- 
nence that went with membership in an organization, 
leading tribesmen to seek admission. When this was no 
longer feasible, native imagination created comparable 
societies under the spell of prestige suggestion. "These 
are the strange phenomena treated by Stoll in his book 
on suggestion, and rather more profoundly by Tarde in 
his book on the Laws of Imitation. ' ' " Incidentally, we 
note here the birth of the ** pattern" theory, which 
played a prominent part in later discussion. 

The emphasis on nonrational determinants of group 
behavior and belief, while an ever-recurring principle of 
Boas ', was far from being restricted to him. Among his 
approximate contemporaries we may here single out R. 
E. Marett (1866- ), Tylor's successor at Oxford. 
Steeped in classical and metaphysical studies, this ur- 
bane and balanced spirit has persistently shed light on 
the primitive mind, and especially on the psychology of 

*^Ibid., 109, 411, 413. 

■'i Franz Boas, * ' Die Entwicklung der Geheimbiinde der Kwakiutl- 
Indianer," Festschrift fur Adolf Bastian, 442, 1896. Cf. idem, The Mind 
of Primitive Man, 114, New York, 1911. 


faith. That the savage is ''not perpetually spook- 
haiiiitod"; that religion serves the function of restoring 
contidenoe in crises; that it is too complex to be reduced 
to a single root; that *'the bane of the psychological 
study of human belief is a shallow intellectualism" — 
these are utterances that many ethnological writers 
might well have taken to heart. Specifically, Marett 
wisely distinguishes the savage's workaday world of 
normal experience from the transcendental phase of his 
being, where a sense of mystery supplants common 
sense. Stressing the subjective states of the latter cate- 
gory, Marett finds them identical, irrespective of whether 
animistic notions occur or not, and hence he unites 
Frazer's "magic" and ''religion" in the wider category 
of "supernaturalism." Recognition of the supernatural, 
he is careful to point out, involves no conception of "na- 
ture" in the sense of modern science. The savage "does 
not abstractly distinguish between an order of uniform 
happenings and a high order of miraculous happenings. 
He is merely concerned to mark and exploit the differ- 
ence when presented in the concrete. ' ' *^ 

Marett thus passes beyond the intellectualistic no- 
tion of magic that ensnared Tylor and Frazer. He 
likewise introduced the useful distinction between 
"animatism" and "animism."" Tylor tended to see in 
all personification of inanimate objects the assumption 
of a spirit, a being modeled on the human soul. This is, of 
course, plausible in many instances, but Marett showed 
that we cannot assume it as a logical corollary. To yell 
at a hurricane is indeed to treat it as though it were 
alive, but it is not the same as to imagine a being of re- 
fined essence dwelling within the tempest and directing 

•»2R. R, Marett, The Threshold of Religion (1st ed., 1909), 11, 102- 
121, London, 1914. 
"/hid., 14, 18. 


it. Here, then, there is ''animatism" but not, without 
further evidence, animism. 

In a sheaf of essays, some of which date back to over 
twenty years ago, Marett clearly forestalls contentions 
latterly supposed peculiar to the functionalist school. He 
deprecates mere antiquarianism in the study of folklore, 
aptly asking, ''How and why do survivals survive?" And 
in another context he returns to the matter more spe- 
cifically. The phenomena described under that label may 
indeed "be referable to antecedent historical condi- 
tions ' ' ; but they may also be explicable from ' ' conditions 
operating here and now." This is the precise position of 
men like Malinowski except that in their less temperate 
moments they would ban all history.*^ 

With characteristic poise Marett makes of primitive 
man neither a logic-chopper nor a chronic mystic : ' * The 
savage turns out to be anything but a fool, more espe- 
cially in everything that relates at all directly to the 
daily struggle for existence . . . common sense is no 
monopoly of civilization." Repeatedly our attention is 
called to that solid core of accurate information acquired 
by early man that contrasts so sharply with his fanciful 
theories and looms as the basis of our modern science. 
Marett 's psychology is also thoroughly up to date in rec- 
ognizing both the force of rationalization and the im- 
portance of individual differences.*^ 

Marett has been willing to forego ethnographical 
field research and has never engaged in literary enter- 
prises of Frazerian dimensions. But in post-Tylorian 
England for poise in the judgment of theories or for a 
sympathetic grasp of primitive values there is no su- 
perior to this philosophical humanist. 

** Psychology and Folk-Lnre, 13, 123-127, Lonaon, 1910. 
*^IMd., 198. Anthropology, 227, 242-246. 



Not even Tarde stressed the irrational factors of 
civilization more vigorously than did a German theorist 
of economic development whose most important publica- 
tion falls within our period. Eduard Hahn (1856-1928) 
received his training in geography under the celebrated 
explorer of China, Baron Ferdinand von Richthofen. To 
Hahn we owe by far the fullest treatment of domesti- 
cated animals.^*^ Later writers have dealt with the sub- 
ject from a mainly zoological angle and have omitted 
species unimportant in our own economic system. Hahn's 
interests embrace zoology, geography, economics, and 
culture; he neglects neither the silkworm nor the mus- 
covy duck, neither the guinea pig nor the bee. 

According to Hahn, domestication involves free 
breeding in captivity, and it is this condition that ren- 
dered the process enormously difficult. All sorts of spe- 
cies have been kept as pets or even, like the elephant, put 
to practical tasks, but they fail to multiply when re- 
moved from a state of nature. As for the original motives 
of animal husbandry, Hahn convincingly eliminates 
several that would occur to a naive speculator. Sheep 
could not have been raised for shearing because in the 
wild condition they lack wool, which is a by-product, not 
an antecedent of domestication. Similarly, a cow natu- 
rally yields milk only for calves, any surplus having 
originally been too insignificant to warrant impounding; 
milch cows, too, are an end result. Correspondingly, wild 
fowls do not lay eggs in such abundance as to tempt early 
man into keeping them for such purposes. What is more, 
we know various chicken-breeding tribes that never eat 
either eggs or poultry, while Eastern Asiatics and Indo- 
nesians are averse to milk drinking. Hahn suggests that 

^^ Die Haustiere und ihre Beziehungen zur Wirtschaft des Menschen, 
Leipzig, 1896. 


people kept poultry originally as alarm clocks or for 
cockfights — both noneconomic motives. This subordina- 
tion of practical consideration Halm carries to an ex- 
treme, as one of his admirers freely admits.*^ At times 
it almost appears from his account as though primitive 
man had no technical problems to solve, as though he 
viewed nature exclusively from the angle of supernatu- 
ralism ; and one marvels at his being able to survive with 
so complete an aversion from common sense. The truth 
is, of course, intermediate between the intellectualism 
combated and the irrationalism put in its place. Hahn's 
ideas on domestic animals, however, form only part of 
a system and must be understood in that setting. 

A cardinal principle of the theory is the repudiation 
of the traditional three economic stages: hunting, herd- 
ing, and farming. Hahn refutes this sequence simply by 
pointing to the innumerable pre-Columbian Indians who 
farmed without owning live stock. Animal husbandry is 
thus certainly not a necessary antecedent of tillage. For 
this idea he yields priority to Alexander von Humboldt, 
who had already cited the same evidence. As a matter of 
fact, in a meritorious historical study Koppers has 
shown that the great naturalist was in turn preceded by 
the Swiss historian I. Iselin, who as early as 1786 chal- 
lenged the accepted views from Polynesian data.** 

From this critique emerged a positive conclusion 
simultaneously arrived at by Ratzel, though more amply 
elaborated by Hahn.*® If American and Oceanian tribes 
tilled without live stock, such cultivation represents a 
distinct type. Thus was conceived the antithesis of primi- 

*THahn, op. cit., 79, 154, 300. Ulrich Berner, "Kationales und Ir- 
rationales in der Wirtschaftsentwicklung primitiver Volker, " ZE 62:210- 
214, 1930. 

48 Wm. Koppers, ' ' Die ethnologische Wirtschaftsf orschung', ' ' A 10- 
11:611-651, 971-1079; 1915-1916. 

^^ Eduard Hahn, Die Haustiere, 388 sq. ; Van der TTackc sum Pflug, 
37, Leipzig, 1919; idem, Das Alter der wirtschaftlichen Kultur, 28, Heidel- 
berg, 1905. 


tive *' hoe-culture" and "plough-culture," the latter 
being the exclusive mark of higher civilizations. Each 
is a set of correlated traits. The simpler type implies not 
merely the hoe but its use by women in relatively re- 
stricted plots. This sociological point, already fore- 
shadowed by Bachofen, Buckland, and Roth, attains 
great prominence in Hahn's scheme, ploughs being char- 
acteristically linked with men. Preferably Hahn identi- 
fies *' agriculture" {Ackerhau) only with ploughing, 
beasts harnessed to a plough being guided by men to 
draw furrows in an extended field. 

The revision of the three-stage scheme rests on this 
dichotomy. ''Agriculture," by definition, follows domes- 
tication; hoe-culture is independent of it. Primeval farm- 
ing grew directly out of feminine gathering in the earli- 
est economic stage, which Hahn describes as omnivorous 
but with a flesh diet rather subordinate. There is thus 
not so much a ''hunting" stage as one of hunting and 
gleaning, with women responsible for the vegetable fare. 
"Women invented work, for early man was an idler, oc- 
cupying himself now and then with useful labor and 
rather as a pastime than with serious intent. Ultimately 
men did come to control various livestock species, but 
this could not happen on the gleaning level because the 
instability of a hunting-gathering life precluded the keep- 
ing of animals until they would reproduce. Agriculture 
rose when primitive hoe-culture was combined with ani- 
mal husbandry, the ox (which Hahn considers the pri- 
mary livestock species) being made to draw a plough. 
This yields, then, the sequence of (a) hunting-gathering; 
(b) hoe-culture; (c) hoe-culture with stock-breeding; (d) 
"agriculture." What, then, about pastoral nomad- 
ism? Here is another characteristic element of the 
scheme. Hahn refuses to regard pastoralism as an inde- 
pendent economic type because herders, in all but a 


handful of exceptional instances,"" depend on neighbor- 
ing farmers from whom they get the vegetable food to 
eke out their diet. Herding peoples, then, are in essence 
degenerate representatives of stage (c) who under spe- 
cial circumstances have lost hoe-tillage, making shift 
with their herds, yet leaning as far as possible on near-by 

Hahn's later writings are unfortunately flavored 
with a persecution mania and a Messianic complex, which 
strangely enough is tempered with a redeeming modesty. 
Irrelevancies abound; the unsuspecting reader learns 
what the author thinks about the celibacy of Catholic 
priests, British free trade, and the wicked German so- 
cialists. Even disregarding these digressions, we are 
often repelled by fantastic hypotheses and dogmatic as- 
sertions. For reasons not at all clear Hahn finds the 
focus of his "agricultural" complex in Babylonia rather 
than in Egypt, blandly admitting the absence of proof 
in the cuneiform inscriptions. With still greater sang- 
froid he confesses his complete ignorance as to Baby- 
lonian pig-breeding, only to add that it certainly 
originated precisely in this area {gerade in diesen 
Gehieten) and certainly in connection with the cult of the 
great national goddess. He cavalierly denies any appre- 
ciable antiquity to the civilization of India — a view now 
refuted by the excavations at Mohenjo-Daro." 

Turning to the essential aspects of his scheme, we 
find that rather serious exaggerations mar the account 
of every one of his basic types. He describes excellently 
woman's economic contributions in the pre-horticultural 
stage, rightly stressing her knowledge of complex pro- 
cedures for rendering vegetable food possible or pal- 
atable; but he minimizes beyond all reason the 

5*5 Die Raustiere, 132 sq. 

SI Vas Alter . . ., 107, 159, 195. 


complementary activities of men." What is worse, his re- 
stricted etlinographical perspective leads to an under- 
estimation of mankind at this stage. Hahn spurns the 
idea that "hunters" could have domesticated livestock, 
because he fails to note the fishermen occupying per- 
manent villages, as in British Columbia, who are thus 
hardly less stable than many rude tillers.''' Gudmund 
Hatt, Wilhelm Schmidt, and Koppers have justly criti- 
cized this feature of the theory. Further, the hoe is 
not the universal implement of simple farmers, since 
many Oceanians and Americans wield dibbles exclu- 
sively. But even if we widen the concept to make it 
include whatever is not a plough, this abstraction still 
fails to be preponderantly a feminine tool. Women did 
probably invent farming as a consequence of earlier root- 
digging, but in the historic era many Africans, Ajneri- 
cans, and Oceanians assign farming wholly or in part 
to men. The reverse proposition — that ploughs are ex- 
clusively masculine, comes much nearer to the truth — 
yet it is not without important exceptions, such as Arthur 
Young observed in traversing France just before the 
Revolution. Again, the ox may have been the first stock 
animal, but prehistoric evidence does not yet prove its 
priority to the ass, the pig, the sheep, or the goat. Fi- 
nally, it is indeed true that herders, like other human 
beings, crave a varied diet, so that they readily trade 
or extort grain from farming populations, but this does 
not in any sense make them dependent on their neigh- 
bors. Arabs can subsist for weeks on camel's milk, the 
Turkic and Mongolic peoples of Asia for months on 
fermented mare's milk. Tribes that have large flocks 
and herds of horses, cattle, sheep, goats, and camels, 
milk the females of all these species, prepare cheeses, 

52 Von der HacJce zum Pflug, 27 sq. 

53 2)a« Alter . . ., 92 sq. 


and eat the flesh of their beasts are certainly as self- 
supporting as any society can be. 

However, Hahn's absurdities and dogmas constantly 
jostle ideas that are sound, striking, and original. As his 
survey of domestication remains unrivaled, so he has 
defined more problems and thrown out more suggestions 
than any other writer on comparative economics. Though 
renouncing priority, Hahn more than anyone else elimi- 
nated the old three-stage theory from serious discussion, 
and his concept of hoe-culture marked a tremendous 
step in advance. It involved far more than a segregation 
of simpler from higher cultivators. Hahn realized that 
his system, like others, could not simultaneously dis- 
tinguish types and take cognizance of transitional forms. 
But he compensates explicitly for this defect inherent 
in all classification. His merit lies precisely in detecting 
both essential difference and essential likeness. He sees 
that in rude farming it is root crops and vegetables, 
not cereals, that preponderate ; but he traces hoe-culture 
down to our kitchen orchards that persist alongside of 
"agriculture." He contrasts Occidental fruit-growing 
with that of primitive peoples on the basis of its graft- 
ing technique. He distinguishes South Chinese "horti- 
culture" (Gartenhau) from "agriculture" and assigns 
to it the highest place in his scheme because its use of 
irrigation and fertilizers ensures a more intensive ex- 
ploitation. But he derives horticulture from hoe-culture 
and is willing to credit aboriginal Peru and Mexico with 
at least an approach to this highest plane of tillage. 

Hahn's discussion of domestic animals is full of 
theoretical import. Like Tarde, he anticipates the "pat- 
tern" principle: the milking technique, for example, 
once invented, was transferred from one species to an- 
other; the horse was ridden only after riding had been 
developed with the ass and camel; reindeer breeders 
modeled their procedure on the experience of cattle and 


liorso Imsbandry. Saturated with the difficulty of domes- 
tication, llaliii inclines to single, or at best a few, centers 
of dilTiision. This he does not necessarily identify with 
an actual importation of beasts; the mere idea of sub- 
jecting a particular species may have been passed on. 
Thus he does not positively assert that China derived 
its pigs from Western Asia, or vice versa, but he con- 
siders some connection conceivable by way of * ' a possibly 
very weak stimulation." 

Hahn strikingly sets forth the contrast betw^een 
Eastern Asia and the Near Orient in the repudiation 
and use of dairying. If he erred in representing herders 
as parasites on horticulturists, he at least helped us 
see the problem of nascent pastoralism in a new light. 
Obviously, the first herders could not subsist on milk; 
insofar as they developed — contrary to Hahn's theory — 
from a hunting condition, they could have utilized their 
beasts only for transport and for their flesh. 

Hahn certainly tried to correlate the several aspects 
of a culture, say, the social position of the sexes with 
the division of labor. In this spirit he also approached 
technology. In what sort of setting, he asks, could a beast 
have been first harnessed to a plough? It is inconceivable 
that a hitherto untrained ox should pull a cultivator; 
we must assume that he had already been accustomed to 
drawing a wheeled cart. In visualizing the origin of the 
wheel, to be sure, Hahn is at his worst. Eager to vindi- 
cate the role of supernaturalism, he assumes that the 
cart originated as "a model by which the votaries of 
the Babylonian astral faith imitated on earth the move- 
ments of their celestial deities." This miniature con- 
veyance, sprung from the brain of ''an idle priest," 
was constructed of a spindle with whorls. In enlarged 
guise it was later taken beyond the temple precincts, and 
streets were built on ** which the gods might roll along 
in chariots. Very gradually, like many other things, the 


divine carriage was degraded to a utensil of daily 
life."'* This far-fetched association of ideas recalls 
Bachofen, but it likewise demonstrates an essential func- 
tionalism. Hahn, however, was also a historian. Imbued 
with the complexity of culture, he deprecates simple evo- 
lutionary schemes, a position strengthened by his stress 
on irrational motives. Finally, while not an extremist, 
he constantly applied diffusionist principles, as already 

Hahn's contribution is easily summarized. He raised 
comparative economics to a new plane ; and for his work 
as a whole there is not yet an adequate substitute. He 
stimulated Boas and a host of other writers. Specifically, 
he exerted a deep influence on Laufer, some of whose 
most characteristic views are admittedly derived from 
Hahn. Thus his correlation of handmade pottery with 
women, in contrast to the association of men with the 
wheel is obviously modeled on Hahn's allocation of the 
hoe to woman and the plough to man (page 116). Not- 
withstanding psychological disparity, Hahn bears in his 
reputation some resemblance to Morgan. Both had a 
restricted range of interests, a crotchetiness that at 
times led to absurdity; but each concentrated on his 
favorite field with unflagging zeal, and both remain 

R AT z E L 

Friedrich Ratzel (1844-1904) shared with Hahn not 
only a similar starting point but even one essential re- 
sult, independently arrived at — the antithesis of hoe and 
plough cultivation. Nevertheless, the totality of his work 
lies in quite a dijfferent direction. Trained at first as a 
zoologist, he soon turned to geography and came to 
occupy the chair for that subject at Leipzig. His ap- 
proach to civilization thus had a broad scientific basis, 

^*Da3 Alter . . ., 122-127. 


and his literary productivity — second only to Bastian's 
— ranged over several disciplines." 

Contrary to some of his expositors, Ratzel did not 
exaggerate the potency of physical environment. In- 
deed, he repeatedly warns against this pitfall and is 
still further removed from those geographers who see 
in climate an overshadowing determinant. What saves 
him from such naivete is recognition of the time factor: 
recent immigrants into an area cannot be so well adapted 
as natives of longer standing. Two further considera- 
tions, he contends, preclude an automatic response to 
environment: the incalculable effect of the human will; 
and man's limited inventiveness, of which Fuegian dress 
is cited as an illustration. No one could emphasize more 
than Ratzel the force of past history. At a pinch, he 
argues, early New England could be understood apart 
from the country, but never without reference to the 
Puritans who settled it. Again, he asks whether without 
Indian contact the lotus flower could have become the 
symbol of Buddhism in arid Mongolia. Ratzel also knows 
that sentimental factors deter men from exploiting avail- 
able resources and make them reject alien offerings."^^ 

Unquestionably there are lapses. Ratzel does in- 
terpret Pueblo Indian life in terms of their country 
without explaining why their immediate neighbors in 
identical circumstances enjoy a dfferent culture. But, as 
a rule, a crass environmentalism is foreign to him. In 
principle he approaches modern anthropology and the 
unexceptionable anthropogeographic position formulated 
in Brunhes' discussion of dwellings: *'If houses are far 
from wholly explicable by geography, this category of 

s^Vid. Viktor Hantzsch, " Katzel-Bibliographie, " in Fr. Eatzel, Kleine 
Schriften, 2:v-lxii, Miinchen und Berlin, 1906. 

56 F. Eatzel, Anthropogeographie, 1:13-110; 2:713, Stuttgart, 1899. 
Idem, Kleine Schriften, 2:36, 63, 407 f., Miinchen und Berlin, 1906. Idem, 
Volkerkunde, 1:91, 101, 322, 351, Leipzig und Wien, 1894. 


human phenomena cannot, at least, be perfectly under- 
stood without recourse to geography."" 

It should be noted that scholars naturally start from 
the phenomena most familiar to them. A geographer, 
approaching civilization with the concepts of the desert, 
steppe, and the ocean, asks : What can I explain by these 
elements? The ethnographer, primarily facing cultures, 
asks : How can the differences between such and such 
societies be interpreted? At times geographical condi- 
tions may fit the case ; often they are ruled out because, 
being identical, they cannot explain variations. 

Temperate on geographical causation, Ratzel was 
equally moderate in appraising biological heredity. He 
strongly believed in the unity of our species, explicitly 
repudiated Gobineau and Chamberlain, and denied to 
no branch of humanity the capacity for furthering the 
common aims of mankind. Not that he held all races 
to be equally endowed, but he rejected major differences. 
On this question he is not completely consistent. Ratzel 
is unprejudiced, but since he accepts psychological judg- 
ments far less critically than Waitz, his suggestibility 
involves him in contradiction. There is caution in the 
abstract, a neat exposition of the entanglements of Na- 
ture and Nurture, a defense of the Chinese against the 
charge of unimaginativeness. On the other hand, the 
Negro figures as a childish, imitative creature rarely 
capable of '' intellectual heights" {geistigen Hohen) ; and 
the Canadian Athabaskans are introduced as 'Vigorous, 
but poorly endowed" [wenig hegaht). Our objection is 
not that these statements are wrong, but that they rest 
on unscientific evidence. What reconciles us in a measure 
is the fairness that shines through Ratzel 's opinions. "^^ 
In this spirit he likewise treats the relations of colored 

^"^ Jean Brunhes, La geographie humaine, 111, Paris, 1912. 
^VoUerTcunde, 1:470, 671, 1914; 2:12-15, 661 f., 1895. Kleine 
Schriften, 2:127, 406, 409, 462-487, 495-497. 


peoples to white civilization, defining the aims of ''ap- 
plied ethnography" in 1900 — after Bastian, but long 
before the Africa Institute." 

Rat/.el's Volkerkunde, first issued in three volumes 
(1885, 188G, 1888) and reduced to two in the completely- 
revised second edition (1894-1895), is unquestionably a 
significant work. The lavish and excellent illustrations 
from museum collections far surpassed anything hitherto 
presented; and it gave precisely what one could not get 
from Tylor — a geographically arranged description of 
living peoples. When, some fifteen years later, Georg 
Buschan edited the next German equivalent, his ''Illus- 
trierte Volkerkunde" (1910), specialization had advanced 
so far that even this first one-volume edition required 
five collaborators. Ratzel still surveyed the whole field 
unaided, and, judged by what was then known, he offered 
a balanced picture. The higher civilizations were neces- 
sarily treated briefly, and if Africa received dispropor- 
tionate space — a whole volume in the first edition — this 
amplitude is warranted by the epoch-making discoveries 
recently made by Schweinfurth, Pogge, Stanley, and 
Nachtigal. Partly it was due to the dearth of material 
from other areas, which extenuates the skimpy treat- 
ment of America and Siberia. 

Such a survey afforded Ratzel opportunity for both 
geographical and historical reflections. Here, then, are 
some of his distinctive ideas, drawn, however, from other 
writings as well. 

As previously indicated, Ratzel did not invent the 
principle of diffusion, which Tylor and Pitt-Rivers applied 
quite as radically (page 72). But Ratzel did give special 
emphasis to the uninventiveness of mankind, though this 
idea, too, was clearly enough voiced by Tylor.^° Ratzel 

^^Kleine Schriften, 2:402-419. 

60 ' ' Old Scandinavian Civilisation among the Modern Esquimaux, ' ' 
JRAI 13:348-356, 1884. 


added the notion that the globe, so far as inhabitable, is 
actually everywhere occupied by man, whence he inferred 
far-reaching migrations dating back to a very early 
period. ' ' The earth is small ' ' and must have been again 
and again traversed by primitive groups, whence the 
constant spread and observable leveling of culture. Ratzel 
thus transfers to civilization his doctrine of the essen- 
tial biological unity of Homo sapiens. Granted these con- 
ceptions, he naturally reverses Bastian's principle that 
resemblances are merely evidence of a common mental- 
ity. Accepting psychic unity, Ratzel will have none of it 
as an interpretation of similarities. The uninventive hu- 
man beings that were constantly migrating hither and 
yon simply transported what they had picked up as their 
cultural inventory. Consistently, Ratzel dropped the re- 
quirement that diffusion can in the main be inferred only 
by a continuous or otherwise traceable distribution. Bows 
on the Kassai may be affiliated with those from New 
Guinea irrespective of whether the path of migration is 

Given this attitude, a world survey must prompt 
many specific historical hypotheses. The most remote 
Australians and Africans, we learn, have traditions that 
go back to India and Egypt ; Australian religion suggests 
decay of a higher form. South American bows are con- 
nected with those of Oceania ; the head-protectors of the 
Gilbert Islands with those from the Northeastern Siber- 
ians; the ornamental art of British Columbia — indeed, 
the Northwest of North America as a whole — with the 
Arctic regions of the Old "World, as well as with Polynesia. 
Negro Africa is related to India, Southern Asia, and New 
Guinea. Extremely characteristic are Ratzel 's ideas on 
the higher American cultures. He envisages no wholesale 
importation of Toltec, Maya, or Quechua elements by 
priestly Asiatic colonists. These complexities come not 
from a specific center in the Old World; their roots go 


back to a primeval conimiiiiity {uraltcn Gemeinschaft) 
of cultural goods, carried here and there over the earth 
in the course of many prehistoric millennia {im Laufe 
vieler vorgeschichtlicher Jahrtausende) .°^ 

This view shows typically good intuitive sense in tlie 
rejection of tlimsy attempts to derive American high cul- 
tures in toto from a particular region, and a less admir- 
able but equally typical vagueness. For what requires an 
explanation is how Mexico and Peru came by their pyra- 
mids, their developed agriculture, their bronze ; these fea- 
tures are not a common heritage of mankind and are 
known from nowhere many millennia ago. A correspond- 
ing stricture often holds against Ratzel's views. It does 
not suffice to indicate even striking resemblances; their 
history is established only when we know the several 
stages, and until then there is merely a problem that may 
or may not lead to a sound reconstruction of what hap- 
pened. A generic theory of incessant migration with po- 
tential contacts of all peoples with all other peoples is 
a meager substitute for those specific relations that have 
actually obtained. 

This vagueness may spring from one of Ratzel's most 
acceptable doctrines, the unity of human history, because 
he oddly exaggerates it into a uniformity of culture. He 
has no more sense for differences than the more naive 
parallelists, a failing he shares with later diffusionists. 
"At some period," we read, "an ingenious (sinnreiche) 
mythology was thought out and imagined: parts of it 
are found scattered everywhere ..." There follow sim- 
ilarities from Indo-Germanic, American, Polynesian, 
West African peoples. In a strange inventory of man's 
common heritage are included objects and ideas Ratzel 
perfectly well knew to have a restricted distribution, a 
fact he evidently ignored under the spell of his leading 
principle. It includes, e.g., spear-throwers and bows, 

o^Volkerlcunde, 1:38, 138 f., 352, 353, 499, 525, 533, 595-597, 668-670. 


farming, puberty rites, and shamanism. Indeed, from the 
Volkerkunde we learn that North Asiatic shamans and 
African rain makers, American medicine men and Aus- 
tralian magicians, are alike in essence, aims, and even 
part of their means. ''To speak at length about the priests 
of these [American] peoples, would be repeating with 
minor variations what has been described for Polyne- 
sians and remains to be described for Africans. "^^ 

In the abstract, of course, Ratzel knew the danger of 
equating what is only superficially alike, and he specifi- 
cally warned against it,^^ but his ethnological practice was 
little affected by this knowledge. Evidence of glaringly 
different cogency is constantly offered as equally grist 
for the mill. But without a well-developed sense for sig- 
nificant differences it is impossible to develop a sound 
typology of separate traits and still less to define culture 
areas. What we painfully miss in Ratzel, then, is precisely 
what might be expected from a geographer — sharp de- 
markation of regional boundaries. The African data, 
which he controlled best, are admittedly refractory, and 
little was known of New World cultures when he published 
the Volkerkunde; yet something better might have been 
expected than the perfunctory classification of Negro 
groups, while the segregation of at least the Pueblo tribes 
from the category of ''Forest and Prairie Indians" was 
imperative even in 1894.®* 

In one of Ratzel 's essays there occurs a curious com- 
parison of Darwin with Herder, in which the German 
classic is described as "much deeper and precisely for 
that reason less popular. . . , but also, to be sure, less 
successful in the solution of specific problems than the 
English savant." To us it seems that Ratzel 's phrase 

^"^Kleine Schriften, 2:136-138, 230 f. VdU.erlcunde, 1:54, 583, 669. 

^^Kleine Schriften, 2:519, 

^Virchow, reviewing the first edition, makes the same basic criticism, 
illustrating by Ratzel 's juxtaposition of New Foundlanders and Arauca- 
nians, Tupi and Apache (ZE, 18:291, 1886; 20:248, 1888). 


fits his own scientific character — a capacity for conceiv- 
ing eomyn'ohonsive ideas coupled with a comparative de- 
ficiency in the formuhition of definite problems. This 
statement is meant to describe, not to deny, his contribu- 
tions. He certainly supplemented the topical surveys of 
the evolutionists by his areal description, introducing 
factors naturally suggested by his training, but not at all 
obvious to the nongeographer. Thus, he properly stresses 
a people's situation on the earth's surface, correlating iso- 
lation with poverty and developing the idea of ^'marginal 
zones" (Eandlander), which has been utilized by later 
writers.^"^ His systematic regional consideration also 
threw into relief some historical views completely vindi- 
cated by later research, such as the influence of India on 
Africa or the connection of Australian with Oceanian 
puberty rites. And while many of his suggestions suffer 
from looseness, he was not uniformly averse to formu- 
lating definite problems. He did consider the resemblance 
of Congolese and Papuan bows sufficient proof of former 
contact, but he saw the value of confirmation by a study 
of the associated arrow types and other features. In other 
words, he grasped the ''quantitative criterion" as a tool 
for the comparison of complete cultural provinces. Eat- 
zel's wide reading further affords his readers informa- 
tion on a great diversity of topics. He brings out the role 
of the pastoralists in Old World history, the instability of 
primitive farming, the tendency of emigrants, illustrated 
by Scandinavian- Americans, to settle under familiar cli- 
matic conditions. 

From the foregoing remarks we should not expect an 
abundance of rigorous new concepts. Too often Ratzel 
is content with a traditional classification. Folk tales are 
mainly fragments of myth; totemism remains unana- 
lyzed ; and a barely existing or even lacking totemic sys- 

65 See e.g., F. Boas, "Die Eesultate der Jesup-Expedition, " Separat- 
Abdruck aus ICA 16:10 f., Wien, 1909. 


tern is made responsible for the animal names of Plains 
Indian military societies; the animation of all nature is 
assumed as a universal trait notwithstanding Tyler's dis- 
criminating prophylactic analysis ; matrilineal descent is 
confused with gynecocracy. On the positive side, Ratzel 
shares with Hahn the fruitful distinction between hoe- 
and plough-farming.*'*' But his chief contribution probably 
lies in certain more general ideas — the conception of hu- 
manity as a unit, the tempering of environmentalism with 
a historical perspective, the demand for a conversion of 
space into time relations, the deprecation of spectacular 
migrations in favor of slow, continuous infiltration, the 
postulation of marginal peoples. 

^^Eleine Schriften, 137 f,, Vdlkerkunde, 1:39, 113, 561, 564. Schmidt 
and Koppers, Volker und Kulturen, 390. 



Since Tyior no one has exerted on ethnology an influ- 
ence comparable to that of Franz Boas (1858- ). 
Born in Minden, Westphalia, he studied physics and geog- 
raphy at Heidelberg and Bonn, whence he followed Theo- 
bald Fischer, his major professor, to Kiel. There he took 
his doctor 's degree in 1881, his dissertation dealing with 
the color of sea water. Fischer himself had turned to 
geography from physics, and Boas acquired a natural 
scientist's control of mathematics which enabled him to 
follow the development of biometrics. His earlier writings 
include a discussion of psychophysics in a physiological 
journal and a proof of Talbot's law in the Annolen der 
Physik und Chemie. We must note, however, that Fisch- 
er's interests embraced anthropogeography. In the year 
of Boas' doctorate his teacher published a long essay de- 
scribing the role of the date palm in North African and 
Western Asiatic life.^ This paper describes aboriginal 
methods of raising the tree, of preparing dishes and 

1 Theobald Fischer, ' ' Die Dattelpalme, ihre geographische Verbreitung 
und culturhistorische Bedeutung, ' ' in Erganzungsheft No 64 zu Peter- 
mann's Mitteilungen, Gotha, 1881. 



stimulants from it ; it defines the commercial intercourse 
fostered by the spread of the palm, and fixes Arabia as 
its probable home. The treatment lends conviction to 
Fischer's own statement that he directed his pupil's in- 
terests towards ethnography.^ 

A decisive factor in determining Boas ' lif ework was 
an expedition to Bafiin Land in 1883-1884, where contact 
with the Eskimo yielded a rich ethnographic harvest. 
From 1885 to 1886 he was assistant at the Royal Ethno- 
graphic Museum in Berlin and Privatdocent at the Uni- 
versity. Thus came relations with Bastian and also with 
Virchow, the dominant spirit in the Berlin Society for 
Anthropology, Ethnology, and Prehistory. In 1886 Boas 
began those investigations of British Columbia tribes that 
engaged his attention for a long time to come. They also 
brought contacts with Tylor, then on a committee of the 
British Association for the Advancement of Science that 
promoted the exploration of that region. Three years 
later came Boas' first American position at Clark Uni- 
versity, followed from 1892 to 1894 by work on the an- 
thropological exhibits of the Chicago World's Fair. In 
1895 began a decade's connection with the American Mu- 
seum of Natural History, overlapping a lecturership that 
grew into a professorship at Columbia University, from 
which Boas retired in 1936. 

Boas' historical position is unique. He is the first an- 
thropologist who combined ample field experience with 
an unrivaled opportunity to train investigators. A. L. 
Kroeber, A. B. Lewis, F. G. Speck, R. H. Lowie, A. A. 
Goldenweiser, P. Radin, E. Sapir, F.-C. Cole, L. Spier, 
M. Herskovits, G. Herzog, A. Lesser are among those who 
took their degrees under him, but his immediate influence 
extended much further. It includes men like A. M. Tozzer, 
R. B. Dixon, C. Wissler, S. A. Barrett, J. A. Mason, J. R. 

2 See Fischer's letter to Laufer, the editor of Boas Anniversary Vol- 
ume, viii, New York, 190C. 


Swanton, H. Linton, who cither studied under him for a 
limited time or pursued lield research under his guidance. 
Still another category is made up of those who, like B. 
Laufer, P. E. Goddard, E. C. Parsons, G. liatt, T. Michel- 
son, came to Boas as mature scholars. Boas' curatorial 
duties at Berlin and New York further enlarged the scope 
of his activities, making him a practical administrator, a 
theorist on the functions of museums, an organizer of 
expeditions and of publication series. Sharing Tylor's 
abiding concern with philology, he had the added oppor- 
tunity to record and analyze the speech of many illiterate 
peoples, so that this geographer-phj^sicist grew into an 
investigator of language whose work commanded the re- 
spect of the linguistic specialist. And as there is nothing 
amateurish in his philological treatises, so he has done 
a full-fledged professional's work in physical anthropol- 
ogy, criticizing the methods of its votaries, measuring in- 
numerable Indians, investigating growth by biometrical 
techniques, independently establishing the fact that the 
stature of mixed breeds is not intermediate between that 
of the parental stocks. He contributed even to archeology 
by stratigraphic excavations in Mexico, while his ethno- 
logical work, descriptive and theoretical, is in its totality 

Such manifold achievement of high quality, coupled 
with his German origin and relationships, his frequent 
trips to Europe and regular attendance at scientific con- 
gresses, has made Boas a towering figure in international 
science. On the other hand, his influence on the world at 
large has not been commensurate with his intellectual 
stature. This is due to several circumstances. Boas is a 
man of research and has never been interested in render- 
ing truth palatable. He appeals neither to the masses nor 
to that part of the cultivated public which looks for aes- 
thetic enthrallment. He is thus essentially a writer of 
monographs, not of books. From the point of view of the 


laity, the few slender volumes that pretend to address 
the general reader are not so much books as negations of 
the idea of a book. The best known of these, The Mind of 
Primitive Man (New York, 1911), is in the main a collec- 
tion of essays previously issued in technical journals; 
one-third of Primitive Art (Oslo, 1927) is a highly tech- 
nical analysis of specimens from a single area. Indeed, 
the inspirer of dozens of anthropologists has never com- 
posed a single textbook. Finally, Boas is not linked with 
an easily condensed or soul-stirring scientific message. 
All this, set down by way of description, not criticism, 
explains why Boas has not left a deeper mark on the 
intellectual life of the world than that made by many of 
lesser stature. 

F lELD Work 

Boas must be understood, first of all, as a field 
worker. Here, too, we are struck by the all-inclusiveness 
of his interests. He himself has referred to ''bad gaps" 
in his early account of the Eskimo, but what impresses us 
as truly remarkable is the multitude and kind of detail 
this novice, schooled in another discipline, succeeded in 
recording. He noted string-figures long before the cat's 
cradle game had become a fashion in ethnographic inves- 
tigation and took down the music as well as the words of 
Eskimo songs. From such observations sprang his con- 
viction that the savage "is sensible to the beauties of 
poetry and music. " ^ As a matter of principle, Boas came 
to insist more and more on the thoroughgoing description 
of all cultural data as the sole warrantable scientific atti- 
tude. House types, basketry, social structure, beliefs, and 
tales must all be registered faithfully and with the fullest 
detail possible. 

3F. Boas, "Das Fadenspiel," AGW-M 18:85, 1888. "Poetry and 
Music of Some North American Tribes," Science, 9:383-385, 1887. 


So far there was not yet a sharp cleavage between his 
procodnro and tliat of competent predecessors. But Boas 
raised lield work to an entirely new level by demanding 
that the ethnographer's technique must equal that of a 
student of Chinese, Greek, or Islamic civilization. This 
implies some control of the aboriginal tongue, for which 
neither pidgin English nor an interpreter's rendering is 
a suitable substitute: ". . . we must insist that a com- 
mand of the language is an indispensable means of ob- 
taining accurate and thorough knowledge, because much 
information can be gained by listening to conversations 
of the natives and by taking part in their daily life, which, 
to the observer who has no command of the language, will 
remain entirel}^ inaccessible. ' ' * 

This ideal. Boas admitted, could not be fully realized 
by most ethnologists because native languages are so 
difficult that hardly any trained investigator is able to 
master them. But fortunately there are substitutes that 
approach perfection. The field worker can at least learn 
as much as time permits and, above all, he can secure 
authentic records of native thought by phonetically tran- 
scribing tales, prayers, poems, set speeches; by then 
reading them to his informants and rereading them for 
revision; and by carefully translating such documents 
with the aid of an interpreter. Thus, Boas' passion for 
texts springs from the need for material vying in docu- 
mentary value with our sources for Periclean Athens 
or the Italian Renaissance. The natives' ipsissima verba 
represent an ultimate datum of reality without the blur- 
ring screen of a free translation, which should be used 
only as a supplement to the interlinear rendering. Of 
course, an intelligent nonprofessional observer familiar 
with the language and collaborating by correspondence 
with an ethnologist may learn to interpret native life 
''from within" while simultaneously answering the spe- 

* Handboolc of American Indian Languages, 1:60 (Washington, 1911). 


cialist's queries. Boas early recognized this possibility, 
and his encouragement of James Teit, a squaw man settled 
in British Columbia, led to a series of splendid mono- 
graphs on Salish tribes. 

Another approach yields even better results. A na- 
tive who has become literate and even educated can be 
taught to write spontaneously , preferably in his vernacu- 
lar, what he knows or can gather from qualified elders 
about tribal lore. Boas accordingly not only himself re- 
corded innumerable texts in the aboriginal tongues, but 
stimulated an enormous amount of high-grade recording 
by Indians. Foremost among his earlier students was the 
part-Fox William Jones, who transcribed a superb series 
of Fox and Ojibwa texts. In continuance of this work, 
the use of a syllabary by the Fox Indians subsequently 
enabled Dr. Truman Michelson to amass a tremendous 
wealth of material sent in by tribesmen. In recent years 
Boas has trained Miss Ella Deloria to take down Teton 
stories among her people and to furnish them with a set 
of linguistic, stylistic, and ethnographic notes that make 
her work a classic of descriptive literature. The Nez 
Perce myths dictated by an old native woman to her 
college-bred son form another notable instance.^ 

Material of this sort has the immeasurable advantage 
of trustworthiness, authentically revealing precisely the 
elusive intimate thoughts and sentiments of the native, 
who spontaneously reveals himself in these outpourings. 
Boas aims at ascertaining the true inwardness of aborig- 
inal life, not by the uncontrollable intuitions of romantic 
outsiders but by objective documentation. Better than 
by answering direct questions, a blue-blood in the caste- 
ridden Kwakiutl society disfjlays his attitude toward the 

5 William Jones, "Fox Texts," AES-P 1, New York, 1907; idem, 
''Ojibwa Texts," AES-P 7, New York, 1917, 1919. Ella Deloria, "Da- 
kota Texts," AES-P 14, New York, 1932. Archie Phinney, "Nez Perc6 
Texts," CU-CA 25, New York, 1934. 


successful upstart iu Boas ' dramatic tale of the parvenu's 
rise and uKiniate humiliation." 

Again, how do aboriginal artists approach and solve 
their problems? Dozens of scholars had been working on 
museum specimens when Boas started this totally new 
lead. He attacked the problem by getting James Teit, 
long resident among the Salish Indians, to interview 
dozens of native basket-makers about the details of their 
artistic careers, the effect of home training or alien con- 
tacts, their judgment of the efforts of others. The same 
urge to see aboriginal mentality in all its phases has made 
Boas encourage work by trained women. Since primitive 
peoples often draw a sharp line between the sexes socially, 
a male observer is automatically shut out from the native 
wife's or mother's activities. A woman anthropologist, 
on the other hand, may naturally share in feminine oc- 
cupations that would expose a man to ridicule. Women 
have made important contributions independently of 
Boas, but probably nowhere have they achieved so much 
work as under the stimulation of the Columbia atmos- 
p]iere — witness the publications of Drs. Elsie Clews Par- 
sons, Ruth Benedict, Ruth Bunzel, Gladys Reichard, Erna 
Gunther, Margaret Mead, Gene Weltfish, Ruth Underhill. 

To consider another topic. Boas translated his accept- 
ance of Galton's individual differences into field practice. 
The scientist dealing with culture must exclude nothing on 
sentimental or aesthetic grounds. He is interested in the 
speculations of a native metaphysician but not a whit less 
in the gross drolls of a primitive raconteur. Only if we 
know the whole gamut of individual responses to the so- 
cial setting — the varying mediocre participants as well 
as the creative leaders — can we understand that complex 
entity we call a tribal culture."^ This point of view, among 

8F. Boas, "Ethnology of the Kwakiutl," BAE-R 35:1104 sq., Wash- 
ington, 1921. 

7 F. Boas, "The Ethnological Significance of Esoteric Doctrines," 
Science, 16:872 sq., 1902. The gro-ndng interest in individual psychology 


other things, justifies and even requires transcription of 
several variants of important myths. 

Another technique is that of securing the reminis- 
cences of informants, not as a substitute for the cus- 
tomary methods but as a valuable supplement. An 
autobiography fills gaps in our information by naturally 
bringing out facts no inquirer would dream of asking 
about; notably it elucidates personal relations and sub- 
jective responses to the cultural setting, enlarging once 
more our knowledge of primitive individuals in relation 
to their society. To be sure, remarkable attempts in this 
line have emanated from other schools. Mrs. Gudmund 
Hatt (then Miss Demant), for example, induced a Lapp 
to write out his recollections, which she subsequently 
translated into Danish, whence English and German edi- 
tions have been issued.^ It is also reported that nowadays 
Russian ethnographers are extensively applying this 
method. But probably nowhere has so much material of 
this type been systematically collected as in America, 
under the direct or indirect stimulus of Boas. No one has 
applied the technique with more zest or skill than Radin, 
while to Truman Michelson we owe illuminating autobi- 
ographies of Indian women.^ In the same category, though 
not identical in scope, belongs the Tewa Indian's diary 
kept at Dr. Elsie Clews Parsons' suggestion. As she 
points out in her introduction, Pueblo data, overabundant 
in some respects, had suffered from one serious defi- 
ciency, a view of the culture ''from within," whence the 
restriction of "our insight into the ideas, feelings, and 

among writers of all schools is brilliantly exemplified in Hilde Thurnwald's 
MenscJien der Siidsee, Stuttgart, 1937. 

^ Emilie Demant, Das Buch des Lappen Johan Turi, Frankfurt am 
Main, 1912. 

^ Paul Radin, Crashing Thunder, New York, 1927 (originally pub- 
lished as "The Autobiography of a Winnebago Indian," UC-PAAE 
16:381-473). Truman Michelson, "The Autobiography of a Fox Indian 
Woman," BAE-R: 40: 295-349, 1925; idem, "The Narrative of a Southern 
Cheyenne Woman," SI-MC 87, No. 5, Washington, 1932. 


volitions of the people, into those collective representa- 
tions, as Lev^'-Bruhl has called them, which are a sip:nifi- 
eant factor in social behavior. Therefore this journal is 
peculiarly welcome; it fills psychological interstices."^" 
We can imagine how Tylor would have rejoiced over 
material of the caliber garnered by the Boas school. Per- 
haps he would not have preferred to ignore the emotional 
phases of supernaturalism if he had enjoj^ed access to 
adequate collections of native prayers and visionary ac- 
counts. A scholar is dependent for his interpretations on 
the available techniques, the facts made known by them. 
As refined instruments of precision open new realms to 
the astronomer and physicist, so every enlargement of 
knowledge about man in society brings novel insights to 
the investigator of culture. 

Views on Mental Pbocesses in Relation 
TO Culture 

Let us, then, turn to Boas' general views. In his 
estimate of races he has commonly been cited as an egali- 
tarian. This, however, is an error of undiscriminating 
readers. He explicitly rejects the dogma that "there are 
no differences in the mental make-up of the Negro race 
and of other races . . . On the contrary, if there is any 
meaning in correlation of anatomical structure and phys- 
iological function, we must expect that differences exist. ' ' 
What Boas insists on is that certain differences have been 
alleged without evidence and, further, that whatever dif- 
ferences may ultimately be determined are minor, hence 
do not * ' unfit an individual of the Negro race to take his 
part in modern civilization. ' ' " For obvious reasons Boas ' 
championship of the simpler peoples and his exposure of 
race charlatanism or immature biological arguments have 

"Elsie Clews Parsons, "A Pueblo Indian Journal," AAA-M 32:6, 

n The Mind of Primitive Man, 271 f. 


attracted wider attention than other aspects of his work. 
While sympathizing with his position, we cannot, how- 
ever, attach importance to it for an estimate of his 
achievement, because it represents no vitally new idea. 
A critical attitude towards the ''evidence" for inequality 
had been repeatedly anticipated, notably by Waitz. Boas 
must be credited, however, with bringing the argument 
up to date; and, it is necessary to repeat, the extrava- 
gances of his opponents do not betray him into equally 
dogmatic egalitarianism. 

Boas' originality appears when he elucidates primi- 
tive mentality, especially as compared with that of civil- 
ized man. Expounded at length, his views on the subject 
might have gained him a larger following. As it is, he has 
in the main uttered his principles aphoristically, with a 
minimum of illustration. The student who has sat under 
him can supply the missing instances, but the layman is 
hard put to it in trying to invest these ideas with positive 

In order to appreciate the advance due to Boas, let 
us once more revert to Tylor. In Tylor's treatment sav- 
ages are essentially intellectuals grappling with their 
problems under the handicap of limited information. He 
describes games as largely "only sportive imitations of 
the serious business of life"; and while admitting the 
existence of ecstatic conditions accompanying supernat- 
uralism, he explicitly restricted his survey to ''the intel- 
lectual rather than the emotional side of religion. ' ' ^^ 
Moreover, not even Tylor wholly rose above the smug 
conviction that Western nineteenth-century civilization 
was the only conceivable goal of social development. 

Tarde, we noted, had transcended evolutionary com- 
placency and had also attained a proper sense of human 
irrationalism. Boas shares both points of view and sup- 
ports them with the amplitude of his wider ethnographic 

12 E. B. Tylor, Primitive Culture, 1:72; 2:339. 


kiiowlcdiio, indicatinii: and solving a host of special prob- 
lems by the way. To begin with his approach to Western 
civilization, he writes: *'In no case is it more difiScult to 
lay aside the 'Culturbrille' — to use von den Steinen's 
apt term — than in viewing our own culture. "^^ That, 
however, should be precisely our supreme aim — not 
merely to see other peoples from their own angle but to 
see ourselves as others see us. This "non-Euclidean" ap- 
proach is immanent, if often latent, in Boas' work. Par- 
ticularly he exposes the popular fallacy that the individual 
civilized person is more rational in his processes than 
the savage; actually both accept traditional judgments 
uncritically, our conclusions being better founded only 
because the tradition of our civilization has become pro- 
gressively more scientific." 

The ''irrationalism" of simpler cultures consists 
largely in their classifying concepts differently from our- 
selves and especially in their curious association of — to 
us — disparate phases of mental activity. With them fash- 
ion and ethics, myth and history, music and poetry fuse 
in a manner strange and in part intolerable to us ; ritual, 
in particular, pervades all their social activities, entering 
innumerable queer combinations. 

Such linkages, Boas contends, as now confront us in 
primitive life are for the most part not primary realities 
but secondary associations which constitute an extremely 
important anthropological phenomenon. What is more, 
in the majority of cases these associations involve second- 
ary interpretations — rationalizations in current psychi- 
atric parlance — of customs or thoughts that arose 
independently and in the main unconsciously. To take a 
hypothetical case, a people's economic life may prevent 
simultaneous eating of venison and seal flesh, but when 

13 F. Boas, "The History of Anthropology," in Science, u.s, 20:517, 

" The Mind of Primitive Man, 204-206. 


an individual breach or a contact with an alien group 
teaches the hitherto unconceived possibility of transgres- 
sion, the rule rises into consciousness, whence the need 
for justifying it. Thus, a matter-of-fact usage is sanc- 
tioned ex post facto by virtue of some supernatural de- 

Boas himself applied this principle to the field of art. 
As noted, his predecessors accepted the names given to 
decorative designs as proof of a former effort to portray 
the species or object designated (page 93). Boas asks: 
What right have we to assume a primary connection? 
Possibly the geometric forms arose automatically from 
technical processes, as Holmes suggested, or from a 
craftsman's craving to play with his technique — another 
recognition of nonrational factors. On this assumption, 
the name or symbolic meaning may be simply a secondary 
feature, an afterthought which bars conventionalization. 
But how can we prove the reality of this process? 
The most satisfactory attempt is probably the discussion 
of Eskimo needle cases.^*' A survey of available specimens 
demonstrates a fixed nonrealistic type of needle case, 
with flanges and tiny knobs, as the model floating before 
the craftsman's mind when he sets out to carve the con- 
tainer. A needle case as a whole is never conceived as the 
image of an animal, yet individual pieces show the minia- 
ture knob of the ''standard" form transmuted into a 
seal's head, while sporadically the flanges turn into wal- 
rus heads with tusks, or into lemmings. The artists cer- 
tainly did not severally try to create walruses, lemmings, 
and again seals, with the miraculous result that these 
diverse forms were all conventionalized into the identical 
traditional tube with its flanges and knobs. Rather must 
the geometrical portions of the implement have stimu- 
li /Md., 197-243. 

18 " Decorative Designs of Alaskan Needle-cases: A Study in the His- 
tory of Conventional Designs," USNM-P 34:321-344, 1908. 


lilted tlie artiticer into altering them realistically as mo- 
mentary fancy suggested. 

Corresponding research by Kroeber and Wissler 
under Boas' direction yielded similar results among the 
Plains Indians. Kroeber found that the Arapaho assigned 
ten distinct interpretations to a rhomboid. Wissler 's male 
informants gave a military significance to designs em- 
broidered by their wives, who associated them with quite 
different ideas. ^' Here, secondary reading in of interpre- 
tations is an established fact. 

Just as sjTubol and decoration join as an after- 
thought, so do ritual and myth. Here, once more. Boas' 
theory grows out of early field experience. The same cere- 
monial, he observed, had spread over British Columbia, 
but each tribe had a different explanation of its origin. 
Evidently, the rituals are not dramatizations of the aetio- 
logical myths, which on the contrary are merely rational- 
izations sanctioning the performance. The primacy of 
ritual, though not universal, has since been abundantly 
confirmed in other areas. Further, complex ceremonials 
prove to represent no basic unity, being welded together 
of historically diverse elements.^* 

In mythology it had been fashionable to interpret 
the plots in cosmic terms — as the adventures of solar or 
lunar heroes. With the advent of Boas' principles these 
facile explanations dissolved into nothing. A given tradi- 
tion turns out to be primarily a story which may some- 
times acquire a cosmic flavor by being ascribed to the 
sun or moon. The association is proved secondary when- 
ever the same plot is linked with totally distinct heroes, 

17 A. L. Kroeber, "The Arapaho," AMNH-B 18:144, 1902. Clark 
Wissler, "Decorative Art of the Sioux Indians," ibid., 273, 1904. F. Boas, 
Primitive Art, 88-143, 1927. 

1' F. Boas, "Die Entwicklung der Geheimbiinde der Kwakiutl-In- 
dianer," Festschrift fiir Adolf Bastia7i, 441, Berlin, 1896. Cf. Paul Radin, 
"The Ritual and Significance of the Winnebago Medicine Dance," JAFL 
24:149-208, 1911. Leslie Spier, "The Sun Dance of the Plains Indians: 
Development and Diffusion," AMNH-AP 16:451-527, 1921. 


so that in at least all variants but one the association 
could not possibly be primary and on analogy became 
suspect for the residual case. Here, also, the phenomenon 
hitherto naively accepted as a unit assumed a quite dif- 
ferent appearance. Any particular variant was analyz- 
able into episodes borrowed from hither and yon. A 
direct interpretation in, say, solar terms is not permis- 
sible because it would be quite arbitrary to single out 
as the original form that version which happens to refer 
to the sun. As Boas vindicated the potential primacy of 
ornament, so he showed that the plot — the equivalent of 
our novelette — could well be the basic phenomenon which 
might or might not later unite with an explanatory ele- 

In the same spirit Boas approached totemism, on 
which befuddlement had reached a maximum — notwith- 
standing Tylor 's admirable words of warning. Totemism, 
too, appears as an artificial unit ; the catchword has been 
applied to diverse phenomena presenting superficial anal- 
ogies. In reality, these several associations have neither 
a single psychological nor a single historical origin. Boas ' 
ideas were elaborated by Goldenweiser with some individ- 
ual additions.^** 

Some scholars felt the dissolution of traditional con- 
cepts as a loss; Boas' procedure seemed mere criticism, 
not constructive work. Nothing could be further from the 
truth. When a design and its name, a folk tale and its 
cosmic hero, a ritual and its origin myth, a clan and its 
totemic designation, are once recognized as spurious 
units, the sham problem of a generalized origin for any 
of these phenomena vanishes; the facts are seen in a 

1^ F. Boas, Indianifsche Sagen von der Nord-PacifiscJien Kiiste, Ber- 
lin, 1895. T. T. Waterman, "The Explanatory Element in North Ameri- 
can Mytholo/ry," JAFL 27:1-54, 1914. 

20 F. Boas, "The Origin of Totemism," AA 18:319-326, 1916. A. A. 
Goldenweiser, "Totemism, an Analytic Study," JAFL, 1910, reprinted 
in his History, Psychology and Culture, 213-332, New York, 1933. 


more natural arrangement ; and a series of legitimate spe- 
cial problems is unfolded. The question how "Totemism" 
evolved is recognized as nonsensical ; instead we are led 
to ask how and why such and such clans bear animal 
names, why they are linked here with magical rites, there 
with artistic carvings, and so forth. The intellectual lib- 
eration due to Boas is the same that we gain whenever 
science substitutes a sound phenomenalism for an arid 
conceptual realism based on premature classification. 


Analysis, however, has for its complement the pos- 
tulate that cultures are not mere aggregates of separate 
elements but integrated wholes. In his functionalist ap- 
proach, Boas — though, of course later than Bachof en and 
Fustel de Coulanges — antedates others by decades. As 
early as 1887 we find him warring against the curatorial 
practice of synoptic museum exhibits, because if a speci- 
men is isolated ''we cannot understand its meaning." A 
rattle, for instance, may be a musical instrument or an 
implement of ritual; two objects identical in outward ap- 
pearance may thus have vastly different connotations. 
**The art and characteristic style of a people can be under- 
stood only by studjdng its productions as a whole." To 
this conception Boas has steadfastly adhered in principle, 
and it formed the central theme of a treatise by one of his 
favorite pupils.^^ 

In registering Boas' functionalism we are simply 
chronicling plain facts, not expressing unqualified assent. 
Doubtless Boas' warning was timely and served a useful 
purpose, but the doctrine lends itself to exaggeration in 
the hands of less cautious followers. Certainly many of 
the cohering elements in the life of a people are not chance 

21 F. Boas, * ' The Occurrence of Similar Inventions in Areas Widely 
Apart," Science, 9:485 f., 1887; idem, "Museums of Ethnology and 
Their Classification," ibid., 587-589. Herman K. Haeberlin, "The Idea of 
Fertilization in the Culture of the Pueblo Indians," AAA-M 3:1-55, 1916. 


concomitants, but it has never been proved that all the 
traits are linked, and the attempts hitherto made to dem- 
onstrate integration have never gone beyond showing 
that, say, art is connected with supernaturalism, or eco- 
nomic pursuits with magical ideas. Further, how can we 
ever attain knowledge of the underlying quintessence of a 
culture? Evidently only from an intensive study of its 
elements; there is no royal road unless divination is to 
supersede sound scientific analysis. Finally, while it is 
true that no isolated fact is significant, it does not follow 
that the only vital correlations lie within the supposedly 
closed system of a particular culture. These fairly obvious 
reflections, directed not against Boas but against distor- 
tions of a relatively valid position, vindicate studies which 
are anathema to the militant functionalist of today. A 
rattle may, indeed, be invested with all sorts of meanings, 
but there remains a reality describable under that head; 
and this can be profitably, though not exhaustively, in- 
vestigated from a technological and distributional angle. 
Otherwise we are arbitrarily restricting the scope of our 
investigations. This is obviously Eadin's intention when 
he prescribes for study only such part of the data ''as 
bears directly upon the culture as a whole. ' ' "^ But if 
ethnology is the science of culture, it cannot rest content 
with this approach. The tribal life then appears as merely 
a segment, arbitrarily delimited for convenience' sake, 
of human culture, and correlations are permissible in all 
directions, with associated intratribal traits, with neigh- 
boring cultures, with physical environment. 

To return to Boas, his functionalist position must be 
understood in relation to the ethnological practices that 
evoked it. Many of the lesser anthropologists were at 
bottom antiquarians who collected curious oddments of 
custom or belief, placing them in convenient pigeonholes 

22 Paul Radin, The Method and Theory of Ethnology, 27, New York, 



according to a rule-of-thumb classification. Frequently, 
disparate phenomena were thus brought together — as in 
the case of totemism — because of a superficial resem- 
blance. Refusing to follow the path of least resistance, 
Boas, with a keen sense for differences and for the com- 
plexity of social life, was able to distinguish like and un- 
like features. He insisted that before equating phenomena 
we must first be sure of their comparability, which could 
be determined only from their context. Specifically, he 
insisted that we must not group together peoples on the 
strength of similar behavior but that the associated sen- 
timents are an essential part of any phenomenon to be 
studied. Thus, the sacrifice of a child on behalf of one's 
community cannot be properly set down with other in- 
stances of "murder," but with forms of self-abnegation." 
This is evidently very different from totalitarian mysti- 

Geography and History 

The quality of Boas' mind appears nowhere more 
clearly than in his attitude towards geography and his- 

Starting as a geographer, he was disillusioned by his 
Eskimo experience as to the potency of physical environ- 
ment, to which he has since then ascribed a preponder- 
antly limiting rather than creative importance. He points 
out the association of vastly different cultures with the 
identical environment and, like Eatzel, but more con- 
sistently, the overlaying of geographical by historical 
factors. In one of his very early papers he contrasts two 
scientific tendencies: the physicist tries to reduce reality 
to simple elements, while the cosmographer dispenses 
with generalization, seeking to comprehend a complex 
phenomenon as a whole. Both tendencies are equally 

23 F. Boas, The Mind of Primitive Man, 192; idem, "The Limitations 
of the Comparative Method of Antliropology, " Science, 4:901-908, 1896. 


legitimate, but with the intricacies of geography — 
and, by extension, ethnology — the latter is the more 
fruitful procedure. To quote Boas: "I aligned myself 
with those who are motivated by the affective appeal 
of a phenomenon that impresses us as a unit, al- 
though its elements may be irreducible to a common 
cause. ' ' Because of the complexity of cultural data. Boas 
deprecates the quest of laws : ' ' Cultural phenomena are 
of such complexity that it seems . . . doubtful whether 
valid cultural laws can be found. ' ' Such laws as may be 
propounded will be ''necessarily vague and ... so self- 
evident that they are of little help to a real understand- 
ing. ' ' Tylor, we noted, also recognized the complication of 
events in most of cultural phenomena, but he reserved an 
undelimited residue to which general laws were appli- 

From Boas' point of view, however, a cultural phe- 
nomenon is intelligible only from its past ; and because of 
the complexity of that past, chronological generaliza- 
tions, like those of physics, are as impracticable as are 
timeless generalizations. Paternal descent, for example, 
need not spring out of maternal descent, as Morgan and 
McLennan assumed. Stimulated by Boas' fresh outlook 
on this question, which apparently had been settled once 
and for all, Swanton undertook an investigation that 
culminated in a rejection of the traditional sequence and 
remains a landmark in the study of American Indian 
societies.'^ He proved beyond a reasonable doubt that in 
North America the clanless condition preceded unilateral 
descent. Methodologically, he refused to accept the posi- 
tion frequently held as axiomatic that Australian phe- 

2* F. Boas, * ' The Aims of Anthropological Eesearch, ' ' Science, 
76:611, 1932; idem, "History and Science in Anthropology," AA 38:137, 
1936; idem, "The Study of Geography," Science, 9:137-141, 1887. E. B. 
Tylor, Researches, 3 f. 

25 John R. Swanton, "The Social Organization of American Tribes," 
AA 7:663-673, 1905; idem, "A Reconstruction of the Theory of Social 
Organization," Boas Anniversary Volume, 166-178, 1906. 


nomona are the rudest extant and that the succession of 
events in Australia represents the norm for all human 
groups. In insisting that first of all "each region must be 
worked out by itself so far as practicable," Swanton 
voiced an essential feature of Boas' programme of study. 
This sense of diversity separates Boas on principle from 
the parallelists. A closely connected point of his is the 
reality of convergence : in culture not only like causes but 
also unlike causes produce like effects. **It is of very rare 
occurrence that the existence of like causes for similar 
inventions can be proved, as the elements affecting the 
human mind are so complicated ; and their influence is so 
utterly unkno\\Ti, that an attempt to find like causes must 
fail, or will be a vague hypothesis. " ''' 

Boas' historical outlook must be considered in the 
light of this overpowering sense of cultural diversity, 
for it explains his reluctance to accept the generalized 
pictures of extreme diffusionists as well as his repudiation 
of parallelist schemes. For these diffusionists resolve 
culture history into the interaction of a very few ultimate 
culture complexes. Surveying the whole span of human 
existence. Boas sees no warrant for such simplicity of 
formulation. Each group has its own unique history, due 
partly to inner causes, partly to extraneous influences; 
and these differentiations must date back to an extreme 
antiquity. If, for instance, Elliot Smith interprets prac- 
tically all arts and customs of savages as decadent relics 
of Egyptian civilization that spread as complexes with 
the rise of navigation. Boas objects that while single 
detached elements may persist indefinitely, ''the coherent 
survival of cultural features that are not organically con- 
nected is exceedingly rare." Thus, in the course of cen- 
turies Elliot Smith's complexes would inevitably be torn 

26 F. Boas, ' ' The Occurrence of Similar Inventions in Areas Widely 
Apart," Science, 9:485 f., 1887; idem, "The History of Anthropology," 
ibid., 20:513-524, 1904. 


asunder and rearranged differently in whatever region 
they might reach." 

As a result of this attitude Boas does not reconstruct 
history on a major scale. Probably more familiar than any 
of his contemporaries with the ethnography of the world, 
he has never traced the sequences of culture history as a 
whole ; and even for America he has merely sketched the 
course of development with utmost brevity in an after- 
dinner speech to which he himself attaches no impor- 
tance.^^ Given a diversification of culture myriads of 
years ago and an instability working differently on every 
group because of specific conditions and specific contacts, 
precise reconstruction is possible only after the most in- 
tensive investigation. Hence no one scholar can establish 
sequences for more than a minute fraction of the total 
number of peoples ; and Boas prefers connecting North- 
eastern Siberia with British Columbia to formulating a 
scheme for all of the New World, let alone for both hemi- 
spheres. His restraint is partly due to his rejection of 
certain principles applied by others in reconstructing the 
past. He does not believe that the area of distribution is 
proportionate to the antiquity of traits ; nor does he con- 
sider the area of greatest intensity to indicate the original 
focus of dissemination. 

Boas deprecates the criticism that such abstention 
from historic synthesis implies indifference to ''the ulti- 
mate problems of a philosophic history of human civiliza- 
tion. ' ' He doubtless feels that a satisfactory synthesis may 
some time emerge from a sufficient number of intensive 
regional studies. In the meantime his approach offers 
two compensations — a rigorous demonstration of his- 

27 F. Boas, Primitive Art, 6 f . ; idem, ' ' The Methods of Ethnology, ' ' 
A A 22:311-321, 1920; idem, "Evolution or Diffusion?" AA 26:340-344, 
1924; idem, "The Social Organization of the Tribes of the North Pacific 
Coast," A A 26:323-332, 1924. 

-*"The History of the American Race," Annals N. ¥. Academy of 
Sciences, 21:177-183, 1912. 


torieal relationships and new light on the processes in- 
volved in such contacts. 

Boas' approach to concrete historical problems is 
best exemplified by his mythological investigations and by 
the work of the Jesup North Pacific Expedition. 

As repeatedly pointed out, there was never a period 
in which diffusion was generally denied. But in mythology 
parallelism was carried rather far by Brinton, who ex- 
plained resemblances even between neighboring tribes on 
the principle of psychic unity. Boas's early work in 
British Columbia completely demolished this interpreta- 
tion. If Brinton 's position were tenable, proximity would 
be a negligible factor, hence the most remote variants of 
a cycle would not differ any more than those among con- 
tiguous tribes. Actually, the Raven myth turned out to 
be most elaborate in the north and to taper away pro- 
gressively as one proceeds southward. Hence, the plot was 
not the independent product of a common psychology but 
had been evolved in one center and transmitted from 
tribe to tribe. In other words, the combination of adven- 
tures found in the subdivisions of this area is a function 
of geographic position, not of a common mentality. The 
beauty of this logic has never been surpassed either by 
Boas himself or anyone else. What impresses us is the 
conversion of barren abstract dispute about parallelism 
versus diffusion into a scientific problem that admits of 
definitive solution.^ 

It is the capacity for defining problems so as to pre- 
pare a clear-cut solution that distinguishes Boas from 
Ratzel. Not content with general migratory movements of 
peoples and ideas that would lead to an indiscriminate 
international communism of culture, he is interested less 
in the possibility than in the demonstrability of contact. 

» Daniel G. Brinton, The Myths of the New World, 172 f., Philadel- 
phia, 1868. F. Boas, Indianische Sagen von der nord-pacifisohen Kuste 
Amerikas, 329 sq., 1895. 


Methodologically, proof consists not in indicating isolated 
resemblances, but in tracing a number of complex cultural 
phenomena over a continuous territory beyond which they 
are lacking or fragmentary. Only cogent evidence war- 
rants assuming a break in geographic continuity. Thus, 
Boas rejects Ratzel's theory of an Oceanian- Aonerican 
connection not as inconceivable but as undemonstrated, 
in the same spirit refusing to ally certain mythological 
incidents of Oceanian and American mythology: "We 
desire to find uncontestable evidence of transmission, not 
alone the possibility or plausibility of transmission ; and 
for this purpose our safeguards [complexity of the traits 
compared and continuous distribution] must be insisted 

What, then, are the results based on these premises 
and the investigations of the Jesup Expedition? The 
tribes of Northeastern America are proved to have had 
at one time intimate relations with the Paleo-Asiatics. 
Especially is there a parallelism of complex folktales 
between East Siberians and the Indians of British Colum- 
bia, while no such similarities ally the East Siberians with 
the Eskimo, who now occupy an intermediate position. 
Hence, Boas infers, the Eskimo, coming from the east, 
must have driven a wedge between these two groups which 
once formed a continuous major block of population. This 
comparatively recent contiguity of the Eskimo and the 
East Siberians explains their sharing many features of 
economic life. True, the Eskimo did not take over the 
Asiatic idea of domesticating reindeer, but that is because 
the tribes near Bering Strait had themselves only ac- 
quired reindeer a few centuries ago. Correlating all the 
findings with those from other sections of the New World, 
Boas lays down in broad terms the history of American 

■'"' F. Boas, * ' Die Eesultate der Jesup-Expedition, ' ' ICA-P, Separat- 
Abdruck, 6 f., Wien, 1909. Idem, "Mythology and Folk Tales of the 
North American Indians," JAF 27:381 sq., 1914. 


cultures. Basically there are only two areas : the marginal 
tribes of the Arctic, Northwest, and California, repre- 
senting the primeval hunting culture in its several local 
variants; and the Middle American farming civilizations 
from Peru to Southern Merico. The remaining cultures 
are due to ^Middle American influences blending with the 
archaic features in varying degree. 

These are historical inferences of no mean import, 
and they have remained fundamental to Americanist 

Characteristically Boas was not content with his 
theory of Eskimo migration, but proposed a crucial ex- 
periment. Eskimo skulls happen to have several highly 
distinctive features. Hence, it is proposed, we must ex- 
cavate Alaskan sites in search of calvarial remains ; and 
should older strata harbor non-Eskimo types, the in- 
trusion of the Eskimo from the outside world would be a 
demonstrated fact. 

One further result of the Jesup Expedition must be 
cited. Various tribes of the Salish family were shown to 
have become secondarily assimilated to the higher coastal 
cultures of British Columbia. Specifically, they had 
borrowed a clan system to supersede their older family 
organization. Hence, Morgan's sequence is reversed for 
this series of peoples and refuted as a general law. Again 
we note that vital conclusions are the direct outgrowth not 
of abstract reasoning but of concrete research. 

Before leaving Boas' attitude toward diffusion, we 
must note one other point. In contrast to those satisfied 
with establishing the fact of a historical connection, Boas 
regards this as merely an initial step. It is important to 
ascertain why traits were borrowed and how they were 
incorporated into the borrowing cultures. What, for 
instance, is the role of captives, male or female, in intro- 
ducing novel ideas? Why are some cultural features re- 
jected while others are readily adopted? How are bor- 


rowed elements remodeled? What innovations do they 
evoke? These are the questions Boas studies as ''the 
dynamic conditions of change, " ' ' the dynamics of primi- 
tive life." They lead to still another extension of tradi- 
tional field research; in addition to the pigeonholing of 
facts on the model of our standard monographs we need 
supplementary data ' ' on the way in which the individual 
reacts to his whole social environment, and to the differ- 
ences of opinion and of mode of action that occur in prim- 
itive society and which are the causes of far-reaching 
changes. ' ' *^ 


It is one of the most difficult tasks to expound Boas ' 
greatness to those who have failed to come into personal 
contact with him. His achievement is full of paradoxes. 
Here is a man lacking and scorning any artifice that might 
attract students, yet he has trained and influenced the 
greatest number of professional anthropologists. Here is 
a scholar who controls the ethnographic literature of the 
world as well as anyone, yet he has never summarized his 
views in a comprehensive treatise comparable to Ratzel's 
Volkerkunde. His critics suggest an incapacity for syn- 
thesis; his intimates know that he forms opinions on all 
anthropological questions but refrains from utterance 
when the evidence seems indecisive. That even the pro- 
visional syntheses of this independent and erudite thinker 
would shed floods of light is unquestionable; it is not, 
however, Boas ' method of procedure. It is still more para- 
doxical that this indefatigable collector, who has con- 
sistently preached the totalitarian view of culture, has 
produced not a single full-length monograph of any one 
tribe. Notably he has published thousands of pages on the 
Kwakiutl, but he has never integrated his data in a single 

"F. Boas, "Evolution or Diffusion," AA 26:341, 1924; idem, "The 
Methods of Ethnology," A A 22:315 sq., 1920. 


work. Here, too, he is doubtless deterred by puritanical 
motives. On the one hand, our duty is to gather the raw 
facts before they disappear; on the other, when can one 
be sure of having all the data that would warrant defini- 
tive interpretation? 

It is this deliberate aversion to systematization that 
is the despair of many readers and precipitates misunder- 
standing. The craver of systems cannot understand a 
scientist's progress from problem to problem without at 
once generalizing a particular solution achieved. For ex- 
ample, Boas once propounded the view that in British 
Columbia totemism had evolved through allowing inheri- 
tance of the individual guardian spirit. This hypothesis 
was discussed as though it were a theory of totemism as a 
whole, an interpretation indignantly rejected by Boas: 
'' ... it is entirely opposed to the methodological prin- 
ciples to which I hold to generalize from the phenomena 
found among the Kwakiutl and to interpret by its means 
all totemic phenomena. " ^^ This attitude is the scientist's 
as opposed to the philosopher's; it has been lucidly set 
forth by Ernst Mach: ''To the scientist who always de- 
tects new features in every major solution of a problem 
systematizing and schematizing always appear premature, 
and he gladly leaves it to the more practised philoso- 

More disturbing to those who share Boas' conception 
of anthropology is his failure to expound at length the 
reasons for a change of opinion. Thus, in 1909 he still 
considers it conceivable that America might have been 
peopled by the transatlantic migration of ancient Mongol- 
oids coming from Europe. In later discussions this possi- 
bility is ignored or eliminated.^* Again, his earlier 

32 F. Boas, "The Origin of Totemism," AA 18:320, 1916. 

33 Ernst Mach, ErJcenntnis und Irrtum, p. vi, Leipzig, 1906. 

34 F. Boas, "Die Kesultate der Jesup-Expedition, " 15. Idem, "The 
History of the American Race," Annals N. ¥. Academy of Sciences, 


accounts credit the British Columbians with a belief in 
possession. Nothing could be more explicit than the state- 
ment: "In the second dance the novice appears wearing a 
mask, which represents the spirit which possesses him." 
In later publications this phenomenon figures as belong- 
ing distinctively to the Old World: "On the other hand 
it seems quite foreign to the beliefs of American 
tribes. . . . The spirits may attack man, but they do not 
enter his body. " ^^ So far as I am aware, this discrepancy 
is nowhere explained. 

A more vital matter relates to contradictions in ex- ' 
planatory principles. We have already cited Boas' view 
that primitive cultures are not stable. This is unquestion- 
ably correct, but his application of it to a critique of dif- 
fusionism appears puzzling in the light of one of his own 
major conclusions. He chides the British diffusionists for 
assuming that "ancient Mediterranean customs could be 
found at the present time practically unchanged in differ- 
ent parts of the globe. ' ' ^^ Yet we have found him arguing 
for an exceedingly ancient connection {uralte Verbind- 
ung) between the Indians and the Paleo- Asiatics on the 
basis of mythological resemblances. This obviously im- 
plies that the tales have remained stable over a period of 
thousands of years. If there is a way of harmonizing these 
two positions, it has not been indicated. Finally, there is 
a curious indefiniteness with regard to areal range as a 
criterion of antiquity. On the one hand. Boas objects to 
making a general principle of this point of view "which, 
with due caution, may be applied here and there"; and 
specifically he protests against Spinden's, Wissler's, and 
Kroeber's reconstruction of American chronology on \ 

21:178, 1912. Idem, "America and the Old World," ICA 21:21, 1924. 
Idem, Scientific Monthly, 110, 1929. 

35 F. Boas, "The Tribes of the North Pacific Coast," Annual Archaeo- 
logical Eeport, 1905, 246; Toronto, 1906. Idem, "America and the Old 
World," ICA 21:27, 1924. 

«8F. Boas, "The Methods of Ethnology," AA 22:317 f., 1920. 


this assumption. Yet because songs and tales are found 
universally, they ligure in Boas' account as ''the primary 
form of literary activity" and in the same spirit he sets 
down exogamy as older than totemism. "The recognition 
of kinship groups, and with it of exogamy, is a universal 
phenomenon. Totemism is not. It is admissible to judge 
the antiquity of an ethnic phenomenon by its universality. 
The use of stone, fire, language, is exceedingly old, and 
it is now universal. On this basis it is justifiable to assume 
that exogamy also is very old. ' ' From a more recent state- 
ment we gather that: "In a few cases it seems justifiable 
to infer from the world-wide diffusion of a particular 
cultural achievement its great antiquity. This is true 
when we can prove by archaeological evidence its early 
occurrence."" However, Boas evidently does not mean 
that only archeological evidence has cogency since he 
admits the antiquity of language and song. We are thus 
left without guidance as to the applicability of the prin- 
ciple. * * Here and there ' ' and " in a few cases ' ' provides no 
acid test without further specification wherein trust- 
worthiness may lie. Eepugnance to systematized exposi- 
tion may thus be carried a bit too far. Like Tylor, then, 
when he fails to define the sphere of "general laws" as 
against that of specific history. Boas leaves us here in a 
methodological quandary. 

An objective recital of Boas' achievements is one 
thing, an appraisal involves as much personal taste as an 
artistic judgment. Those who cannot divorce ethnology 
from belles-lettres will find nothing to attract them in 
Boas, whose bald exposition never aspires to literary 
graces. But even among those who view anthropology as 
a science, responsiveness has varied. As Boas wrote in 
his obituary of Virchow: "There are but few students 

s^Zdem, Primitive Art, 5 f., 301. "The Origin of Totemism," AA 
18:323 f., 1916. "The Aims of Anthropological Research," Science, 
76:609, 1932. 


who possess that cold enthusiasm for truth that enables 
them to be always clearly conscious of the sharp line 
between attractive theory and the observation that has 
been secured by hard and earnest work. "^ 

Those ethnologists who crave bold generalizations are 
certainly doomed to disappointment. Boas' greatness 
lies not in the systematic elaboration of facts, but in his 
independent approach to that material, his novel classifi- 
cation of it, his capacity for defining problems hitherto 
undreamt of, his insistence on a methodologically rigor- 
ous solution. From that angle he stands unrivaled, and 
all his contemporaries seem shallow in comparison. 

2^ F. Boas, ' ' Kudolf Virchow 's Anthropological Work, ' ' Science, 
16:441-445, 1902. 



Historical Schools 

''History" is a term legitimately applied in several 
senses. It signifies either the course of events or its de- 
scription, some scholars restricting it to written reports. 
The historical ethnologists, however, must largely dis- 
pense with documents because they mainly deal with il- 
literate tribes whose past is at best fitfully illuminated by 
written sources. These scholars have accordingly been re- 
proached with relying on mere conjecture, but they com- 
mand a twofold defense. Archeology, so far as it can be 
used, yields more objective evidence than the biased nar- 
rative of ancient chronicles. Secondly, every historian 
synthesizes his documentary evidence by his interpreta- 
tion ; and the ethnologist may plead for a similar latitude 
provided that he uses canons of inference which ensure 
reasonable accuracy. 

Our historical schools in anthropology, however, 
must be viewed historically, i.e., with reference to the 
condition, actual or putative, that evoked them. They are 



in conscious revolt against ''evolutionism," rejecting its 
doctrines as subjective simplifications, hence distortions, 
of the real events. Actual history is too complex for such 
neat summaries as Lewis H. Morgan's. As Laufer put it, 
''Development does not take place according to the sub- 
jective classiticatory scheme of the ethnological school 
that has gone astray in evolutionist paths. " ^ In principle 
the avowed historian recognizes that development is in- 
tricate, that each people experienced a distinctive set of 
influences, specifically as a result of unique contacts with 
neighbors. To determine the nature of such intercourse, 
then, is certainly an initial and, according to Graebner, 
the basic problem.^ 

Now the first thing to note once more is that this 
doctrine was not invented when Ratzel challenged Bas- 
tian's elementary ideas, let alone when Graebner and 
Ankermann delivered their lectures on African and 
Oceanian culture circles.^ Tylor, we cannot too frequently 
insist, explicitly accepted the complexity of culture and 
repeatedly suggested how the life of peoples had been 
molded by importations. To be sure, his theory of religion 
allocated certain beliefs to definite stages in a hypotheti- 
cal scale, but various less bulky though not less important 
publications exemplify his faith in diffusion. As for Boas, 
his work in British Columbia shows above all the inter- 
play of tribal groups ; and it is precisely because of such 
influences that he rejects "simple" explanations of 
myths as though they had been conceived on the spot as a 
direct response to nature. 

Recent movements known as "diffusionist" par ex- 
cellence, then, were not the first to propound a metamor- 
phosis due to alien contacts. Nor can we easily distin- 

1 B. Laufer, Dolcumente der indischen Kunst, 31, 192 f ., Leipzig, 1913. 

2F. Graebner, Methode der Ethnolngie, 107, Heidelberg, 1911. 

2 F. Graebner, ' ' Kulturkreise und Kulturschichten in Ozeanien, ' ' ZE 
37:28 sq., 1905; B. Ankermann, "Kulturkreise und Kulturschichten in 
Afrika, " ibid., 54 sq. 


giiisli the inodorates from the radicals by the criteria 
used to establish relationships. Graebner, the methodolo- 
gist of the German school, has announced two means of 
determining cultural connection — a formal resemblance 
neither inherent in the nature of the phenomena com- 
pared nor due to geographical causes; and the chance 
association of a whole series of elements in two regions 
(''quantitative criterium"). Now these are criteria 
clearly set forth and applied in Tylor's discussion of the 
patolli game. This is not remarkable, for they have prob- 
ably been applied ever since man began to institute com- 
parisons at all. 

The extremists, however, do stress two additional 
principles. The uninventiveness of the human mind, to 
which Tylor alludes only incidentally, becomes a cardinal 
dogma. While these scholars do not deny that a dupli- 
cated invention is conceivable, they are so skeptical 
about human originality as virtually to exclude the as- 
sumption. Graebner ingeniously adds the auxiliary point 
that there are only criteria of connection, none of inde- 
pendent development; hence, at best, parallelism might 
be inferred at the very close of investigation, from the 
failure to prove transmission. The second principle, 
which dates back to Ratzel, is the irrelevance of distance 
or continuity. Not of course that even the radicals allege 
action at a distance; but unspanned remoteness is not 
felt as an obstacle in the face of similarities. As easy- 
going investigators of faunal distribution invent land- 
bridges to suit their purpose, so the diffusionists decree 
at their convenience a former continuity no longer 

So far, then, moderates like Boas differ mainly by 
their greater caution. For them the degree of human or 
racial inventiveness is as yet unknown, hence cannot be 
invoked as an ultimate principle on either side ; and they 
treat intermittent resemblances not as worthless, but as 


lacking cogency so long as the paths of communication 
remain obscure. However, we must note another major 
difference among ethnological historians that is not quite 
coterminous with the division into a right and left wing. 
Tylor is content with establishing the transmission of 
particular features; Boas connects whole cultures but 
only as a rule in adjacent regions. Ratzel abandons conti- 
guity but concentrates on single elements — types of bow 
or armor. But a disciple of Ratzel, Frobenius, going be- 
yond his master, traced to a common origin the whole 
culture of two remote areas, West Africa and Oceania.* 
Graebner, starting from a similar survey, came to work 
out a scheme for the whole of culture-history. Similarly, 
Father Schmidt, acknowledging his indebtedness, revised 
and amplified Graebner 's doctrines so as to create a 
veritably new system; and a recent attempt, elaborating 
Graebner 's and Schmidt's principles, is due to Montan- 
don. Completely independent of the Germans is the pan- 
Egyptian theory of Elliot Smith and his disciple Perry. 

There are thus at present two main schools, a 
German and a British, which purport to define the course 
of culture throughout the globe and throughout human 
existence. To that extent they supplant Morgan's doc- 
trines with a system similarly all-embracing; but avow- 
edly their procedure differs in reconstructing history 
inductively. The question is how far they have succeeded. 

Lengthy discussions of methodology are avoided at 
this point because the relevant treatises, while estimable 
contributions to the logic of science, seem ethnologically 
unimportant.' The crucial point in practice is not whether 

■*Leo Frobenius, Der Ursprung der afrikanischen Kulturen, Berlin, 

5 Cf . H. Pinard de la BouUaye, L'etude comparee des religions, 2:183- 
282, Paris, 1925. F. Graebner, Methode der Ethnologie, Heidelberg, 1911. 
Schmidt und Koppers, Volker und Kulturen, Eegensburg, 1924. E. Sapir, 
TiTne Perspective in Aboriginal American Culture; A Study in Method, 
Ottawa, 1916. W. Schmidt, Handbuch der MetJwde der Kulturhistorischen 
Ethnologie, Miinster, 1937. 


the phenomena compared should be simihir, but whether 
alleged formal resemblance actually exists. Graebner 
himself admits the subjectivity of this criterion when he 
rejects Von Luschan's alignment of Papuan headrests 
with the Ionic capital. He contends, however, that the 
difiSculty is overcome by the quantitative criterion. Yet 
obviously this is significant only if the several compari- 
sons hold water: two dozen features of dubious char- 
acter are less cogent than two beyond cavil. As Graebner 
says in another context, everything hinges on whether a 
scholar is facing natural units or artificial constructs of 
his brain. And Sapir, almost duplicating his phraseology, 
writes: "The constant danger that besets the investiga- 
tor is to make historical or psychological actualities out 
of merely conceptual abstractions. ' ' ° 

Accordingly, diffusion is best considered not in the 
abstract, but in the light of several schemes. In the 
interest of simplicity rather than of chronology we begin 
with the British school. 

Elliot Smith 

G. Elliot Smith (1871-1937), a justly distinguished 
anatomist, at some time decided that ethnology was in a 
bad way and required a savior. Being a forceful person- 
ality possessed of a vigorous style, he soon attracted a 
group of apostles. To plunge into his writings is un- 
questionably a stimulating experience. Let no one, how- 
ever, approach with the picture of British science formed 
on a reading of Faraday or Darwin. Here there is no 
humble quest of the truth, no patient scrutiny of difl&- 
culties, no attempt to understand sincere criticism. Vehe- 
ment reiteration takes the place of argument. Elliot 
Smith is more cocksure than Haeckel, more contemptuous 
of opposition than Dr. Samuel Johnson. Everything is 

8 Graebner, op. oit., 86, 118, 155. Sapir, op. cit., 38. 


grist for his mill, everything is either black or white. This 
procedure is not limited to ethnology, but there is a vital 
difference : in physical anthropology Elliot Smith controls 
the facts, hence — right or wrong — his judgments com- 
mand respect, while in ethnography his crass ignorance 
darkens counsel. 

Being at one time stationed in Cairo, this reformer 
of ethnology decreed that Egypt must be the source of all 
higher culture. Had he tarried on the Euphrates, we may 
reasonably surmise him to have fathered a pan- 
Babylonian theory. Be that as it may, his actual scheme 
rests on a few dogmas that are easily summarized : 

(1) Man is uninventive; hence culture arises only in 
exceptionally favorable circumstances, practically never 
twice independently. 

(2) Such circumstances existed only in ancient 
Egypt; hence elsewhere culture, except some of its 
simplest elements, must have spread from Egypt with 
the rise of navigation. 

(3) Civilization is naturally diluted as it spreads to 
outposts ; hence decadence has played a tremendous role 
in human history. 

Of "Natural Man," unstimulated by Egypt, we get 
an engaging picture. This idyllic creature, honest, peace- 
able, improvident, was ''almost wholly devoid of 
anything worthy of the name of culture. ' ' In 4,000 b.c. re- 
ligion and social organization, marriage and burial cere- 
monies, houses and clothes, all arts and crafts except 
those used to make hunting equipment, were lacking 
everywhere outside of Egypt and vicinity. Human be- 
ings, we learn, lived essentially like the anthropoid apes.^ 

The ancient Egyptians were favored by the growth 
of wild barley in their country, which led to its deliberate 
cultivation, the inundations of the Nile prompting the 

■^ G. Elliot Smith, In the Beginning ; the Origin of Civilization, 20-31, 
New York, 1928. 


natives to imitate its process by irrigation. Having to 
store food, the people invented pottery and granaries, 
the latter evolving into dwellings. "The leisure enjoyed 
by men who stored up food in their settled homes" was 
devoted to inventing basketry, matting, and weaving; 
and, incidentally, cattle came to be domesticated. Eeli- 
gion rose out of the embalmer's art: the king-engineer 
who controlled fate by accurately predicting the move- 
ments of the Nile was mummified and henceforth treated 
as inmiortal. The practices performed to ensure the royal 
corpse against corruption gave rise to drama and cere- 
monialism, to dancing and music, also stimulating archi- 
tecture and carpentry.* 

This account of Egyptian history can be definitively 
judged only by Egyptologists; it impinges on ethnog- 
raphy because Elliot Smith virtually denies any inde- 
pendent developments. The American Indians, according 
to him, lived like apes until the beginning of the Christian 
era; their first pyramids were erected five or six cen- 
turies later as copies of Cambodian and Javanese models, 
themselves traceable to Egyptian prototypes.^ American 
initiation ceremonials and secret societies go back to the 
mummification ritual of the Nile; Australian totemism 
and social organization are "the degraded and otherwise 
modified results of the adoption of alien [i.e., Egyptian] 
practices and beliefs."" 

To start with the beginning: What about the ape- 
like condition of man sixty centuries ago ? It is so unwar- 
ranted an assumption that at lucid intervals Elliot 
Smith's unguarded statements completely nullify this 
cardinal principle. Thus, on prehistoric evidence he 
credits the ape-like Natural Man who on one page has 
"neither arts nor crafts beyond the making of imple- 
ments of the chase" with "an aptitude for pictorial art 
and craftsmanship." Obviously the second statement di- 

8 /bid., 30-46. ^ Ibid., 83. ^o Ibid., 25, 67. 


rectly contradicts the first. As for supernaturalism, El- 
liot Smith admits that "long before the use of gold," i.e., 
long before Egyptian civilization could travel round the 
earth, men believed in imitative magic: "The teeth of 
ferocious animals were believed to protect the wearer, 
since they conferred powers of aggression upon their 
original owners." " If so, why could not the Australians 
have arrived at such conceptions independently? Yet ac- 
cording to the chief apostle of the creed it is probable 
that "prior to the coming of this [Egyptian] civilization, 
the native peoples were devoid of any magical or reli- 
gious practices or ideas. "^^ 

Obviously an Egyptian origin for any trait does not 
follow from even the most rabid insistence on human un- 
inventiveness. What does follow is simply a single center 
somewhere; the feature, however, might be of Chellean 
antiquity, a parallel signifying merely the persistence of 
an extremely ancient cultural element on both the Nile 
and the Darling. This, for instance, might explain the 
distribution of animism in time and space without re- 
course to multiple origins. 

It is true that Elliot Smith and Perry categorically 
deny an "original idea of a soul that persisted after 
death" except in association with mummies; but they 
offer no evidence for this amazing allegation.^^ 

The pan-Egyptian obsession of these writers runs 
counter to established historical facts. When two peoples 
meet there is not an automatically irreversible stream of 
culture from the higher to the lower. Europeans brought 
livestock and wagons to the American Indians, but in re- 
turn they acquired maize, squashes, potatoes, and a 
host of other plants. The Chinese were not uniformly 
donors in relation to ruder neighbors, but took over 

"Zfttd., 48. 

12 W. J. Perry, The Children of the Sun, 480, New York, 1923. 

13 Idem, Gods and Men, 69, London, 1927. 


seafaring from the Malays, felt from northern herders. 
As for the ancient Egyptians, in the nascent stage of 
their civilization, they were little superior to surround- 
ing tribes ; hence contact with them certainly implied an 
exchange rather than one-sided benefactions. 

Yet the extremists constantly argue that: (a) some 
tribe has demonstrably borrowed a particular trait; 
therefore (b) its inventory of crafts and customs must 
come from the same source. The logical fallacy is patent. 
Admitting that the Mexican pyramids are Indo-Chinese 
and that the patolli game is Hindu, this would prove 
nothing more than contact of America with Asia; it 
would imply that other traits might have been borrowed 
by the New World, not that they actually were. In the 
light of the instances cited, loans in the opposite direc- 
tion are wholly plausible; indeed, reputable botanists 
believe that the sweet potato was carried by the Poly- 
nesians from South America. The reverse conclusion 
does not follow from contact, but only from the dogma 
that the aborigines of America, being chimpanzee-like, 
had nothing to give." 

This gross fallacy pervades the entire treatment of 
the Western hemisphere. Perry proclaims: ''It is well- 
known . . . that the agricultural tribes of the United 
States owe their customs to Mexico and the neighboring 
countries. The proof is simple and conclusive. All these 
tribes cultivate maize, and other southern plants, such 
as squashes and gourds. Maize is indigenous in Central 
America, so its cultivation must have been propagated 
thence in all directions." ^^ The preposterous inference is 
simply an a priori deduction from the principles already 
expounded. American agriculture comes from Mexico, 
hence everything in the New World comes from Mexico ; 
but agriculture came to Mexico from the outposts of 

" Elliot Smith, In the Beginning, 29. 
15 Perry, Gods and Man, 18. 


Egypt ; hence ultimately Egypt is the universal fountain- 
head. As a matter of fact, Americanists increasingly 
doubt that Mexico is the center of aboriginal husbandry. 
It now appears wholly plausible (a) that maize may have 
been raised first in South America; (b) that another 
staple, such as manioc, may have preceded maize in the 
New World; (c) that American tillage has several inde- 
pendent foci.^^ The speculations of the British diffusion- 
ists collapse on any of these postulates. No doubt they 
will deny the last and assert that the others involve only 
minor details: the essential fact for them is ''agricul- 
ture," and agriculture, no matter where or how it orig- 
inated in America, must have come from abroad. This 
is, of course, again a priori reasoning, but even apart 
from that it reveals a further cardinal error. 

To the diffusionists * ' agriculture "is an ultimate ir- 
reducible verity; to unprejudiced minds the term illus- 
trates an artificial construct of the scholastic classifier. 
There is more resemblance between the Ionic capital and 
a Papuan headrest than between the sowing of cereals 
and the planting of a banana shoot ; and when cultivation 
implies the laborious extraction of poison, we are again 
dealing with quite a distinct matter. For some purposes 
a common label may be convenient, just as at times we 
may speak of ''keeping animals." But such classification 
does not prove an underlying common reality; bee- 
keeping is not the same as training elephants or herding 
horses ; and sowing seeds is not equivalent to planting a 
side-shoot or a tuber, let alone ridding a tuber of its 
prussic acid. If, then, bitter manioc should prove the old- 
est cultivated species in the New World, it could not be 
derived from alien forms of "cultivation" that share 
with it nothing but the arbitrarily assigned name. 

18 A. V. Kidder, "Speculations on New World Prehistory," ALK, 
150 f., 1936. Carl Sauer, "American Agricultural Origins: A Considera- 
tion of Nature and Culture," ibid., 291 f. J. Eric Thompson, Archaeology 
of South America, 16, Chicago, 1936. 


Elliot Smith and Perry misapply the form criterion 
still more gravely when they trace a ''dual organization" 
over the whole globe and back to Egypt. They actually 
equate the division of a society into intermarrying halves 
with any stressing of the number "two": the halving of 
a village community by a street running north and 
south is evidence of a "dual organization";" and so is 
the very division into opposing sides at games. How 
competitive games were to be played without such ar- 
rangements is unexplained. Here their claim for a single 
historical phenomenon can be directly refuted. Whenever 
one of three exogamous clans becomes extinct — as hap- 
pened in a Hopi village — two clans remain as a secondary 
phenomenon that is evidently an independent growth. 
Similarly, the ranging of Crow Indian men in two rival 
military organizations about 1870 was simply due to the 
extinction of several societies recorded among them two 
generations earlier. The dual grouping is a consequence 
of special conditions, and is neither conceptually nor 
historically related to the normal exogamous moieties. 
Again, the Angami Naga of Assam are divided into 
moieties that are not now^ but traditionally were at one 
time exogamous. However, in one village recently in- 
habited only by members of one of the old moieties, 
the people all belonged to either of two clans which 
intermarried freely but tabooed marriage of clans- 
folk. That is, these local clans were simply exogamous 
moieties ! Further, in a still more recent period one of 
these newfangled moieties was broken up into six seg- 
ments, while the other remained undivided, with the re- 
sult that a member of one of the new seven clans might 
marry into any one of six others. Thus, the "dual organ- 
ization" is demonstrably a fluid phenomenon which dis- 
appears and reappears in the course of history.^® 

i'^ Perry, op. cit., 56. 

18 J, H. Hutton, The Angami Nagas, 110 sq., 125-132, 418 sq., Lon- 
don, 1921. 


This case leads us back to the problem of inventive- 
ness. Obviously, when the extinction of one clan leaves 
two, the consequent dualism is not a sign of ingenuity 
but an automatic end result. Accordingly, the question 
of human or primitive originality does not enter at all. 
Diffusionists persistently fail to discriminate between 
difficult technological achievements which are by general 
admission rarely, if ever, made independently and cul- 
tural features that impose little or no strain on the intel- 
lect. They further belittle the inventiveness of primitive 
men, stressing their failure to originate devices that 
would add to their comfort. It is true that the Australians 
and Fuegians make shift with inadequate shelter and 
clothing; but other primitive groups have been more re- 
sourceful and certainly did not get their technique from 
Egypt. How could the Egyptians who have never felted 
themselves, transmit felting to the Central Asiatic no- 
mads ? Is is seriously suggested that the Siberian ski and 
snow goggle, the tailored clothing of Arctic peoples, the 
snow house of the Eskimo are directly or indirectly pat- 
terned on Egyptian models'? An intensive study of 
aboriginal technology reveals many processes that com- 
mand our profound respect.^^ If Messrs. Smith and Perry 
neglect them, it is because of their unfathomable igno- 
rance of elementary ethnography. Incredible as it is, when 
Elliot Smith lists the ape-like peoples of the present time 
C almost wholly devoid of anything worthy of the name 
of culture") he includes the reindeer-breeding, butter- 
churning, cheese-making Lapps ! ^^ 

Our conclusions as to inventiveness may then be 
summarized as follows. Human groups, the most sophis- 
ticated as well as the rudest, have often failed to invent 
what in retrospect seem obvious improvements ; even the 

18 See, e.g., Erland Nordenskiold, Comparative Ethnographical Studies, 
8:1-124, Goteborg, 1930. 

'■'^ Elliot Smith, In the Beginning, 20. 


Greeks muddled along with a wretched arithmetical no- 
tation. Accordingly, we cannot for any people assume an 
unlimited stock of creative ideas. But from this it is a far 
cry to deny all significant originality to mankind at large, 
or to limit it to the ancient Egyptians. Especially should 
we beware of ascribing to higher civilizations social fea- 
tures and cosmic or religious conceptions which may 
arise spontaneously, either like the dual organization 
from external factors or by a simple association of ideas 
akin to the classifications of language. No one argues that 
the Shoshone Indians of Wyoming derived their dual 
number from Greek grammar ; in both instances, paired 
occurrence of phenomena happened to strike the primary 
speech-making minds as significant and established the 
dual category. In the same way, such celestial phenomena 
as eclipses may quite readily prompt a similar explana- 
tion in distinct parts of the world. This is not to assert 
that they have evoked independent resemblances, but to 
insist that the analogy of an eclipse with a devouring and 
regurgitating monster is not an intellectual feat to be put 
on the same plane with bronze-casting or the domestica- 
tion of livestock. 

After the preceding remarks it is hardly necessary 
to comment on the exaggerated role ascribed by Elliot 
Smith to degeneration as the normal process everywhere 
outside Egypt. But it is worth while to point out certain 
chronological implications of the scheme. Since the Indo- 
Chinese pyramids are used for dating higher American 
civilizations, our diffusionists are obliged to crowd an 
incredible number of novel developments within the 
period of, say, 500 a.d. to 1492. They must assume that a 
profusion of distinct pottery styles suddenly sprang up 
in Peru, Mexico, and the southwestern United States; 
also that innumerable varieties of plants peculiar to 
America were domesticated during a few centuries, a 
conclusion at variance with botanical evidence. Further, 


the study of tree-rings in the southwestern United States 
has achieved a trustworthy chronology for this area. It 
appears that in northern Arizona the hypothetical ape- 
men made pottery by at least 500 a.d. ; in the southern 
part of the state possibly by the beginning of the Chris- 
tian era. As for the higher civilization of Yucatan, the 
best authorities find the Maya calendar in full swing in 
300 A.D.^^ To turn to another aspect of the case, it is al- 
leged the Indians of the United States did not become 
farmers until several centuries ago, i.e., they remained in 
the unclad, houseless, and nonreligious ape-like state until 
then. If so, their cultural development subsequent to 
southern stimulation was staggering, for they certainly 
could not acquire from Mexico what Mexico itself lacked. 
To single out religion, they might take over Mexican 
ritual, but not that highly personal faith which the Aztec 
failed to display. The tremendous differentiation of rele- 
vant beliefs and observances would thus have to be 
crowded into a pitifully short span. And even if this were 
conceded as possible, we should still not be dealing with 
decadence but with a wholly novel series of original de- 
velopments. Sapienti sat. 

Perry requires no separate treatment. Better read in 
ethnography than his master, he is incapable of seeing 
the evidence naively so that all the items gleaned are 
dropped into the preordained pigeonholes. On basic facts 
of culture history he is as prone to error as Smith: we 
merely note his crediting the Maya with an "alphabet." ^^ 


Towards the end of his life W. H. R. Rivers (1864- 
1922) came to form a triumvirate with Elliot Smith and 

21 Kidder, op. cit., 148. 

22 Perry, Gods and Men, 75. For an appreciative survey of Elliot Smith 's 
anatomical work, see T. Wingate Todd, "The Scientific Influence of Sir 
Grafton Elliot Smith," A A 39:523 sq., 1937. 


Perry, but he deserves consideration for real achieve- 
ments which were independent of this association. Rivers 
was medically trained, investigated the physiology of the 
senses, and thence turned to experimental psychology. 
In 1898, aided by 0. S. Alyers and W. McDougall, he was 
the first to subject aborigines, to wit, the Torres Straits 
Islanders, to a thorough set of laboratory tests. Of these 
researches a recent specialist says that they ''might well 
serve as a model for present-day investigators."^^ They 
established certain sensory racial differences, but none 
that indicated a wide chasm between Papuans and other 
races. Above all, irrespective of particular findings, the 
approach was exemplary in its scientific caution, its anal- 
ysis and rejection of popular errors. 

Unlike his ultimate allies. Rivers did notable ethno- 
graphic field work. In the Torres Straits he developed the 
most useful technique known for recording kinship 
nomenclatures and systematically studied the sociology 
of the Islanders. Later he published a valuable tome on 
the Toda, one of the simpler peoples of southern India, 
and a considerable body of new material on Melanesia."* 
These investigations must be rated high, but they suf- 
fered from certain deficiencies. Rivers views natives as 
an outsider, so that his writings lack the intimacy of 
Rasmussen's accounts of the Eskimo or even of the better 
monographs on American Indians which consciously 
strive to afford glimpses of the inwardness of tribal life. 
Moreover, with a one-sided emphasis on sociological 
phenomena there goes the virtual neglect of material 
culture and even of religion. Finally The To das, falling 
into Rivers ' earlier period, ignores the possible influences 

23 Reporis of the Cambridge Anthropological Expedition to Torres 
Straits, Vol. II, Cambridge, 1901. Florence L. Goodenough, "The Meas- 
urement of Mental Fimctions in Primitive Groups," AA 38:1-11, 1936. 

2-1 W. H. R. Rivers, The Todas, London, 1906. The History of Mela- 
nesian Society, Vol. I, Cambridge, 1914. 


of higher civilizations in India, where almost any naive 
observer would have suspected them. 

Topically, Rivers was right in assigning first place to 
his sociological researches. These must not be judged by 
the posthumously published work on that subject ^^ 
(based as it was on meagre lecture notes) but by sundry 
special essays. Rivers veritably revived the sociological 
study of relationship terms, which had virtually ceased 
through revulsion from the rasher of Morgan's theses. 
In 1909 Kroeber, while fruitfully paving the way for 
work on the linguistic categories embodied in kinship 
systems, denied any social determinants. Five years later 
Rivers took up cudgels not on behalf of Morgan's scheme 
but in vindication of his basic postulate that the nomen- 
clatures had sociological correlates."® More suo, Rivers 
was carried away by his enthusiasm, proclaiming that 
every detail of relationship terminology was determined 
by social conditions, a conclusion certainly at best un- 
proved. But he advanced our insight in several ways. 
He criticized Morgan's typology — the inappropriateness 
of the concept of "descriptive" systems as applied to the 
common Indo-European terminologies. He proved be- 
yond question that Melanesian forms of cousin marriage 
are functionally related to the designation of kin. Fi- 
nally, he formulated in a new way the problem, previously 
broached by Tylor, of how far exogamy was linked with 
the more usual of Morgan's "classificatory" types. The 
definitiveness with which such questions could be at- 
tacked led to a rebirth of interest in relevant researches, 
so that vast bodies of material bearing on the basic issues 
were accumulated and remain valuable irrespective of 

25 Social Organization, edited by W. J. Perry, London, 1924. 

26 W. H. E. Eivers, Kinship and Social Organisation, London, 1914. 
A. L. Kroeber, " Classificatory Systems of Relationship," JRAI 39:77- 
84, 1909. 


their relation to Rivers' conclusions.^' Naturally, since the 
terms were presumptively linked with social institutions, 
all kinship customs, taboos and privileged familiarity, 
for instance, were scanned so that our knowledge of them 
progressed by leaps and bounds. 

Medically and psychologically trained, Rivers did 
army service during the War, treating cases of shell 
shock. His alert and suggestible mind was affected by the 
rise of psychoanalysis, and on that basis he attempted to 
ally ethnology with psychology .^^ Whatever he may have 
added to psychological science in this way, he hardly ad- 
vanced ethnology; to us, at least, he does not seem to 
have done more than to paraphrase ethnographic facts 
in psychiatric argot. 

With his ability and energy, his fusion of theoretic 
interests and field experience. Rivers at one time loomed 
as the prospective leader of British anthropology. Yet 
for a variety of reasons he failed to wear Tylor's mantle. 
First of all, manifold as were his interests, he never made 
them converge upon culture as such. Unlike Tylor and 
Boas, he never devoted himself intensively to the study 
of language ; he never mastered the history of culture as 
a whole, never advanced the study of either technology or 
comparative religion or art. His rather late incursion 
into ethnology left him a relative stranger, and he never 
took steps to acclimatize himself. He fostered the com- 
mon delusion that classical English anthropology had 
been unaware of diffusion; and only personal Oceanian 
experiences converted him to a belief in the complexity 
of culture.^^ Whether Rivers ever read Tylor's Researches 
or his discussion of the patolli game is a fair question. 

27 E.g., E. W. Gifford, "California Kinship Terminologies," UC- 
PAAE 18:1-285, 1922. 

2« " Conservatism and Plasticity," Folk-Lore, 32:10-27, 1921; Mind 
and Medicine, Manchester, 1919; Dreams and Primitive Culture, Man- 
chester, 1917- '18. 

29 The History of Melanesian Society, 2 : 1, 1914. 


Even in the special field of sociology he strangely ignored 
Tylor's treatment of exogamy in relation to classifica- 
tory systems. Rivers was often like a modern mariner 
shouting with glee over the discovery of America. 

Another handicap was a lack of mental poise that 
perhaps still more sharply distinguished Rivers from his 
great predecessor. In his later work he displayed an 
amazing autosuggestibility. An idea that sprang into 
his mind forthwith assumed the character not of an as- 
sumption to be tested, but of an axiom from which star- 
tling conclusions might be deduced. Thus, Rivers cites 
several interesting cases of the disappearance of useful 
arts in Oceania. In one island, which must of course have 
been reached by water, the natives no longer manufac- 
ture canoes ; elsewhere pottery, revealed in archeological 
sites, is a lost art ; while in still other localities the bow, 
once an important weapon, has been abandoned. 

Certainly the instances — especially the loss of 
canoes — are striking; and Rivers made a contribution 
here because, though Tylor recognized the reality of de- 
cline,^" he rather minimized its likelihood in case of ''arts 
which belong to the daily life. ' ' Further, Rivers plausibly 
explains such decadence by irrational factors : if a craft 
is practiced — as in Polynesia — by religious officials, their 
death would suppress it because the proper ritual, a 
priestly privilege, is a prerequisite. Thus, Rivers gave a 
new support to Hahn's insistence on the potency of non- 
utilitarian motives.^^ 

These conclusions, however, were not deemed suf- 
ficient. With exemplary candor Rivers explains his 
interest in the matter. He wants to analyze Oceanian 
culture into several complexes carried over wide regions 
of the globe, and for such analysis it would often be nec- 

^ Researches, 180 sq. 

31 W. H. E. Rivers, "The Disappearance of Useful Arts," FestsTcrift 
tillagnad Edvard Westermarck, 109-130, Helsingfors, 1912. 


essary to assume loss of elements; ''and the probability 
and stability of any analytic scheme will be greatly pro- 
moted if one is able to assign motives for the disap- 
pearance. ..." Because even an art so useful as naviga- 
tion can disappear, Rivers at once assumes that it has 
disappeared over and over again: the Tasmanians may 
have reached their island in craft superior to their rude 
balsas, the South Americans may once have owned boats 
of which no vestige remains, and so forth. The fallacy of 
this argument is clear. Though men have abandoned 
utilitarian industries, they have not done so everywhere 
and all the time. As Tylor rightly felt, the general course 
of technology has been progressive rather than retrogres- 
sive. The contrary assumption, consistently carried out, 
might lead to the inference that the rudest hunters of to- 
day w^ere once superior to ourselves in the material arts 
— a conclusion not supported by their prehistoric re- 

In a supplementary paper ^' Rivers advances the 
idea, equally useful for extremist inferences, that a few 
immigrants possessed of a superior technology can im- 
pose their customs on a large autochthonous population. 
In characteristic fashion our author combines this con- 
ception with his earlier theory of decadence, applying his 
synthesis to Australia. Australians inter their dead, or 
put them on platforms, or embalm them, or cremate the 
corpse. Whence this amazing diversity of an emotion- 
ally tinctured practice? exclaims Rivers; only a succes- 
sion of distinct immigrations could account for such 
variety, and only a superior material culture would in 
each case cause the aborigines to adopt the new form of 
burial. Yet how can this repeated infusion of new strains 
be reconciled with the racial homogeneity of the present 
Australians I And if the technology of the newcomers was 

32 ' < The Contact of Peoples, ' ' Essays and Studies Presented to Wil- 
liam Eidgeway, 474-492, Cambridge, 1913. 


superior, what has happened to itf Here is indeed a para- 
dox, a set of intricate problems to delight the heart of a 

The solution is ingenious. If the immigrants were 
uniformly few in number, their racial strain would fail 
to assert itself in the total population. Their material 
arts must indeed have transcended the level of the na- 
tives, but of course useful arts disappear! Each time, 
then, a group of new immigrants came they degenerated 
technologically, whence the persistent lowliness of Aus- 
tralian culture as we know it. ". ... it is essential to 
the argument . . . that this disappearance of useful arts 
should have taken place in Australia, and on a scale 
perhaps unrivaled anywhere else in the world." Burial 
rites and other irrational elements remained, whence the 
combination of sociological and religious complexity with 
technological crudity and physical unity. 

This argument is characteristic of Rivers' "histori- 
cal" thinking. From beginning to end it rests on pure 
fantasies ingeniously interwoven. The coexistence of 
several burial practices, with which the discussion starts, 
is no cause for amazement because it is a common situa- 
tion among very primitive peoples. To cremate or bury 
a corpse is not the same as inventing pottery. Calif ornian 
tribes both inter and burn the dead; Canadians have 
platform and tree disposal, inhumation, cairn-burial, and 
cremation. The attitude of peoples toward death varies 
so much that we cannot predict how tenaciously they 
would cling to a traditional form of burial. Indeed, 
Kroeber has plausibly argued that the disposal of the 
dead is largely a matter of fashion.^^ Certainly the ap- 
pearance of a superior technology is not the only factor 
that would cause a change in ritual; the Ghost Dance of 
1891 started from the lowly Paviotso of Nevada and was 
eagerly borrowed by the superior Plains tribes. Further, 

33 A. L. Kroeber, "Disposal of the Dead," A A 29:308 sq., 1927. 


conceivably the same early band of immigrants intro- 
duced all the modes of disposal, fitting each to a special 
circumstance, and so forth. In short, alternative solu- 
tions abound. There is no real problem, but a sham 
problem resting on the dogma of aboriginal uninventive- 
ness ; and it is solved by interweaving possible but unde- 
monstrated determinants into a scheme supported by not 
a single verifiable fact. 

After this comparatively harmless sample we may 
forego the intricacies of Rivers ' ambitious scheme as elab- 
orated in the second volume of The History of Melane- 
sian Society. Hypotheses are here reared upon hypotheses 
until the bewildered reader asks himself in what possible 
sense this could be called ''history." 

Let us, then, briefly summarize the contributions of 
British diffusionists. Those of Elliot Smith and Perry 
are probably nil. Rivers did excellent work, but independ- 
ently of his "historical" reconstructions. He looms in 
the annals of ethnology for his pioneer work on the 
psychology of a colored race and, above all, for his 
investigations of relationship systems and kinship 




In the range and solidity of their knowledge, the 
German diffusionists are incomparably superior to their 
British counterparts. Prolix and pedantic as Graebner 
frequently is in the Methode, his special papers reveal a 
wealth of ethnographic information; and his '* Ethnolo- 
gic," discussing the whole gamut of civilization, 
geographically and topically, explodes many of the tra- 
ditional criticisms leveled against him. It certainly nei- 
ther considers only museum objects nor ignores the 
psychology of diffusion. As for Father Schmidt, he con- 
stantly astonishes us by the width of his reading, while 
his interests embrace technology no less than sociology, 
comparative religion, linguistics, and prehistory. The 
mutual relations of these two leaders have been repeat- 
edly set forth in Schmidt's writings; their differences, 
from some angles significant, may be provisionally ig- 
nored when their position is to be constrasted with that 
of other schools. Let us, however, first note some points 
of agreement with their contemporaries.^ 

1 For an appreciation of Graebner, see : Julius Lips, ' ' Fritz Graeb- 
ner," AA 37:320-326, 1935, and Wm. Schmidt, Der Ursprung der Got- 



Like Boas, Rivers, and Ratzel, then, the modern Ger- 
man diffusionists oppose as oversimplified the evolution- 
ary schemes of an earlier day and explain the complexity 
of real history in terms of contact metamorphosis. De- 
velopment, they further find, is not uniform, so that a 
people with simple technology may have an advanced so- 
cial structure or form of worship. We must shun a priori 
schemes and determine tlie actual course of events. 

With the British school, but to a much lesser degree 
with more moderate diffusionists, the Germans share a 
belief in the uninventiveness of man. Independent origins, 
while abstractly conceivable, are therefore almost al- 
ways ignored, though Schmidt and Koppers explicitly 
except the conical roof of Asiatic nomads and of higher 
hunters." Unlike Elliot Smith, the culture circle theorists 
propound a multiple development, not a single evolution 
on the Nile, followed by universal degeneration else- 
where. What unites the Graebnerians with the pan- 
Egyptians is merely the attempt to present the totality 
of culture history, an attempt not made by such histo- 
rians as Boas, Nordenskiold, Hatt, or Wissler. 

The German diffusionists, then, are not the only 
diffusionists or historians. Their differentia lies in a 
particular system of culture strata, by which they explain 
the growth of civilization in all periods and all regions. 

Graebner and Schmidt picture primeval man as liv- 
ing in small groups, presumably somewhere in Asia. 
Isolated and without means of transportation, these pop- 
ulations evolved a number of distinctive cultures {Kul- 

tesidee, 1:743 sq., Miinster i. W., 1926; idem, A, 1935. For a vindication 
of both wings, see Clyde Kluekhohn, ' ' Some Eeflections on the Method and 
Theory of the Kulturlcreislelire," AA 38:157-196, 1936. 

Graebner 's most significant works, besides those already cited, prob- 
ably are: " Ethnologic, " in Die Kultur der Gegenwart, Teil 3, Abteil 
5:435-587, Leipzig, 1923; "Die melanesiche Bogenkultur und ihre Ver- 
wandten," A 4:726-780, 998-1032, 1909; Das Weltbild der Primitiven, 
Munchen, 1924. 

2 Schmidt and Koppers, V biker und Kultur en, 488, 53? 


turkreise). As modes of travel improved, the influence 
of these centers began to radiate mainly in the aggregate 
rather than by the spread of single elements. When two 
of these systems clashed, they either blended or one de- 
stroyed the other. Lacking boats, the earliest immigrants 
into America and Africa were obliged to enter by narrow 
passages, whence they were successively pushed farther 
by newcomers into the most remote and inhospitable 
regions, to persist there as marginal populations. Such 
wanderings successively carried primitive modes of life 
as complexes to the four quarters of the globe. Losses 
and modifications were inevitable, but they do not pre- 
clude identification of the cultural totalities even today. 
Thus, Graebner relates the ''Melanesian bow culture" 
to the Neolithic of Central Europe because both share 
pile dwellings, a rectangular ground plan, coiled pottery, 
a special mode of hafting adzes, and spoons. The lack of 
cattle and other traits in Oceania is ascribed to geo- 
graphical deterrents.^ 

These writers are aware of the intricacy of inter- 
tribal influences and repeatedly allude to secondary 
complications. Though Graebner, like Elliot Smith, de- 
rives the civilizations of Mexico and Peru from the out- 
side, he does not imagine an irreversible stream of 
elements pouring in from a single western focus. He 
concedes that some Polynesian traits may have been 
imported from America; and the residual parallels of 
the New and the Old World are not derived from a sin- 
gle source. Again, the African distribution of totemism, 
headrests and conical roofs is not explained by one early 
dissemination, but is partly ascribed to fairly recent 

This very fairness, however, involves the school in a 

3 The exposition given above follows mainly Schmidt and Koppers, 
op. cit., 64 sq., 71 sq., 111. 

* Graebner, " Ethnologie, " 464, 496. 


quandary. They grant that during the thousands of years 
of culture history any people has been exposed to varied 
alien ideas. By what process, then, can we recognize the 
several basic complexes as integral units? And to this 
epistemological query there is no satisfactory answer. 
The Kreise are before us as ultimate axioms and by in- 
genious shuffling of their constituents the whole of civili- 
zation is explained. But what led to the definition of the 

Graebner began his researches with Oceania, where 
he recognized six successive layers, which he labeled, 
respectively, the Tasmanian, Old Australian, Totemic, 
Moiety, Melanesian Bow, and Polynesian.^ Of these his 
first and last are geographically defined; and all scholars 
w^ould agree that the Tasmanian culture is ancient, the 
Polynesian recent. But what about the intermediate parts 
of the scheme! Let us select two distinctive complexes 
for scrutiny — the Totemic and the Moiety culture. 

In Australia the Totemic complex is primarily pe- 
culiar to the central and eastern areas. It embraces the 
foUow^ing material elements: penis-sheath, stiff bark 
girdle, conical-roofed huts, dugouts, headrests, spears 
\\ stone points or wooden barbs, and spear-throwers. 
Sociologically, there is totemism, always associated with 
patrilineal hordes. Burial is on platforms, boys' initia- 
tion rites probably involve circumcision, in the decorative 
style looms a band with marginal triangles or semicir- 
cles. Mythology is astral, the sun playing a major part. 

Now, first of all, how does Graebner arrive at this 
combination of traits? Avowedly, he is no longer de- 
scribing a continuous geographical unit as in the case of 
Tasmania. On the one hand, he admits that elements of 
the complex occur probably all over Australia ''because 
of secondary movements." On the other hand, criteria 
of this culture are said to be lacking in Australia, viz., 

5 Graebner, op. cit., 449 sq. 


the penis-sheath, headrests, and bark girdles; that is, 
they never spread from Asia beyond Melanesia. The 
conical roof, too, is not found in Australia, but Graeb- 
ner suggests that the center post of Australian dwell- 
ings may be a survival of the true Totemic type of hut. 
For such phenomena he offers alternative explanations. 
Features extending from the north may simply have 
originated after the main southward movement of the 
complex; or they may represent local Melanesian dif- 
ferentiations; or they did migrate to Australia but 
proved unviable because not in harmony with the pre- 
existing Australian culture. All these explanations are 
plausible if we admit the reality of the Totemic complex 
to begin with. But why Graebner assumes that this ar- 
bitrary association of traits once marked a coherent area 
in Oceania is not clear. When he traces the complex to 
remote regions, the persistent elements naturally di- 
minish; for Africa Graebner admits its extreme at- 
tenuation. Totemism itself, he concedes, is far rarer 
there in its hypothetically primary local and patrilineal 
form, because it has spread widely through secondary 
transmission.® But how do we know empirically what 
is primary and what is secondary diffusion? We observe 
merely that certain specific features cohere in particular 
localities. Only subjective abstraction from the immediate 
data establishes the hypothetical complex; and other 
scholars might well combine different features into an 
equivalent Kreis. This is not pure empiricism, but em- 
piricism largely diluted with a priori speculation. 

To turn to the Moiety complex, according to the 
scheme moieties are primarily matrilineal and go with 
cultivation of the soil. The bearers of this culture raise 
yams, navigate in plank boats, build gable huts, saw 
fire instead of drilling it, and wield heavy clubs. The 
dual organization further accompanies men's secret so- 

* Graebner, op. cit., 464. 


cieties, whose members don masks to impersonate and 
worship the dead. Myths are predominantly lunar, dec- 
orative art curvilinear, musical instruments include the 
Panpipe. Here, again, Australians have been selective 
borrowers, rejecting farming, masquerades, and musi- 
cal instruments. When the matrilineal and the totemic 
circles fuse, the totem groups are arranged in comple- 
mentary moieties, which may be either patrilineal or 
matrilineal. Some Melanesians thus have totemic moie- 
ties, others multiple totem groups of the matrilineal 

A West African counterpart is recognized, though 
this is conceived as a union of the Moiety complex with 
another matrilineal culture kept separate in Melanesia. 
Weaving, the varied forms of musical instruments, and 
metallurgy mark the African equivalent as more recent ; 
it lacks the Oceanian clubs and plank boats. Sociologi- 
cally, the western Negroes have achieved distinctive in- 
novations by blending totemic and moiety ideas, their 
totems remaining patrilineal, while exogamy is regulated 
through the mother's line.^ Corresponding explanations 
are offered for America. 

Here we face one of the fallacies of the system. It 
is dogma to treat totems as primarily patrilineal and 
moieties as matrilineal, and then to say that deviations 
from this norm must be due to blending.® Empirically, 
North America alone presents so many formidable ex- 
ceptions that such explanation becomes incredible. Thus, 
the Southern Siouans have patrilineal moieties; the 
Hopi lack moieties but have multiple clans with totemic 
names and matrilineal descent. Of course, the facts can be 
fitted into the scheme by an auxiliary hypothesis for each 
deviant, but such supplementary assumptions progres- 

^ Graebner, op. cit., 452 f . 

* IMd., 464 sq. 

® Schmidt and Koppers, 87. 


sively weaken the dogma. Thus, Father Schmidt suggests 
that in America matrilineal moieties emanate from the 
Athabaskans, who transmitted them in the north to the 
coastal tribes of British Columbia and in the south to the 
Pueblo tribes/" who are conceived as formerly patri- 

Now such an explanation would hardly occur to most 
Americanists. The Northern Athabaskans, so far as un- 
touched by coastal influences, are notoriously simple folk 
who completely lack any clan system. This holds spe- 
cifically for people as far west as the Great Bear Lake. 
Dr. John M. Cooper, one of our most trustworthy in- 
vestigators, informs me that his detailed inquiries as to a 
clan organization among the Chipewyan of the Great 
Slave district yielded wholly negative results. Indeed, so 
experienced an observer as Father Morice generalizes 
the denial for all the eastern tribes from Hudson Bay to 
the Rockies and ascribes matrilineal clans only to those 
Far Western Athabaskans ^'who live in regularly con- 
stituted villages, had adopted matriarchy, with all its 
consequences, after the example of the coast Indians." ^^ 
As for the Pueblos, there has indeed probably been a 
fusion of the moiety idea with a clan system, but accord- 
ing to a quite different pattern from that assumed by 
Graebnerians. The Southern Athabaskans — Navaho and 
Apache — are matrilineal but have multiple clans; it is 
the Hopi with an ''all-penetrating matrilineal clanship 
system" who lack moieties; while some of the Eastern 
Pueblos have moieties but with patrilineal descent.^^ 

10 /bid., 231. 

" C. B. Osgood, ' ' The Ethnography of the Great Bear Lake Indiana, ' ' 
Annual Report, 1931, National Museum of Canada, 74, Ottawa, 1933. A. 
G. Morice, "The Canadian Den6s, " Archaeological Report, 1905, 201, 
Toronto, 1906; idem, "Are the Carrier Sociology and Mythology Indige- 
nous or Exotic?" Transactions Boyal Society of Canada, Section II, 109- 
126, 1892. 

12 E. C. Parsons, "The Religion of the Pueblo Indians," ICA 21 
(First Part): 140, The Hague, 1924. 


In short, though the Kulhirkreis theorists allow for 
greater complication than the parallelists and Elliot 
Smith, they still oversimplify the facts in practice, if not 
in theory. Tribal crosscurrents are constantly producing 
effects that cannot be defined by means of a few cultural 
complexes. Only intensive work in each major area can 
determine what really took place. 

This leads to another point. All the diffusionists 
from Ratzel on are right in treating the history of man- 
kind as one unit ; this implies ipso facto that no Monroe 
Doctrine can segregate America from the rest of the 
world. Actually Boas and his collaborators on the Jesup 
Expedition ignored continental boundaries and proved 
far-reaching relations between Siberians and American 
Indians. Similarly, Speck and Cooper have traced scapu- 
limancy from northern Europe through Asia to eastern 
North America; and Gudmund Hatt connects Lapp and 
Asiatic traits with New World phenomena." If a divin- 
atory practice in Albania can be derived from the same 
source as its parallel in Labrador, there can be no ob- 
jection to establishing contacts over any distance what- 
soever. The essential thing, however, is to compare only 
strictly comparable phenomena. Here lies the error com- 
mon to all the extreme diffusionists : they mistake anal- 
ogies for homologous features. 

This is of course the fallacy underlying Perry's 
treatment of the ''dual organization," an error not even 
shared by his ally Rivers.^* Graebner and Schmidt are not 
so crassly crude as to subsume under one head whatever 
can be considered in pairs ; but they do fail to see that a 
moiety system need not be conceptually the same in Aus- 
tralia and in New York and that it can arise independ- 
ently in distant regions. And as with moieties, so with a 

"F. G. Speck, Naskapi, 158 f., ISTorman, Oklahoma, 1935, John M. 
Cooper, " Scapulimancy, " ALK, 29 sq., Berkeley, 1936. Gudmund Hatt, 
ArktisTce STcinddragter i Eurasien og Amerika, Copenhagen, 1914, 

"W. H. E. Elvers, Social Organization, 31, New York, 1924, 


number of other concepts. One feature shared by the 
marginal Fuegians of South America with Graebner's 
Old Australians is the type of dwelling, a wind-screen or 
a dome-shaped hut/^ These structures, however, are so 
crude and architecturally undefined that their designa- 
tions are only classificatory labels ; and because the 
dwellings are so simple they may have been invented over 
and over again. When Koppers pictures the very origin 
of human habitations, he asks, "What was more obvious 
than to perfect these natural protective roofs" (viz., 
trees and shrubs)? If the procedure is so obvious to 
primeval man, it can evidently have occurred also to his 
descendants, so that from its distribution we can infer 
nothing as to dissemination. To take another Graebner- 
ian trait, "skin clothing," a hunter is very likely to uti- 
lize the skin of his quarry, and no historical conclusion 
seems warranted from this common practice. 

To summarize the main objections to the scheme: 
(1) It is not clear how the fundamental complexes are es- 
tablished as historical realities; (2) the component traits 
— including such vital elements as the moiety principle — 
are in part classificatory devices without historical real- 
ity; (3) some of the elements may very well have arisen 
independently; (4) the complexity of actual events is too 
great to be described by the interaction of a small number 
of cultures, though their deft dialectic manipulation can 
of course give a self-consistent scheme through the in- 
troduction of auxiliary hypotheses ad hoc. 

At this point we must take cognizance of an interest- 
ing French offshoot of the Graebnerian movement. It 
emanates from Professor George Montandon of the Nicole 
d 'Anthropologic at Paris.^® Originally a physical anthro- 
pologist, this author broaches ideas obviously tinctured 

1^ Schmidt and Koppers, op. cit., 80, 439 sq. 

18 George Montandon, L'ologSndse culturelle; traitS d 'ethnologic 
cyclo-culturelle et d'ergologie sysUmatique, Paris, 1934, 


by his earlier training. To him the Kulturkreise appear 
as the conceptual equivalents of races, though he ex- 
pressly repudiates the idea that races and cultures must 
coincide. As somatology no longer clings to a geographi- 
cal limitation of race, so ethnology must break with the 
tradition of defining its major units in territorial terms. 
Thus he accepts the notion of cycles culturels consisting 
of elements that remain associated par compagnonnage 
traditionel rather than by an organic affinity. In short, 
Montandon deliberately enrolls himself under the Graeb- 
nerian banner. 

The biological flavoring of this variant appears in 
the distinction of "higher" and ''lower" samples of 
civilization. Beyond the primeval stage, Montandon 
ranges all cultures in either the rameau tardif ("back- 
ward branch") or the rameau precoce ("precocious 
branch"). The former grow slowly but surely, culminat- 
ing in modern civilization ; the latter are capable of much 
complexity but lose themselves in blind alleys. Under the 
spell of phylogenetic theories the author is evidently ap- 
plying to ethnology what zoologists tell us about anthro- 
poids in contrast to hominids. 

This logically precipitates a relapse into the sub- 
jective evaluations of the Lubbock period. As proof of 
the merely relative and specialized development of higher 
American Indian cultures Montandon cites "the spirit- 
ual aberration attested by excessive human sacrifice." 
He concludes : ' ' Such a civilization was bound to col- 
lapse." But, apart from the absence of major human 
sacrifices in Peru, the ritual killings of the Aztec seem 
paltry indeed beside the wholesale massacres of Euro- 
pean warfare. Science has nothing to do with such rank 

Disregarding these quaint idiosyncrasies, we readily 
concede to Montandon some of the strong points of Ger- 
man diffusionism. He, too, views mankind as a connected 


unit and seeks to combine ethnographic with archaeo- 
logical facts. He deserves credit for being less intransi- 
gent than his compeers on the subject of independent 
development. In principle he even regards it as an open 
question whether whole cultures may have become simi- 
lar independently. As to special parallels, his discussion 
of the coiled basketry {vannerie spiralee) of Australia 
and Fuegia is instructive : they could be safely assigned 
to one center only if there were resemblances of complex 
varieties and shades {nuances compliquees) . With this 
principle not even the most moderate of ethnological 
historians could quarrel. Montandon likewise escapes the 
crass monism of Elliot Smith by insisting that a culture, 
instead of arising in one definite point, may result from 
a fusion of traits within a larger area. The documented 
facts about Egypt, Babylonia, and China support this 

On the other hand, Montandon shares the character- 
istic errors of diffusionists. He lightheartedly assumes 
that 'Hotemism" — at best a convenient label — is a phe- 
nomenon of unitary character and origin. Surely its var- 
iations are more remarkable than those of Australian 
and Fuegian basketry? Finally, Montandon is no clearer 
than the German theorists in explaining how the cultural 
cycles are derived and why it would not be equally legit- 
imate to combine other features into comparable com- 
plexes to be traced over the globe. 

But we must not allow skepticism to blind us as to 
the real merits of the Graebnerian movement and we 
must specifically repel certain unfair strictures. For ex- 
ample, it has been alleged with vehemence that Graeb- 
ner's outlook is mechanical, excluding an understanding 
of the dynamics of diffusion or indeed of any psycho- 
logical aspect of social life. This is a grave error. Minds 
that see only heroes and villains in life naturally indulge 

1^ Montandon, op. cit., 41, 501. 


in such unrealistic antitheses as History vs. Function- 
alism or History vs. Psychology. Graebner and Schmidt 
are avowedly interested, above all, in the sequence of 
events, but this does not automatically bar a function- 
alist or a psychological approach in principle. 

Looking, for example, into one of Graebner 's later 
publications, we find him facing — precisely as did Boas, 
Thurnwald, and Radcliffe-Brown — the process of contact 
metamorphosis.^® He shows that mechanical juxtaposition 
does not inevitably evoke a transfer of arts and beliefs ; 
that borrowing proceeds selectively, with varying speed 
for elements of distinct categories ; that some phenomena 
have an inherent affinity for others, predisposing to as- 
similation, while other traits are negatively correlated 
and bar or hinder adoption. Finally, Graebner indicates 
how borrowed elements acquire novel significance among 
the recipients: how lunar myths turn into vegetation 
myths, how curvilinear designs adapt themselves to a 
rectilinear style, and so forth. The constantly reiterated 
stricture that Graebner deals with transmission in a 
purely mechanical way is as baseless as the gibe about 
his exclusively museological outlook. Goldenweiser, for 
instance, alleges that the interrelation of associated 
traits is ^' quite beyond Graebner 's horizon"; and by 
contrast he praises Rivers for observing *'the psycho- 
logical interplay of cultural features." ^® As we have just 
indicated, these are precisely the points made by Graeb- 
ner himself, and Goldenweiser 's critique rests on ig- 
norance of the ''Ethnologic." All that can be urged is 
that Graebner evinced a disproportionate preference for 
historical reconstruction; but that surely is a matter of 

In Father Schmidt's writings, too, psychological 

"Graebner, " Ethnologie, " 577 sq. 

18 A. A. Goldenweiser, History, Psychology, and Culture, 84, 150, New 
York, 1933. 


points of view are far from negligible. While he is not 
always willing to curb irrelevant ethical appraisal, his 
attitude toward primitive races is exemplary in its gen- 
eral appreciativeness. In the matter of native endowment 
he is on the side of Waitz as against Klemm. More sig- 
nificant still is his unremitting insistence on the individ- 
ual variability of primitive man — in opposition to the 
traditional view that on simpler levels personality is 
merged in society."*' Here once more Schmidt's views 
coincide with those of Boas; and it may be noted that 
Graebner expresses himself to the same effect, if less 

Further, both Graebner and Schmidt — contrary to 
superficial criticisms — recognize an interrelation of cul- 
tural phenomena. This is an inevitable consequence of 
their following Grosse's correlation of family types with 
economic activities,^^ a position that verges on economic 
determinism. A Kulturhreis is not a random series of 
traits ; in part the organic bond uniting the concomitants 
is specially stressed. Masks, for instance, are treated as a 
natural correlate of a spirit cult; the realism of African 
carvings is derived from their mimetic magical purpose ; 
the totemistic world-view naturally expresses itself in 
animal figures.^^ Again, whatever one thinks of Father 
Schmidt's account of the Moiety complex, it bristles with 
correlations on the model of that good old functionalist, 
Bachofen. In its primary form he derives matrilineal de- 
scent from feminine tillage and links it with feminine as- 
cendancy, girls ' puberty rituals to the exclusion of boys ' 
initiations, and female deities.^^ 

Finally, in spite of themselves, these arch-historians 

20 See e.g., W. Schmidt, Ber Ursprung der Gottesidee, Band II, 2. 
Abteilung: 173 sq., Miinster i. W., 1929; Volker und Kulturen, 39, 59. 

21 Ernst Grosse, Die Formen der Familie und die Formen der Wirt- 
schaft, Leipzig, 1896. 

22 Graebner, "Ethnologic," 556, 562 f. 

23 Schmidt and Koppers, op. cit., 86 sq., 264 sq. 


and arch-critics of parallelism firmly believe in evolution. 
For, to recognize that cultures change in time and to see 
single traits as organically related is to admit the pos- 
sibility of a definite sequence. Schmidt, to be sure, has a 
phobia of the word ''evolutionism," but when he lays 
down "the stages of the whole development" {Stufen 
der gauzen Entiuicklung) of matrilineal institutions, it 
is mere quibbling to deny that he throws himself into the 
arms of Evolution, Schmidt differs from Morgan mainly 
in denying universal parallelism, unilinear evolution. In 
other words, he believes that only certain peoples pass 
through a matrilineal condition. Within these groups, 
however, his scheme implies clear-cut evolution. Femi- 
nine tillage ushers in an economic revolution, producing 
matrilineal descent; men first retain their independence 
by mere visits to the homes of their wives; later they 
adapt themselves to the new conditions by matrilocal resi- 
dence, which in turn favors monogamy, the cult of a fe- 
male deity, girls' puberty rites, gynaecocracy ; there 
follows bride-service with strengthening of avuncular 
authority; service, however, passes into bride-purchase, 
with a reversal of parts, woman being now degraded to 
the status of a chattel. This result is promoted by sub- 
versive masculine organizations, the "secret societies," 
whose mummers terrorize women and substitute worship 
of male ancestors for the cult of a human goddess. 

This sequence, to be sure, does not impress us as 
valid. Until these events are actually demonstrated some- 
where on the globe we must reject the scheme as not a 
whit more empirical than Morgan's. But unconvincing as 
it is, the underlying idea remains sound : certain phenom- 
ena are connected so that the presence of one renders 
the occurrence of another more probable. Only this re- 
acts on the basic assumption of radical diffusionism, the 
virtual exclusion of independent multiple occurrences. 
For let us take the nascent stage of the assumed Moiety 


complex. Women have just begun to plant ; then the spe- 
cies planted will begin to spread, as in historic times ; and 
naturally they will do so as an aspect of feminine re- 
sponsibilities. Now, ex hypothesi, tillage by women evokes 
matriarchy; hence the transmission of the first species 
cultivated prior to the development of the whole Moiety 
complex in the original focus will repeatedly produce 
similar consequences and ''the steps of the entire devel- 
opment" may be traversed not once but over and over 
again. It is not possible to couch functional relations in 
temporal terms without postulating the possibility of 
parallelism, a limited rather than a universal parallelism, 
yet parallelism for all that. 

Thus Graebner and Schmidt are by no means so in- 
transigent as appears from selected facets of their writ- 
ings; and a reconciliation with the views of many 
contemporary and supposedly hostile colleagues is not at 
all barred. 

In justice to Father Schmidt, he must not be con- 
sidered solely or even mainly as a follower of Graebner. 
Even in his championship of the culture circle concept, he 
is no satellite, but an independent thinker who materially 
departs from his predecessor's tenets. Thus, he regards 
not Tasmanians but the Pygmies as the nearest group 
to cultural origins among races extant. Let us note that 
Schmidt excels Graebner in his sense of cultural to- 
tality so that when he deals with a geographically de- 
limited population — say, the Central Asiatic nomads — 
his picture is at once vivid and sound. His conception of 
the rise of higher civilizations is also noteworthy from 
the same angle; and their derivation from a fusion of 
hoe-tillage, animal husbandry, and specialized crafts- 
manship seems essentially convincing. 

Among the important by-products of Schmidt's anti- 
parallelism is his conclusion that very primitive groups 
may have a conception of a Supreme Being, the core of 


the argnmoiit in liis notable work Der Ur sprung der Gott- 
esidec (G vols., 192G-1935). In our opinion the contention 
is borne out by the facts; and though Schmidt had a 
forerunner in Andrew Lang, the Austrian scholar's amaz- 
ing erudition has done much to elucidate this point. One 
qualifying comment is necessary. We must not ignore the 
significant fact that the notion frequently fails to domi- 
nate religious consciousness. The Ona, e.g., admittedly 
solve most of their problems by an appeal to shamanism 
rather than to their otiose high-god. In other words, their 
capacity to conceive a Supreme Being stands vindicated, 
but it is rather an intellectual achievement than a phe- 
nomenon of practical importance. As Kierkegaard might 
say, Schmidt examines the natives in philosophy and 
then gives them a high mark in religion. 

Ethnology owes much to Schmidt for the establish- 
ment of Anthropos, a journal second to none in the field. 
With unsurpassed energy Schmidt enlisted the services 
of missionaries scattered over the globe and thereby se- 
cured priceless descriptive reports from men resident in 
remote regions for a long span of years, hence thor- 
oughly conversant with the customs and language of 
their native parishioners. Major contributions of this 
sort have been published in monographs of the ' ' Anthro- 
pos-Bibliothek." The journal has not by any means been 
restricted to missionaries, but has numbered among its 
contributors such scholars as Birket-Smith, Ernst Grosse, 
Berthold Laufer, A. L. Kroeber, F. G. Speck, and Erich 
von Hornbostel. 

While Schmidt has never studied natives in the field, 
he has directly promjoted some very important investi- 
gations. Realizing the significance of the most primitive 
cultures, he has sent some of his disciples to the Pygmies, 
others to the Fuegians. The most conspicuous result of 
these researches is Gusinde's magnificent monograph on 


the Ona,^* now supplemented by his parallel study 
of the Yaghan. It is worth noting that Schmidt's the- 
oretical interest in individual variability has also stimu- 
lated observations along these lines.^^ 

Anticlerical critics have suggested that the field work 
undertaken under Father Schmidt's auspices has been 
unduly colored by Catholic or personal prejudices. This 
is an unfair criticism; let him that is without bias cast 
the first stone. Schmidt has naturally exerted a deep in- 
fluence on his pupils, as any scholar of strong person- 
ality and marked attainments is bound to do. There is no 
evidence, however, that the results have been twisted 
except through the common foibles of humanity. Cer- 
tainly some of his disciples' findings do not harmonize 
with his conclusions. For example, it is one of his pet 
views that the Pygmies represent one basic culture, yet 
in the single Congo area Father Schebesta has observed 
much diversity, due to borrowings from different Negro 
groups, so that the possibility of reconstructing a proto- 
Pygmy world culture seems definitely lessened.^" We 
must therefore exonerate Schmidt of an unjust accusa- 
tion. He doubtless errs at times from overindulging with 
a virtuoso's gusto a natural gift for dialectics, but it is 
unfair to impugn his good faith. 

Our final balancing of the books thus leaves the 
German diffusionists with very considerable assets. What 
Elliot Smith attempted with amazing lack of information, 
they have attacked with much fuller knowledge; and 
what Ratzel left largely undefined, to wit, a generic 
world-wide intercourse, they have formulated in a set of 
definite historical problems. Their initial postulate is ac- 
ceptable. Man did at one time occupy a restricted ter- 
ritory; and when he spread he unquestionably carried 

24 Martin Gusinde, Die Selk'nam, Modling bei Wien, 1931; idem, Die 
Yamana, 1937. 

25 W. Koppers, Vnter Feuerland-Indianern, 216 sq., Stuttgart, 1924. 
28 Paul Schebesta, Bambuti; die Zwerge vom Kongo, Leipzig, 1924. 


with him part, at least, of his cultural inventory. Amer- 
icanists such as Boas, Wissler, and Kroeber explicitly 
concur in this assumption as applied to the earliest im- 
migrants to the Xew World. When, therefore, the Graeb- 
nerians allege the survival of some of these traits in 
Tierra del Fuego, they are not talking arrant nonsense. 
It has, indeed, been objected that cultures, being fluid, 
could not maintain their status through millennia; but 
the matter of stability is a moot problem, and Cooper has 
argued cogently that very rude cultures have a tendency 
to persist, as suggested by prehistoric findings.^' 

Let us, however, mention specific questions illumi- 
nated by the Graebnerian approach. One problem to 
which the German diffusionists have fruitfully directed 
attention is the possibility of an ancient contact between 
Tierra del Fuegian and Calif ornian (or Great Basin) 
culture.^^ Some of the parallels cited are of the most 
specific nature, notably, the obligatory use of a head 
scratcher at a ceremonial occasion and the notion of 
pristine immortality, ensured by washing a dying person. 
We may add the striking parallels of a Yaghan myth and 
the Calif ornian story of Wolf and Coyote, both of which 
contrast a benevolently inclined hero with a marplot 
brother, resemblances not nearly so pronounced between 
the Yaghan Two Brother Myth and that of other South 
Americans. Possibly still more striking is the occurrence 
of the Lecherous Father motif among the Ona of Fuegia 
and the Calif ornians (and other North Americans) ; in 
both areas a man, lusting for his daughters, feigns ill- 
ness, predicts his death, directs his survivors to leave 
his body unimpeded, and urges them to leave and marry 
a man he describes ; after their departure he rises, meets 

27 John M, Cooper, "Analytical and Critical Bibliography of the 
Tribes of Tierra del Fuego and Adjacent Territory," BAE-B: 227, Wash- 
ington, 1917. 

28 W. Schmidt, Der Ursprung der Gottesidee, Band II, 2. Abteilung: 
1031 sq., 1929. 


them from the opposite direction, and attains his end. 
These are not vague hints of intercourse in all sorts of 
directions after the fashion of Ratzel ; here a differential 
relationship is alleged for two widely remote areas and 
supplementary evidence in support has been accumulated. 

It is also directly to the influence of the KuUurkreis- 
lehre that we owe a number of other promising re- 
searches. Von Hornbostel supports the Fuegian-Califor- 
nian relationship on musicological grounds; and he has 
made of Melanesian-Peruvian intercourse at least a de- 
batable question by his comparison of the absolute pitch 
of the Panpipes played in these regions.^^ Solid investiga- 
tions have thus resulted from the stimulus of Graebner 
and Schmidt. 

Our conclusion, then, is not to forego the recon- 
struction of human history in its entirety, but to pursue 
Graebnerian aims after purging his scheme and his 
methodology of their unsound elements. We must use 
only rigidly definable concepts and realize more keenly 
how complex is historical reality. Ours should be the 
caution of a geologist who, having established his strata 
in one continent, does not promptly leap to their identi- 
fication with comparable phenomena elsewhere. What 
Graebner and Schmidt have failed to do is to work out 
American, Asiatic, and African history naively — inde- 
pendently of the originally constructed Oceanian strati- 
fication. The sounder approach will be to work out the 
sequence for each area in complete independence of the 
pigeonholes suitable elsewhere and ultimately to combine 
all findings in a world scheme. 

20 Erich M. von Hornbostel, "Fuegian Songs," AA, 38:357 sq., 1936. 
idem, "Ein akustisches Kriterium fiir Kulturzusammenhange, " ZE, 
43:601-615, 1911. 



In France, anthropology took a course of develop- 
ment distinct from that in other countries. It was a 
Frenchman, Boucher de Perthes, who inaugurated the 
epoch-making advances of prehistory; and his continu- 
ators, from Lartet and de Mortillet to I'abbe Breuil, have 
remained pre-eminent. Man as a biological organism has 
also stirred French enthusiasm for many decades, as the 
names of Broca, Topinard, and Boule testify. But for 
some inscrutable reason the arts and manners of living 
peoples have attracted little interest. There were French 
colonies with Oceanian and Negro populations, but the 
accounts published of them long remained few in number 
and inferior in quality to the comparable reports of Brit- 
ish or German officials. As for scholars trained to observe 
in the field — until lately there were none. The remedy 
came from an unexpected quarter. It was not ethnography 
that stimulated the theory of culture, and through it 
other disciplines. On the contrary, the impulse to field 
research finally emanated from philosophy. The Institut 
d 'Ethnologic, whose Travaux et Memoir es, issued since 
1926, at last represent the equivalent of publications in 



other countries, was sponsored by three men, — Profes- 
sors Lucien Levy-Bruhl, Marcel Manss, and Paul Rivet. 
Of these, only Rivet can be reckoned an ethnographer. 
Levy-Bruhl is a philosopher by training, while Mauss 
ranks as the successor of Simile Durkheim, who was in- 
deed the founder of a sociological school but had both 
studied and taught philosophy. It is, above all, Durkheim, 
Mauss, and Levy-Bruhl who have affected theoretical dis- 
cussion in other countries, and accordingly to them we 
now turn. 


:fimile Durkheim (1858-1917) looms as the author of 
several significant sociological treatises, such as De la 
division du travail social and Le suicide, etude socio- 
logique. Ethnologically he became important in 1898 by 
founding L'Annee sociologique which, while embracing a 
variety of topics, consistently stressed the cultures of 
illiterate peoples. It presented detailed and extremely 
careful analyses of ethnographic literature, as well as 
original articles on such definitely ethnological subjects 
as magic, Australian marriage classes, and the origin of 
incest. While most of these essays expounded the tenets 
of the editors, outsiders were not barred, even Ratzel 
figuring in one of the earlier volumes. Thus, Durkheim 
provided a yearly survey of the literature on all phases 
of civilization, not excluding technology and linguistics ; 
and in that period after Tylor's prime when theory was 
languishing or sprouting in hidden nooks, L'Annee 
sociologique offered a welcome opportunity for the airing 
of basic problems. Durkheim 's own contributions, both as 
memoirs and reviews, are distinguished for their range 
of knowledge and penetration, as in his comments in the 
initial yearbook on Kohler's treatise Zur Urgeschichte 
der Ehe. 

The preface to the opening issue of the series clearly 


expounds the founder's programme. He wished to famil- 
iarize sociologists with those concrete findings of the 
social sciences which might provide materials for a soci- 
ology of the future. On the other hand, the benefits were 
not to be one-sided. History, more particularly, was suffer- 
ing from the lack of a comparative point of view to be 
supplied by the sister discipline. '*It is thus that Fustel 
de Coulanges, despite his profound insight into historical 
matters, misunderstood the nature of the gens, seeing in 
it only a vast agnatic family, — because he knew nothing 
of the ethnographic parallels of this family type." A 
primary aim of Durkheim, then, was to synthesize the 
two sciences, to combine the perspective of the one with 
the documentation of the other. ' ^ Fustel de Coulanges was 
fond of repeating that the true sociology was history; 
nothing is more certain provided that history is worked 
sociologically. ' ' 

Durkheim, then, does not disdain history; but his 
special interest lies in the determination of types and 
laws, without which the facts would lack significance. 
This position explains, respectively, his admission and 
his rejection of historical material. He deliberately ex- 
cludes what is not amenable to comparative treatment — 
historic individualities, "innovators of every type," the 
biographical element — for these, we are told, lack utility 
for the sociologist. Here is a highly distinctive element 
of Durkheim 's philosophy. 

Apart from L'Annee, Durkheim 's contribution to 
ethnology is mainly embodied in Les formes elementaires 
de la vie religieuse (Paris, 1912), a work that has played 
a curious role. It has profoundly influenced several 
English-speaking writers, yet Father Schmidt correctly 
states that "possibly no book has reaped so many eulo- 
gies in detail, yet has been so generally repudiated in its 
main propositions. ' ' ^ Goldenweiser, for instance, after 

1 Wm. Schmidt, Ber Ursprung der Gottesidee, 1:579, 1926. 


paying court to the author's ''wisdom, scholarship and 
noted brilliancy," categorically rejects his conclusions 
on the origin of religion and the relation of the individual 
to society: ''Sharp as is the author's wit and brilliant as 
is his argumentation, one closes the book with the melan- 
choly assurance that Durkheim has left these two peren- 
nial problems where he found them. ' ' ^ Why, one naturally 
asks, is there no illumination from brilliance? This is 
surely the sort of critique that fails to explain what we 
wish to understand, — the nature and cause of a writer's 
influence. In order to penetrate the paradox we must go 
back to Durkheim 's basic aims, more obvious from Les 
regies de la methode sociologique (1894; 6th edition, 
Paris, 1912) and some shorter articles than from his 
major work; and we must try to bring him into rela- 
tion with his contemporaries and immediate predeces- 

Culture had been defined by Tylor as embracing those 
' ' capabilities and habits acquired by man as a member of 
society." This implied, then, a distinguishable field of 
knowledge demanding a separate science with distinctive 
procedures. But ethnologists were by no means quick in 
visualizing the implications. Their tardiness becomes 
apparent when we scan the literature of about 1915, which 
confronts us with a veritable cornucopia of pronunci- 
amentos, each vociferously contending for the autonomy 
of ethnology. Rivers, while not in principle spurning an 
ultimate psychological interpretation, insisted that socio- 
logical (= cultural) phenomena must be explained first 
in sociological terms, not in those of a science dealing with 
simpler data. Hocart argued that "the ever-changing and 
endless variety of custom and belief" could not be de- 
rived from the constant mental traits of humanity. Ac- 
cording to Wissler, psychology could teach the ethnolo- 

2 A. A. Goldenweiser, History, Psychology, and Culture, 373, New 
York, 1933. 


Joists what is innate but could never solve his specific 
problems : ' ' All the knowledge of the mechanism of associ- 
ation in the world will not tell us why any particular 
association is made by a particular individual," — why 
some one invented the bow or inaugurated exogamy. In 
poignant sentences Kroeber outlined the sole end of eth- 
nology as the study of culture regardless of organic phe- 
nomena; its sphere is the social, the individual having 
merely illustrative value/ 

Durkheim had proclaimed this independence of soci- 
ology more than two decades earlier and had drawn the 
logical inferences from it. Because sociology concerns 
itself with a distinctive sphere, explanations by any 
science dealing with simpler data are inapplicable. 
Specifically, psychology is as impotent for the purpose as 
are physics and chemistry to explain organic facts. Hence 
the conclusion: ''The determinant of a social fact must 
be sought among antecedent social facts, not among the 
states of individual consciousness."^ This explains the 
recurring emphasis on group ideas in contrast to individ- 
ual ideas, a point later amplified by Levy-Bruhl. To quote : 
". . . whatever is social consists of representations, 
consequently is a product of representations. " ^ At this 
point we parenthetically recall that Tylor partly con- 
formed to Durkheim 's principles in correlating parent-in- 
law avoidance with forms of residence, employing one 
cultural datum to shed light on another. It is only when he 
goes behind this correlation that he resorts to explana- 
tions of a psychological order, such as abound in the 
writings of lesser anthropologists of the era. 

3 W. H. R. Rivers. Kinship and Social Organisation, 92, 1914. A. M. 
Hocart, "Psychology and Sociology," Follc-Lore, 115-137, 1915. Clark 
Wissler, "Psychological and Historical Interpretation for Culture," 
Science, 43:193-201, 1916. A. L. Kroeber, "Eighteen Professions," AA 
17:283-288, 1915. 

* Les regies de la methnde sociologique, 124 sq., 135. 

' Durkheim, * * La prohibition de 1 'inceste, ' ' L 'Annee sociologique, 1 : 69. 


To return to Durkheim, he not only offered a new 
branch of knowledge, but placed it on a footing of dignity. 
It was to rival the objectivity of the older sciences, start- 
ing with directly observable facts, eliminating the sub- 
jective judgments of values to which Lubbock had fallen 
prey, rigorously applying such methods of proof as were 
consistent with the nature of the data. Such objectivity 
had indeed been achieved by some earlier and contem- 
porary ethnologists; but probably none of them had so 
explicitly set it forth as a matter of principle. The appeal 
of such an approach to scholars of a certain mentality is 
thus readily understood.® 

But Les regies de la methode sociologique offered 
more than a programme. There were specific conclusions. 
Though classed as an "evolutionist," Durkheim at least 
in theory revolted against the idea of unilinear evolution, 
which he deprecated as oversimplified: there is not one 
species of society, but a number of qualitatively distinct 
types. Historic development breaks up into ' * a multitude 
of fragments which, being specifically different from one 
another, cannot be united continuously." Durkheim 
certainly does explain in parallelist fashion resemblances 
among remote peoples as symptoms of a definite stage 
{symptomatiques d'un certain etat social).'' Nevertheless, 
he realizes better than many of his contemporaries the 
diversity of primitive culture, hence the absurdity of 
lumping together Homeric Greeks, Zulus, and Iroquois. 
This position is well expressed in Durkheim 's review of an 
essay by Steinmetz: ''People argue as though the so- 
called savages or primitives formed a single identical 
social type. ' ' * 

"With this at least intermittent historic sense is cou- 
pled an insight into cultural dynamics. Durkheim knows 

ejfttd., 52, 54, 159 sq. 

7 IMd., 26, 96, 147, 117. 

* Durkheim, L' Annie sociologique, 4:341, 1901. 


that social phenomena are not normally created by de- 
liberate planning. He also neatly distinguishes between 
the utility of an institution and the historical causes of 
its origin. Phenomena may exist without serving a vital 
need, either as survivals or because they never had a use- 
ful function. "What is more, the function may change sec- 
ondarily. In this connection we may add that in a paper 
already referred to there is a timely warning against 
accepting aboriginal explanations as anything but after- 
thoughts devised to sanction a pre-existing practice. ''We 
know how these theories are fashioned ; they are required 
not to be adequate and objective, but to justify practice. " ® 
Thus Durkheim expresses the very idea elaborated by 

As a philosopher, moreover, Durkheim viewed cur- 
rent concepts with keener criticism than was common 
among the ethnologists of the nineties. He saw both the 
irrelevance of race for the elucidation of culture and the 
lack of any instinct to achieve progress. He warned 
against vague catchwords that mask vital differences: 
because two peoples avoid plural marriages, it does not 
follow that their ** monogamy" represents the same phe- 
nomenon ; everything hinges on whether such restraint is 
obligatory on principle or merely the result of natural 

There are, it is true, less attractive features in Durk- 
heim. The desire to free sociology from the yoke of other 
sciences precipitates a quite unnecessary antagonism to 
the psychological approach. There is also a curious 
anomaly. Though the primary purpose is to vindicate the 
autonomy of sociology, though only certain types of logi- 
cal procedure are held compatible with the nature of 
sociological data, the writer cannot rid himself of the 
prejudice that the new science, like its predecessors, must 

8 Durkheim, L'Annee sociologique, 1:55, 1898. 
lOLes rdgles, 48, 112 f., 120, 129. 


aim at general laws. Finally, he postulates as the simplest 
hypothetical form of society the unsegmented "horde," 
— ' ' a social aggregate comprising none other that is more 
elementary, but which splits up immediately into individ- 
uals. ' ' This sort of society may not be demonstrable now 
or in the historic past, but from the combination of such 
independent hordes must have sprung the historic clan 
system, each horde becoming one segment or clan of the 
new totality. What is curious about this is the deliberate 
exclusion of the individual family as a definite unit pre- 
ceding the clan organization, a dogma contrary to all 
recent research, but constantly recurring in Durkheim's 
writings. Here, as in his conviction, elsewhere set forth, 
that matrilineal necessarily precedes paternal descent, 
Durkheim remained an evolutionist of the old school, re- 
lapsing into the parallelist error of treating Australian 
and American societies as rungs of one ladder.^^ 

These defects, however, must be viewed in historical 
perspective. Durkheim's championship of sociology nat- 
urally led to militancy against psychology as the science 
that threatened to keep the new discipline under its thumb. 
At the same time scientific respectability had to be main- 
tained at all costs — hence the insistence on laws. As for the 
misconceptions of the clan and family, they formed the 
common creed of the period. On the credit side, Durkheim 
was on several vital points abreast of the insights in 
process of achievement by Boas and very definitely in 
advance of most post-Tylorian ethnologists. Here was a 
man of high seriousness, filled with the sense of his mis- 
sion, supported by a wide familiarity with the social sci- 
ences, and striving for major generalizations such as had 
apparently ceased to appear from the ranks of ethnology 

This, then, is the background for Durkheim's major 

^^Ibid., 102 f. Cf. E. Durkheim, "La prohibition de I'iuceste et ses 
origines, L' Annie sociologique, 1:1-70, especially 10 f., 22 f., 28, 53, 1898. 


treatise. He attempts to interpret religion as a social 
phenomenon, choosing for his specific theme the religion 
of Australians because to them he ascribes the lowest 
and simplest social organization, viz., a clan system." Ac- 
cordingly, the essence of the phenomenon, masked on 
higher planes, should appear with supreme clarity there. 
What, then, does Durkheim understand by religion? In 
contrast to received ideas, he treats as immaterial the 
belief in personal supernatural beings. A basic dichotomy 
everj^vhere divides what is sacred from what is profane. 
But it is not merely spirits and deities that loom as sacred, 
but also the impersonal force Melanesians call **mana"; 
indeed, when people venerate the sun or the moon or 
the souls of the dead, it is solely as reservoirs of this 
mysterious impersonal entity, so that it becomes a matter 
of indifference whether this force resides in a personal 
being. Since mana is contagious, diffusing like an electric 
current, any object, any rite, is potentially sacred, i.e., of 
religious nature. Thus, a ceremony for increasing totems 
without worship of spirits or gods is not purely ''magi- 
cal" but truly religious because invested with sanctity. 
Durkheim does not, however, confound magic and religion 
under one head. Though magic has sprung from religion 
and borrowed its technique therefrom, he conceives it as 
its very antithesis. Religion invariably presupposes a 
communion of the faithful, while magic, even when of- 
ficially practiced, lacks a church: the magician is neither 
bound to his clients, nor are his several clients bound to 
one another, by any moral ties ; " his laws are rules of 
expediency devoid of holiness. 

How, then, did this sense of the sacred arise? True to 
his principle of eliminating psychological factors, Durk- 

12 Les formes elSmentaires . . . , 88, 238, 255. 

^^Ibid., 58 sq., 65, 284 f., 430, 455 sq., 459 f., 490, 516 ff. The ideas 
on the sacred and profane are foreshadowed in the opening essay of 
L' Annie sociologique. 


heim spurns the influence of dreams and of natural phe- 
nomena on individual minds. What impressed primitive 
man was the overshadowing force of society, the clan of 
which he was a member, to which he owed protection and 
knowledge, without which he was a nobody. Only by sym- 
bolizing their clans could totems — often singularly incon- 
spicuous in themselves — become sacred; indeed, charac- 
teristically, it is the totemic emblem that is more sacred 
than the animal it stands for; and essentially it is the 
social unit, the clan, that is God. The sentiment towards 
this entity was born at those periodic tribal reunions 
which alternate with the humdrum existence of hunting 
and gathering in minute family units. Practically all of 
Australian religious activity is restricted to these major 
assemblies, which evoke a regularly recurring state of 
effervescence; while the complementary season of sepa- 
ratism is filled with profane tasks. In the major assem- 
blies, then, arose the notion of the sacred as contrasted 
with the profane world then left behind ; and this notion, 
identified with the clan, was totemically symbolized. All 
spiritual beings are simply derivates. The soul, which 
Tylor had put at the base of religion, is merely the to- 
temic principle incarnated in each individual member of 
the clan. The Australian high-god is nothing but the prod- 
uct of tribal sentiment transcending that of the clan. For 
totemism is not elaborated by a single clan but by a whole 
tribe that had attained some consciousness of unity. *'0r 
c'est ce meme sentiment de 1 'unite tribale qui s'exprime 
dans la conception d 'un dieu supreme, commun a la tribu 
tout entiere. ' ' " 

Let us pause to examine this remarkable scheme. As 
a general explanation of religion it is at once ruled out 
by several fallacies. The clan is not the oldest type of 
social unit, having been preceded by the family. Austra- 
lian society is not the simplest known, but it is manifestly 

"Ibtd., 238-424. 


more coiiiiilex than that of the Puegians, the Basin and 
California Indians, the Eskimo, or the Andamanese. To 
learn about the origins of religion from them is accord- 
ingly hopeless; the attempt is a relapse into unilinear 
evolutionism. But even were we to waive this point, an 
unbiased inspection of the Australian data would at best 
show totemism as one ingredient of native religion; its 
ascendancy over others is unproved; that other phases 
are derivatives is an arbitrary assumption. At the present 
stage of knowledge it is thus useless to refute in detail 
the ingenious arguments adduced in support of these 

Equally serious is Durkheim's aversion to psychol- 
ogy. Healthy as is his protest against vulgar interpreta- 
tions in terms of mental states, his devotion to sociology 
as an autonomous science becomes doctrinaire and mis- 
leading. There is no hard and fast line between one branch 
of knowledge and another. What should we think of a 
prehistorian who investigated the origin and spread of 
bronze-casting without knowing that bronze is an alloy 
of tin and copper? The most vital problems would escape 
his attention. But metallurgy is much farther removed 
from culture history than is psychology ! The sociologist 
simply cannot get away from states of consciousness, and 
Durkheim does not really eliminate them but smuggles 
them in at his convenience. The etat d' exaltation, which 
he finds at the base of religious emotion is evidently a 
psychological condition; and the part of wisdom — pre- 
cisely if one considers the state fundamental — would be 
to trace it systematically through all its contexts. Then, 
however, it appears at once that comparable exaltation 
is a frequent concomitant not of periodic tribal gatherings 
but of lonely vigils. The chronological priority of the 
crowd phenomenon is simply a corollary of Durkheim's 
primary axiom — the invariable precedence of the social 
over the individual. This is, of course, true today since 


the individual of any society finds a ready-made tradi- 
tional system of belief and practice to which he must 
somehow adapt himself ; and it is a merit to have thrown 
the fact into relief. But in speaking of origins the dice 
are no longer loaded against the importance of the in- 
dividual. Not that he can create something out of nothing, 
but that his deviation from the norm may become signifi- 
cant in ushering in innovations, as has demonstrably hap- 
pened in recent aboriginal messianic movements. 

Durkheim ignores this set of facts because he disre- 
gards the immense importance of individual variability 
among simpler peoples. He writes: ''The group displays 
an intellectual and moral uniformity of which we find 
only rare instances among higher societies. Everything 
is common to all. Movements are stereotyped ; everybody 
executes the same ones in like circumstances and this 
similarity of behavior only reflects that of thought. All 
minds are swept away by the same eddies, hence the in- 
dividual merges in the generic type. ' ' ^^ This is precisely 
the fallacy refuted by Boas, Schmidt, Hilde Thurnwald, 
and diverse naive field workers, such as the late Reverend 
Junod. To take down five versions of the same folk tale 
in one community suffices to explode once and for all the 
dogma of psychological uniformity on simpler levels. 

Durkheim 's weakness along these lines naturally ap- 
pears in his treatment of dreams.^^ He criticizes Tylor's 
approach as intellectualistic, but lapses into the same 
fault in a more extreme form. Why, he asks, would a 
native interpret the dream vision of a distant friend as a 
real visit if all he had to do in order to check the conclu- 
sion was to ask whether his camp mates shared the ex- 
perience? "During the same time they, too, have had 
dreams, but they are quite different. They have not seen 
themselves taking part in the same scene; they believe 

is/btd., 7 f. 
i6/6id., 78 sq. 


that they have visited quite different places." Had Durk- 
heim paid more attention to these subjective phenomena 
he would have noted that a native who treats a dream as 
veridical is often stirred to the very depths of his soul. 
The experience is an inmiediately convincing datum not 
in need of corroboration, sharing the ineffableness of 
mystic revelations, too sacred to be communicated to 
one's fellows, let alone to be submitted to logical scrutiny. 

Reverting to Durkheim's cardinal proposition, we 
may further expose his psychological insufficiency by 
citing a legitimate stricture by Goldenweiser." If a crowd 
situation precipitates religious behavior and sentiment, 
why does this not hold for all crowds I What distinguishes 
the secular from the ceremonial American Indian dance ? 
Why does the Australian corroboree remain a form of 
entertainment distinct from initiation and totemic rituals? 
Evidently there is a determinant over and above the mere 
factor of assemblage. 

As to the definition of religious phenomena, any 
writer is, indeed, at liberty to define his terms, hence at 
first blush to identify religion with sacredness may seem 
permissible. What Durkheim does, however, is rather dif- 
ferent. He asserts that all societies dichotomize the uni- 
verse into a sacred and a profane half, the sphere of 
religion coinciding with, that of the *' sacred, i.e. segre- 
gated, forbidden things" {choses sacrees, c'est-d-dire 
separees, inter dites) recognized as such by the community 
of believers. Now such an antithesis is, indeed, reported 
from Polynesia, where ''noa" and "tabu" express pre- 
cisely the antagoijism between secular and holy things. 
But if such formalized dichotomy represents a general 
phenomenon, Durkheim fails to present the evidence. Let 
us add, for the sake of clearness, that he expressly does 
not identify the sacred with the supernatural, mysterious, 

1^ A. A, Goldenweiser, Kistory and Culture, 371. 


unknowable.^^ Ingeniously and speciously he argues that 
the notion of the supernatural can appear solely in con- 
trast to that of a natural order of the universe, such as 
has only arisen recently with the advancement of science. 
Once more his nonpsychological orientation has played 
him a trick. For psychologically what is implied is simply 
that the native irrationally responds to a given experience 
as transcending the ordinary routine : thrilled and awed, 
he labels the source of his emotion by some word describ- 
ing the extraordinariness he feels, and cares not in the 
least about accurately conceptualizing it in the abstract. 

Nevertheless, from another angle, Durkheim's ap- 
proach is not devoid of merit. The traditional restriction 
of religious faith to that in personal beings may well ex- 
clude what is better classed under the same head. As we 
properly consider philosophical rather than religious the 
arid description of divinity furnished by metaphysicians, 
so a response strictly comparable to that otherwise as- 
sociated with gods may be evoked by other phenomena. 
If the Australians consider their rites for the propaga- 
tion of totem animals, in the words of a missionary, ' ' as 
a sort of divine service"; if a Bantu soothsayer refuses 
to sell his set on the ground that it is to him what the 
Bible is to the Reverend Junod, then the ceremonials and 
the divining bones are properly classed as ''religious," 
irrespective of their connection with sentient supernatu- 
ral beings. 

On the other hand, little can be said for Durkheim's 
conception of magic. Right, for the reasons just given, 
in classing many magical procedures under the head of 
religion, he is once more misled by his fatal penchant 
for neat bisection of a logical universe. Magic becomes 
in his eyes an antisocial, churchless mockery of religion. 
"Magic puts a sort of professional pleasure into pro- 
faning sacred things; in its rituals it inverts religious 

^»IMd., 33 sq. 


ceremonial." That, however, such an attitude character- 
izes any considerable number of primitive groups remains 
an unproved allegation. The primitive equivalent of the 
Black Mass, to which Durkheim alludes, is not apparent. 
On the other hand, there are many instances of the very 
same person figuring ambivalently as social shaman and 
antisocial sorcerer, in both instances using precisely the 
same methods and being judged merely by the results.^'' 

So far we have said nothing to explain our initial 
paradox, the enthusiasm for the book and the deprecation 
of its doctrines. Yet the solution is clear, though not ap- 
parent from Goldenweiser's review. Durkheim 's contribu- 
tion lies not in those points on which he piqued himself 
but along quite different lines. It is not pure accident 
that the ethnologists of English speech who have been 
particularly drawn to Durkheim — Eadcliffe-Brown and 
William Lloyd Warner — are field workers and specialists 
in the Australian field. Durkheim misconceived the place 
of Australian culture in historical perspective, but having 
once persuaded himself that it presented the simplest 
form of social life, he immersed himself in the relevant 
material with exemplary assiduity and appraised it w^ith 
critical acumen. When Strehlow, for example, adduced 
facts additional to those published by Spencer and Gil- 
len, some writers were inclined to treat the earlier sources 
as superseded. Durkheim fairly weighed the evidence, 
partly reconciled the contradictions, and reduced them 
to their proper, minor proportions. 

This mastery of the raw material, however, enables 
Durkheim to draw a number of conclusions that, whether 
wholly new or not, were sound and important. Thus, the 
contrast between the routine and the ceremonial phases 
of the annual cycle is real, even if it cannot explain the 
origin of religious feeling. His intimate knowledge of 

19 Ibid., 59 f. Cf. Wm. Lloyd Warner, "The Social Configuration of 
Magical Behavior: A Study of the Nature of Magic," ALK 405-415, 1936. 


the area further leads Durkheim to indicate phenomena 
of wider import. He demonstrates that ritual has a less 
serious aspect, merging into a form of entertainment. 
Again, he clearly illustrates the "pattern" phenomenon 
of American authors : the totemic ceremonials of the sev- 
eral clans differ in detail but there is a core of ideas com- 
mon to all their practices, e.g., the bull-roarers. What is 
more, even the initiation festival and the totemic ceremo- 
nies, otherwise distinct in purpose, share vital formal ele- 
ments. Thus, like Boas, Durkheim recognizes that the 
reasons alleged by natives as underlying their perform- 
ances are mere rationalizations, ritual being primary and 
its ostensible aims secondary. It is probably no accident 
that Durkheim, with the behavioristic outlook stressed 
in the exposition of Les regies, should be most successful 
in the discussion of ceremonial.^ 

Two other points merit attention. Durkheim was 
sufficiently independent in his outlook to admit the native 
origin of the Australian high-god idea when it was still 
hotly contested. Finally, his excellent treatment of sym- 
bolism convincingly shows that a symbol may succeed in 
concentrating upon itself all the fervor that properly 
belongs only to the ultimate reality it represents.^^ 

In summing up Durkheim 's work so far as it bears 
on ethnology, we thus note several positive points. He 
saturated himself with the ethnographical data on the 
areas that interested him, critically sifted them, and 
when unaffected by his special bias, arrived at valid the- 
oretical conclusions. Some of these had been reached by 
others, but Durkheim attained them independently and 
illustrated them by other sets of data. He was a thinker 
who wrestled with general ideas. It is true that his a 
priori conceptions led him astray, and part of his ar- 
gument suffers from scholasticism. But if he sometimes 

20 Jfeid., 406, 540 f., 542 f., 550. 
^Ihid., 415, 314 sq. 


lost himself in arid dialectics, he also penetrated beyond 
the obvious. Marett has well defined the contrast be- 
tween him and a famous British author in their treat- 
ment of relig'ion: ''Sir James Frazer's method is simply 
to ask whether a given body of associated beliefs and 
practices is signalized by the presence of a certain con- 
cept. But M. Durkheim's method, going deeper, considers 
whether it fulfils a certain social and moral function." 
And the comparison concludes with the ironic comment; 
''Sir James Frazer's principle of classification has at 
least this virtue — that it is not over-subtle. ' ' ^" We have 
pointed out that part of Durkheim's doctrines coincided 
with those of Boas. But that does not detract from his 
historic significance, for we have seen that Boas' prin- 
ciples are not easily disengaged from his writings. Be- 
sides, minds to which Boas' procedure is radically 
uncongenial cheerfully accept the very same ideas when 
presented as part of a coherent system. 

Durkheim's Followers 

The scholars whom Durkheim rallied round his ban- 
ner adhere so closely to his principles that a detailed 
exposition is unnecessary. Nevertheless, several of their 
productions cannot be passed over in silence. 

Probably the most influential of these is Henri Hu- 
bert and Marcel Mauss' "Esquisse d'une theorie gen- 
erale de la magie, ' ' ^^ whose conclusions, indeed, already 
figure in their master's work. Its erudition and insight 
have been acknowledged by writers of other schools.^* 
With Schmidt, however, we must repeat the stricture 
that the sociological school, here as elsewhere, fails to 
recognize the influence of the individual. This is all the 
stranger because under magic the authors include sha- 

22 R. E. Marett, Psychology and Foil-lore, 188 f ., London, 1920. 

^^L'Annee sociologique, 7:1-146, 1904. 

2*W. Schmidt, Der Ursprung der Gottesidee, 1:514 sq., 1926. 


manism,^*' which so clearly reflects the impress of power- 
ful personalities and of individual differences. If it is 
true that "the magician is normally a sort of maniac" 
and only a limited number of people in a community are 
subject to such experiences, it is absurd to argue that 
public opinion creates the magician. As with other cul- 
tural data, we have to face the interaction of society and 
the individual, and no one can foresee the scope of the 
latter 's influence on events in a given situation. 

Hubert and Mauss deserve credit for their admir- 
able critique of the overintellectualistic interpretations 
of their predecessors. They rightly contend that without 
mysticism magic simply merges in science ; that this mys- 
tical factor is never lacking in magical belief. It can never 
be reduced merely to the laws of association, which the 
magician operates and limits in accordance with his mo- 
mentary wishes : he blinds a frog to transmit its blindness 
to the victim, but does not sympathetically transform 
him into a frog. In other words, he does not put his trust 
in the indefinite automatic functioning of trains of as- 
sociation once set in motion. 

Successfully avoiding the error of confounding 
magic and science, Mauss and Hubert fail properly to 
distinguish magic and religion. They are hindered by an 
initially assumed antithesis between these two sets of 
phenomena, religion being conceived as social, magic as 
typically antisocial or asocial. Not that the two neces- 
sarily incarnate these contending principles, but that es- 
sentially magic tends towards evil sorcery.^^ But since 
shamanism, included by our authors under magic, con- 
stitutes the very essence of innumerable religions, the 
artificiality of this polar contrast is obvious. Actually, 
they conclude by treating mana, or impersonal magic 
power, as the source of religion no less than of magic. At 

25 Op. cit., 22, 28, 30, 35, 37, 94. 
^Ibid., 17 et passim. 


bottom, then, there is no such sharp demarkation. To 
illustrate the crowd origin of magic, they cite the war 
dance of Dyak women, interpreted by the actresses into 
efficient collaboration with their actually fighting hus- 
bands and brothers. Thus, we learn, mana is born. ''On 
the other hand, we do not detect in their spirit this defi- 
nite notion of sacred things which is the sign of the re- 
ligious state."'" But how common is such a precisely 
defined notion of sacredness in religion? And why do 
some social gatherings precipitate a notion of mana, and 
others of sacredness? 

As is usual with the members of their school, Hubert 
and Mauss evolve excellent ideas so far as their thinking 
is not overshadowed by sociological bias. They recognize 
not merely emotion and desire, but unconscious ideas; 
they realize that magic presents regional differences, that 
a given society has a restricted number of ritual forms 
— in other words, tribal patterns; they properly empha- 
size the importance of incantations, rites oraux, comple- 
mentary to the generally recognized rites manuels.^^ 

A very different contribution is represented by 
Mauss' "Essai sur les variations saisonnieres des soci- 
etes Eskimos; etude de morphologic sociale."^ This 
admirably documented paper defines the sharp dualism 
of Eskimo life, with its alternations of caribou hunting 
and the quest of sea mammals. In correlating these 
economic aspects of Eskimo life with social and religious 
seasonal differences, Mauss produces a striking "func- 
tional" picture and is suggestive in the truest sense of 
the term, i.e., in suggesting parallel inquiries on other 

Finally, we must consider Mauss' "Essai sur le don; 
forme et raison de I'echange dans les societes archa- 

27/6id., 137 f. 
28 lUd., 51, 53, 57, 116. 

^^L'Annee sociologique, 9:39-132, 1906. The late H. Beuchat assisted 
in the preparation of this memoir. 


iques."^° It had been one of Durkheim's favorite ideas 
— in contrast to many of his contemporaries — to stress 
the quality of the documentation rather than the number 
of societies compared. ''The essential thing," he wrote, 
*'is to unite not many facts, but facts at once typical and 
well studied. ' ' ^^ This admirable principle guides Mauss 
in his study of primitive gifts no less than in his earlier 
essay on Eskimo seasonalism. On the primitive level, he 
argues, gifts are rendered not freely but as an obligation, 
and there is an obligation likewise to accept the proffered 
present. Further, there is not an individual exchange of 
goods, but one between clans, tribes, or families; and 
what is exchanged represents not purely economic utility 
but a whole system of courtesies, rites, feasts, military 
services, all of which Mauss groups together under the 
head of prestations totales. 

The familiar potlatch of British Columbia, in which 
the contracting parties vie with each other in quest of 
prestige, forms a competitive subtype of this category, 
the type agonistique, which also includes Melanesian 
practices, while elsewhere occur intermediate forms be- 
tween such rivalry and the simple obligatory exchange 
of gifts. Mauss cites Polynesian parallels, hitherto 
neglected, as illustrating the entire potlatch theme minus 
its exaggerated rivalry. For Oceania, generally, he 
argues that though the exchange of goods plays a large 
part, the forms and rationale differ from ours: there is 
no purchase, no sale, no barter in the strict sense of these 
terms. These forms necessarily involve the ideas of credit 
and honor so marked in the transactions of British 
Columbia Indians. The principle of the exchange-gift 
is treated as representing a definite evolutionary stage 
between that of total group prestation and the late stage 
of purely individual contract.^^ 

^^L'Annee sociologique, nouvelle s6rie, 1:30-186, 1925. 
"76id., 4:341, 1901. ^2 Mauss, op. cit., 126. 


While the Diirklieimian propositions as to primitive 
clans and the dominance of society reappear in this 
treatise, they are not obtrusive. On the other hand, Mauss 
duly stresses the irrational, nonutilitarian motives under- 
lying primitive negotiations that take the place of our 
business transactions. With Thurnwald and Malinowski 
he thus becomes one of the leading students of primitive 
economics. Finally, the concluding section of the essay 
explicitly demands the study of cultures as integral 
wholes, indicating that ''economic" facts have their 
social, religious, aesthetic aspects. The point had, indeed, 
been made before, but it is none the less sound when 
voiced not merely as a mystical shibboleth but with 
concrete illustrations. No wonder this timely production 
has met with a more generally cordial reception than 
have most works of the school.^^ 


Lucien Levy-Bruhl (1857- ) is even more defi- 
nitely identified with philosophy than are the preceding 
French authors. He has published a treatise on the 
History of Modern Philosophy in France, as well as 
monographs on individual thinkers — Comte and Jacobi. 
Naturally he is attracted by the primitive equivalents of 
such concepts as causality and the logical law of contra- 
diction. Quite intelligibly, too, his interest takes a some- 
what different direction from that of Durkheim and 
Mauss. Perhaps as widely read, Levy-Bruhl does not 
evince that intensity of concern with particular regions 
which we recognize, e.g., in Mauss' essay on the Eskimo 
cycle. In other words, his orientation is less ethnographi- 
cal, notwithstanding the wealth of his documentation; 
and though his main works were issued as Travaux de 

^ See e.g. Eichard Thurnwald, Die menschliche Gesellschaft, 5 : 48 f ., 
Berlin u. Leipzig, 1934, 


VAnnee sociologique, it is not clear to what extent he can 
be reckoned a full-fledged member of the school. 

Levy-Bruhl's two most important and mutually 
complementary books, Les fonctions mentales dans les 
societes inferieures (Paris, 1910; 5 edition, 1922) and 
La mentalite primitive (Paris, 1922), certainly share in 
part Durkheim's position. Though their aim is to define 
the primitive mind, this is expressly and emphatically 
not attempted by leaning on psychology. Levy-Bruhl, 
like Durkheim, ignores the individual mind and deals 
only with "group ideas" {representations collectives). 
In harmony with the leader of the French sociologists, 
he thus spurns all explanations of primitive attitudes 
and actions that treat their originators as rational beings 
seeking an explanation. In conscious opposition to Tylor, 
he must be placed with Tarde and Boas as among those 
who allow full scope to the emotional side of primitive 
man. He illustrates the importance of the affective factor 
by a wide range of data and not infrequently arrives at 
important reclassifications, as when he shows convinc- 
ingly that the primitive concept of the soul is by no 
means so simple and uniform as the followers of the 
animistic theory assume; and he rightly treats the 
couvade not as an isolated instance of aboriginal bizar- 
rerie, but as merely one of a large set of taboos imposed 
on both parents at childbirth. Again, he demonstrates 
very clearly the nature of mystic numbers; if Malays 
speak of seven souls for each individual, it is not that 
they actually distinguish seven aspects of a spiritual 
essence; *'It is, on the contrary, because in their eyes 
the number seven enjoys pre-eminently mystic virtues, 
becoming a kind of category, to which conform not only 
their magical operations, but also their representations, 
including those of the soul." ^* 

Thus, Levy-Bruhl forcibly brings home to his read- 

34 Les fonctions mentales . . . , 83-93, 296-302, 250. 


ers the way in which a pre-existing norm determines 
individual thinking. Like other members of the school, 
he also displays a sense of cultural dynamics, as when he 
discusses the possible alterations in the native interpre- 
tation of funeral rites or the tortuous route of mythologi- 
cal development.^" A less attractive feature shared with 
Durkheim is the naive evolutionary assumption that 
Australians are the low^est of known peoples.^® Still more 
objectionable is the ever-recurring use of the term 
*4aw," a veritable idee fixe of the French sociologists. 
For Levy-Bruhl believes that he can summarize the 
representations collectives by the "law of participation," 
to which all of primitive mentality conforms. Now it is 
not easy to define this law because the author himself 
nowhere furnisbes a clear-cut definition. But by piecing 
together statements in different parts of his works we 
can make his meaning sufficiently clear. Primitive people 
perceive nothing as we do, conceive nothing as we do, 
dispense with the principles of contradiction and of 
causality. Their thinking is prelogical, though not on 
principle antilogical, in that it is a matter of indifference 
whether the law of contradiction holds or not. In other 
words, the primitive mentality makes an inseparable 
jumble of logical and nonlogical procedures. What, how- 
ever, is the positive meaning of "participation"? Ac- 
cording to our author, logically distinct aspects of reality 
tend to fuse into one mystic unity. An Australian horde 
does not own its hereditary land, for it cannot conceive 
separation from it. "Between its members and this 
locality there is a mutual participation: it would not be 
what it is without them, nor they without it." Similarly, 
a South American Indian who declares himself a parrot 
means precisely that, viz., an inexplicable mystical iden- 
tity of himself and the bird. In primitive hunting, what 

85 Ibid., 383-391, 439 f. 
36 E.g., ibid., 329. 


is essential is not the practical pursuit of the game but 
the magical appurtenances of the chase: ''What is es- 
sential are the mystical operations that can alone ensure 
the presence and capture of the quarry. If they are 
lacking, it is not worth while trying." This trend culmi- 
nates in the group ideas of death. For, in such activities, 
as the chase, a minimum of rational adjustment of means 
is essential to attaining the end sought. Into the notions 
about the dead this logical factor does not intrude, hence 
prelogic here runs riot untrammeled. In his resume Levy- 
Bruhl stresses the point that the very term representa- 
tion is inaccurate in speaking of group ideas. The 
primitive mind does more than represent its object: **It 
possesses it and is possessed by it. It participates in it, 
not merely in a representative, but simultaneously in the 
physical and mystical sense of the word. It does not only 
think but live it."'' 

Even in his earlier work Levy-Bruhl ascribes only 
infinitesimal significance to the notion of causality among 
simpler societies. In his later treatise he defines the 
equivalent among them as indifferent to secondary 
causes: ''The connection between cause and effect is 
immediate. Intermediate links are not admitted, or at 
least, if recognized, are regarded as negligible and re- 
ceive no attention. ' ' ^^ 

In these discussions we accept the emphasis on ir- 
rational associations of ideas on ruder levels, the view 
that a merging of logically irrelevant notions precedes 
their analytical recognition as so many distinct phases 
of reality — incidentally a favorite point of Boas. But we 
resent the term "law" because Levy-Bruhl nowhere 
provides a clue to the nature of the associations expect- 
able in a given situation. Why does the "law of partici- 
pation" make the Malay invest with a mystic halo the 

'"'Ibid., 77, 130, 242, 263, 332, 334, 355, 378, 426 f. 
^^ Ibid., 78. La mentalite primitive, 92, 518. 


number 7, while this same "law" makes the Pueblo 
ascribe a similar significance to 4? Why does this princi- 
ple precipitate the full-fledged couvade in South America 
and merely mild dietary and other taboos in California? 
But, waiving the point, we find serious objections to 
Levy-Bruhl's basic conception of the primitive mind — 
objections we share with such diverse thinkers as Pinard 
de la Boullaye, Schmidt, Goldenweiser, and especially 
Thurnwald, whose masterly critique we indorse in every 
point.^ In the first place, Levy-Bruhl roams over the 
whole primitive world and then presents a composite 
picture of the ''primitive" mind, a category that is made 
to embrace even the highly sophisticated natives of China 
and India. Second, there is a complete neglect of individ- 
ual variability in any one society; yet every society 
includes not only those who follow along the predeter- 
mined paths of tradition, but also leaders who in part 
break away from the past and found new "group ideas." 
Third, primitive man, irrational though he often appears 
in his abstract formulations and the religious phases of 
culture, is often as keen an observer as civilized man, 
and as logical a reasoner from his observations. Intermit- 
tently, Levy-Bruhl himself admits this, but he tries to 
argue the fact away by ascribing aboriginal skiU to a 
sort of intuition such as guides the billiard player "who 
without knowing anything of geometry or mechanics, 
without requiring reflection has gained a rapid and sure 
intuition of the movement to be executed for a given 
position of the balls." *° But, as Thurnwald notes in sup- 
porting Nieuwenhuis' observations in Borneo by his own 
in Melanesia, the longer one associates with "savages," 
the more intimate one's acquaintance with their lan- 
guage, the more do the differences initially felt between 
their mode of thinking and ours tend to disappear. The 

38 E. Thurnwald, Deutsche Literaturzeiiung, 486-494, 1928. 
*<* L6vy-Bruhl, La mentalite primitive, 92, 518. 


** savage" thinks rationally as we do in comparable 
situations, that is, he uses the rule-of-thumb logic that 
suffices for everyday use. What Levy-Bruhl describes as 
the ''law of participation" is the common foible of all 
humanity, not the peculiarity of primitive minds. He 
establishes his contrast not by comparing civilized and 
primitive man, but, in Thurnwald's apt characterization, 
"the highest achievements of the modern intellect" — 
nota bene, only in its professional activities — "with a 
rather vague 'primitiveness.' " 

The net result of Levy-Bruhl's discussions — apart 
from some of the fruitful corrections of earlier classifi- 
cations mentioned above — is to bring home once more 
the importance of social tradition in molding individual 
responses to experience and to stress the overwhelming 
significance of irrational factors not in primitive, but in 
human thought. 


Though an Englishman trained at Cambridge by 
Haddon and Rivers, A. R. Radcliffe-Brown (1881- ) 
is best considered with the Durkheimians, whose influ- 
ence he freely acknowledges. He has lived in the South 
Seas, taught at the universities of Sydney, Cape Town, 
and Chicago, and since 1937 has been professor of social 
anthropology at Oxford. In contrast to Mauss and Levy- 
Bruhl, he has had abundant contacts with primitive men, 
yet the resulting difference is less than might be sup- 
posed. For Radcliffe-Brown is most emphatically not a 
field man by temperament. The only complete descrip- 
tion of a tribe he has ever attempted is based on his 
maiden trip, which took him to the Andaman Islands 
(1906-1908). His report, unduly delayed by the War, is 
creditable enough, but in no way remarkable as an 
increment of factual knowledge if compared with the 
account of his predecessor, E. H. Man. Unlike the mono- 


graphs of other modern investigators, it lacks the per- 
sonal touch: statements are consistently offered in a 
generalized form without reference to individual inform- 
ants' experiences. This procedure, as we shall see, reflects 
more than merely a novice's deficiency in technique. 
Later Radcliffe-Brown did significant work in Australia, 
but none of his relevant writings touch more than 
selected phases of native life. Of original investigation 
among the natives of Polynesia and South Africa as a 
result of prolonged residence in these areas there is no 
record. Paradoxically, this widely traveled scholar is 
at heart an armchair anthropologist who formulates 
problems in the study to be solved in the field by his 
followers. This is said by way of characterization, not of 
criticism, for he has thus stimulated important work by 
others. Indeed, the meagreness of his total output is 
intelligible from the compensatory energy he has ex- 
pended on teaching and the organization of research. 

Radcliffe-Brown 's printed work divides into two 
categories : a series of confessions of faith outlining the 
proper procedure of anthropological research; and his 
actual contributions, which by no means wholly harmo- 
nize with the spirit of his manifestos. 

In his general pronunciamentos " Radcliffe-Brown 
clearly and avowedly stems from Durkheim. ''Social 
anthropology" or "comparative sociology" — the study 
of group behavior — is independent of psychology and 
ignores as irrelevant the individual as an individual. Its 
aim is to discover laws. Since history explains its phe- 
nomena only by "finding wherever possible the particular 
cause or occasion of each change that has taken place," 

"A. R. Radeliffe-Brovrn, "The Methods of Ethnology and Social 
Anthropology," South African Journal of Science, 20:124-147, 1923; "The 
Present Position of Anthropological Studies," Brit. Assoc. Adv. Science, 
Section H, Presidential Address, 1-32, 1931; "Applied Anthropology," 
Australian and New Zealand Association Adv. Science, Section F, Presi- 
dential Address, 1-14, 1930. 


it cannot yield the generalizations current in natural 
science, hence at best the historical method is of com- 
paratively slight value; in proportion as it becomes 
conjectural it approaches worthlessness. Where definite 
historical data are extant as to the origin of an institu- 
tion, that knowledge may indeed be most significant for 
social anthropology, but actually the historical ethnolo- 
gists provide only ''a few, a very few" such facts. The 
social anthropologist, on the other hand, observes and 
explains customs and beliefs by showing *'how each one 
of them is an example of some general law of human 
society." To elucidate totemism, for instance, he dem- 
onstrates it as ^'a, special instance of a phenomenon or 
at any rate of a tendency which is universal in human 
society." Finally, when laws are once determined, they 
can be applied practically to regulate the course of social 
development, in parallelism with the procedure of other 

On another essential point Radcliffe-Brown agrees 
not only with other Durkheimians but also with Boas, 
and, as we shall see, with Thurnwald and Malinowski. 
Any particular culture is ''normally a systematic or 
integrated unity in which every element has a distinct 
function." This implies skepticism, again shared with 
Boas, as to the comparability of single traits from dif- 
ferent areas. Specifically, Radcliffe-Brown opposes the 
"atomic" view of studying the distribution of isolated 
features apart from the context that gives meaning to 
them. He is further alive to the process of selective 
borrowing, thus repudiating the notion that contact 
automatically precipitates diffusion. 

As in the case of other writers, Radcliffe-Brown 's 
real achievement lies in rather different directions from 
those suggested by himself. This herald of the "system- 
atic unity" of cultures has not essayed a single integrated 
cultural picture since his avowedly immature treatise on 


the Andamaiis.'" And tlie **laws" lie propounds must be 
repudiated. The following is a sample: "Things that have 
important elfects on the social life necessarily become 
the objects of ritual observances (negative or positive), 
the function of such ritual being to express, and so to 
lix and perpetuate the recognition of the social value of 
the objects to which it refers." From this principle our 
author infers that hunter-gatherers will perform rituals 
concerning their game animals and edible plant species; 
that where there are clans, the clans will perform 
generically similar but specifically distinct rituals of this 
type, that clanless societies will not have such totemic 
ceremonials but will rather display ''a general undif- 
ferentiated relation between the society as a whole and 
the world of nature as a whole. " *^ In the first place, 
these propositions remain quite as unverified as the 
historical hypotheses against which the author inveighs. 
Secondly, the primary generalization offered is a trite 
statement of certain descriptive facts. It does not even 
pretend to explain those very important rituals which 
do not purport any bearing on the food supply. Certainly, 
hunters execute rites to ensure capture of the game, but 
the Plains Indians also performed the Sun Dance, their 
most conspicuous ceremony, for a variety of other 
reasons, such as revenge, securing the recovery of a sick 
relative, and so on. A law of ritual would have to define 
what species would figure ceremonially; under what 
conditions nondietary factors determine ritual; why 
frequently it is not the clan at all but some other unit 
that conducts ceremonies; and a dozen other things. 

The grandiloquent use of the term ''law" is most 
regrettable and in some circumstances leads to absurdity, 
as when Radcliffe-Brown writes of "a universal socio- 

*2 The Andaman Islanders : A Study in Social Anthropology, Cam- 
bridge, 1922, 

*3 < ' The Methods of Ethnology and Social Anthropology, ' ' 135 f. 


logical law though it is not yet possible to formulate 
precisely its scope, namely that in certain specific condi- 
tions a society has need to provide itself with a segmen- 
tary [clan] organization."" Whoever heard of a 
universal law with an as yet undefinable scope, of a law 
that works in certain specific hut unspecified conditions! 
Is it a law that some societies have clans, and others have 
not ? Newton did not tell us that bodies either fall or rise. 

Like other authors, Radcliffe-Brown achieves his 
most solid work when he forgets about his abstract 
profession of faith, immerses himself in a set of data, 
and extricates sense out of chaos irrespective of any 
doctrinaire principles. Thus, he points out the highly 
important fact that African ancestor-worship is patri- 
lineal, regardless of matrilineal descent ; and he explains 
most cogently why the Northern Thonga of Portuguese 
East Africa use a single word for the maternal uncle, 
the maternal uncle's son, and the mother's father. By 
native usage, it is my maternal grandfather that sacri- 
fices on my behalf to my matrilineal ancestors; on his 
death the duty devolves upon my maternal uncle, and 
after his death, upon his son. Representatives of three 
distinct generations thus share an important ceremonial 
function and are appropriately designated by a single 
term."'* Nothing could be better than this interpretation. 
In the same article he clarifies the concept of avuncular 
authority, showing that it may be counterweighted by 
the status of the paternal aunt. 

Most important of all, however, is Radcliffe-Brown's 
contribution to Australian social organization. His field 
work led to a brilliant discovery, viz., that cross-cousin 
marriage proper invariably accompanies one type of 
kinship nomenclature, while cross-cousin marriage of the 

***'The Social Organization of Australian Tribes," Oceania Mono- 
graphs, No. 1, p. 109, Melbourne, 1931. 

""The Mother's Brother in South Africa," South African Journal 
of Science, 21:542-555, 1924, esp. pp. 552, 554. 


second degree goes with anotlier type/" This is infinitely 
better than a sham law, for it is a real, verifiable cor- 
relation, revealing a connection hitherto unguessed be- 
tween two sets of isolated facts. 

It is in this field, too, that Radcliffe-Brown marks 
the most definite advance over the Durkheimian creed. 
He breaks with the dogma of Australian ultra-primitive- 
ness; he accepts the family, not the clan, as the basic 
social unit ; he no longer assumes the primeval character 
of totemism, nor the necessary priority of matrilineal 
descent. Apart from these deviations from French socio- 
logical doctrine, he establishes a well-defined series of 
concepts to cover Australian phenomena, summarizing 
the essential facts with exemplary terseness and a sane 
appraisal of all sources extant. 

Finally, Radcliffe-Brown 's theoretical intransigence 
on the subject of history wanes before data with which 
he is thoroughly familiar and, notwithstanding some 
qualms, he stoops to chronological hypotheses. The 
Yaralde kinship system "cannot reasonably be supposed 
to have developed independently of those [Aranda sys- 
tems]. ... we must certainly assume some historical 
connection between them. ' ' Again, the Kumbaingeri type 
is a stepping stone from the Kariera to the Arunta 
f orm.*^ Surely this is conjectural history ! 

One aspect of Radcliffe-Brown 's historical thinking, 
rather adumbrated than fully limned, is his notion of 
cultural evolution. It differs from Morgan's scheme in 
renouncing a universally valid sequence, but it does imply 
definite social trends, notably the tendency of wider 
integrations to supersede those of narrower scope.'*® 
This neo-evolutionary phase of Radcliffe-Brown 's philos- 

«" Three Tribes of Western Australia," JEAI 43:143-194, 1913. 
*7"The Social Organization of Australian Tribes," 51, 63, 113, 120. 
■^Ibid., 113-120. 


ophy of culture suggests ideas broached by Thurnwald 
and Schmidt. 

Two recent utterances clarify Radcliffe-Brown's 
present position,*^ especially his conception of ''func- 
tion," which he defines as *'the contribution which a 
partial activity makes to the total activity of which it is 
a part." He explicitly concedes that the implied unity of 
any social system is a mere hypothesis to be verified, a 
conclusion in which we heartily concur. In other words, he 
supplants the dogma that ' ' everything in the life of every 
community has a function ' ' with the legitimate statement 
that ' ' it may have one, and that we are justified in seeking 
to discover it." He admits that at the present time it is 
impossible "to establish a purely objective criterion" for 
the degree of functional unity of a particular society, 
but merely hopes for greater enlightenment in the future. 
Further, Radcliffe-Brown rejects the dogma that "there 
are no discoverable significant sociological laws." 

In this form the doctrine is much more acceptable, 
for what is inadmissible as a set of axioms may remain 
a legitimate programme. // every item of culture has a 
function, if comparative sociology has valid laws to offer, 
this will be of great interest to all ethnologists. In the 
meantime we take cognizance of the message and shall 
watchfully lie waiting for what may come in its wake. 

Of more immediate interest, however, are two other 
points. Radcliffe-Brown now overtly recognizes study of 
the individual as "an essential part of the task"; and 
historical explanations are now regarded as complemen- 
tary to those of the functionalist order, though naturally, 
authenticated events are preferred to merely inferential 
ones. "A sociologist who neglected [documented] history 
. . . would be gravely at fault"; and in accordance with 
this principle a member of the school has recently 

*»"0n the Concept of Function in Social Science," AA 37:394 sq., 
1935; "Kinship Terminologies in California," AA 37:530 sq., 1935. 


examined the kinship structure of Southeastern Indians 
as recorded at different periods."^" It appears that we 
not only may but ought to study the changes going on 
before our eyes: a ''synchronic" approach must be com- 
bined with a "diachronic" one. The only comment 
required is that this thoroughly sound principle is not 
new. Fifteen years ago Dr. Elsie Clews Parsons collated 
data gathered among the Hopi in 1883, 1890, and 1920, 
respectively, concluding her diachronic survey with these 
words : ' * Here under our eyes has gone on an immensely 
interesting process of cultural change of which we have 
as yet but the barest record — to so many of us study of 
the past is so much more appealing than study of the 
present, even the present in which the past repeats itself, 
in terms clearer and more pregnant than archaeology 
can ever use." Parsons' researches on the Pueblo are 
invaluable precisely because they transcend the static 
view of a culture often presented by investigators and 
show exactly what did happen, inevitably revealing not 
only a chronology but also determinants of social be- 
havior. Thus, when a Hopi chief died in 1892 there was 
no automatic succession to office by the theoretical heir. 
Two genealogically possible claimants were frowned upon 
because of their foreign wives ; some candidates aroused 
opposition because of their fraternity membership ; and 
finally the hierarchy installed a reluctant grandson of the 
late chief's eldest sister.^^ 

We note, then, with satisfaction the far-reaching 
agreement of Radcliffe-Brown with other workers. 

To sum up, Radcliffe-Brown is a relatively independ- 
ent offshoot of the Durkheim philosophy who does not 
differ nearly so much from his fellow workers of other 

5° Fred Eggan, ' * Historical Changes in the Choctaw Kinship System, ' ' 
AA 39:34-52, 1937. 

31 E. C. Parsons, "Contributions to Hopi History," AA 24:253-298, 
1922; eadem, (editor), Hofi Journal of Alexander M. Stephen, 1047, New 
York, 1936. 


schools as his earlier utterances might lead one to sup- 
pose. While he is not primarily interested in the recon- 
struction of the past, his practice in a modest way belies 
his theory as to the uselessness of even conjectural 
history. Topically, he has concerned himself almost 
exclusively with problems of social organization, which 
he has advanced in several important respects, while 
keeping abreast of contemporary progress. His earlier 
ukases as to what social anthropology ought to do we 
greet with a shrug ; what he has actually done, restricted 
as it is in scope, we welcome with genuine respect. 




Bronislaw Malinowski (1884- ), professor of 
social anthropology in the University of London, is often 
paired with Radcliffe-Brown. He himself has not rejected 
the classification: Radcliffe-Brown's "tendency to ignore 
completely the individual and to eliminate the biological 
element from the functional analysis of culture" is said 
to constitute "really the only point of theoretical dissen- 
sion" and the only one on which Durklieim's princi- 
ples require supplementing/ Notwithstanding, however, 
Radcliffe-Brown's recent inclusion of the individual as a 
legitimate object of inquiry, the resemblances between 
the two English scholars should not be overrated, and we 
have deliberately divorced them in our treatment. They 
share with each other — but also with Boas, Bachofen, and 
Fustel de Coulanges — a concern with the interrelation of 
the several elements within a given society. Further, both 
have avowed a disdain, largely but by no means uniformly 
indulged in practice, for history. In every other respect 
their lifework diverges radically. For the difference con- 

1 Introduction to H. Ian Hogbin, Law and Order in Polynesia, 
xxxviii, New York, 1934. 



ceded by Malinowski bears not on a casual division of 
opinion but on a chasm between two distinct person- 

In contrast to Radcliffe-Brown, Malinowski is first 
of all a field investigator. His intercourse with savages 
has indeed been less varied, but those three expeditions 
of his to New Guinea, which included a two years' stay 
in the Trobriand Archipelago, marked an epoch in his 
life. All his subsequent publications, descriptive and 
theoretical, stand rooted in that major experience. His 
field technique conforms to Boas' standards: he learnt 
his Trobrianders ' tongue, tried to live their life, garnered 
concrete rather than abstract statements from his in- 
formants, and recorded them in the vernacular. Rivers 
had approached natives as an outsider; Malinowski re- 
formed British methods by stressing the unformulated 
phases of aboriginal culture, *'the imponderabilia of 
actual life and of typical behavior." As Boas had insisted 
on registering the exoteric no less than the esoteric side of 
primitive communities, so Malinowski noted as equally 
significant the norm and the deviations from it. Endowed 
with an unusual literary sense, he thus succeeded in 
creating a ''flesh and blood" picture of his Melanesians. 
For the traditional systematic monograph he has sub- 
stituted a series of books, each devoted to a central 
theme, which is exhibited in its relations to the tribal 
life as a whole. Thus, an account of ceremonial barter 
merges in considerations of the traders' canoes, of 
property rights, class distinctions, rules of inheritance, 
magic. Yam cultivation emerges as the basis of wealth, 
power, and law and as inextricably bound up with magic. 
Inevitably there is repetition, but the reader becomes 
saturated with the Trobriand atmosphere, sees the 
aborigines as human beings, not as puppets designed to 
produce kinship nomenclatures or to illustrate some 
sociological law. Most important of all, this picture is not 


conveyed by rank impressionism: subjective as the in- 
terpretations in part appear, there is an estimable mass 
of textual material — interlinear renderings of native 
statements eked out by free translations.^ 

Immersing himself into the native scene, Malinowski 
is dominated by it, and his theoretical tenets largely 
emanate from direct observation. In justice we must note 
that even before his trip to the Trobriands he had 
grasped the basic character of the family from a library 
study of Australian sources, thus admittedly anticipat- 
ing Radcliffe-Brown's conclusions.^ This idea was, how- 
ever, intensified and amplified by Trobriand life; and 
allying himself with a host of modern writers, such as 
Westermarck, Boas, Swanton, Schmidt, and Kroeber, 
Malinowski came to regard the family as the funda- 
mental unit in all human society.* 

It was field experience that led Malinowski to one of 
his characteristic doctrines — the conflict of practice with 
theory. Here he was preceded by Boas, yet many recent 
authors were still speaking of primitive law as function- 
ing with automatic precision. Malinowski by striking 
case material illustrated how individual natives chafe 
under social restraint to the point of defying tradition. 
Specifically, he has again and again emphasized the 
Trobriand father's predicament in trying to reconcile 
paternal love with the matrilineal law that favors his 
sister's son to the detriment of his own son.'' In the ensu- 
ing conflict of loyalties the decision will hinge on the 
parent's personality — the strength of his character and 
of his fatherly sentiments. Such individual differences, 

^Argonauts of the Western Pacific, London, 1922. Crime and Cusfom 
in Savage Society, London, 1926. The Sexual Life of Savages in North- 
western Melanesia, New York, 1929. Coral Gardens and their Magic, 2 
vols., New York, 1935. 

3 Br. Malinowski, The Family among Australian Aborigines, London, 
1913. Radcliffe-Brown, The Social Organization of Australian Tribes, 103. 

* Sex and Repression in Savage Society, 243. 

" See especially, Crime and Custom in Savage Society, 101 sq. 


then — ignored on principle by the Durkheimians — not 
only help us understand what happens in special situa- 
tions but reveal factors capable of overthrowing the 
social tradition itself and of producing its successor. 

Again, the peculiar network of mutual obligations 
typical of Melanesia led Malinowski to herald reciprocity 
as a basic principle of human society: ''The duty of one 
person is inevitably the privilege of another; services 
rendered are boons received ; gifts and tributes presented 
by one side can be demanded by the other." ° From this 
set of relationships he further inferred the precedence 
of civil over criminal law in society generally. 

Observations in the Trobriands further made Mali- 
nowski reject certain fallacies about primitive man's 
economic behavior and helped him clarify the non- 
utilitarian side of economics. Hahn had already demon- 
strated the religious and sportive facets of animal hus- 
bandry. Malinowski widened the outlook by exhibiting 
the motives that underlie production and exchange 
among his people, and analyzed such concepts as labor, 
wealth, money, and value in correspondence with savage 
realities. Value turned out to rest not on utility alone or 
on utility plus rarity, but rather on the fancy character 
of a product, on which the craftsman lavishes a dispro- 
portionate amount of effort. Correspondingly, the elabo- 
rate ceremonial barter of the natives revolves about the 
irrational transfer "of two meaningless and quite use- 
less objects." And, as our rationalistic notions cannot 
be transferred to Trobriand economics, so the division 
of labor there assumes a quite distinct meaning.'' 

No doubt the avowed core of Malinowski 's philoso- 
phy of culture, his ' ' f unctionalism, " is in one of its 
aspects also the reflection of experience in Melanesia, 

^ E.g., introduction to Hogbin, op. cit., xxxiii sq. 

''Argonauts of the Western Pacific, esp. 49-104, 166-194. Coral Gar- 
dens, 21, 41. 


where the interlocking of a priori disparate phases of 
activity, such as magic and industry, are insistently 
brought home to a sensitive observer. 

As a positive achievement stimulated by reading 
rather than direct observation may be registered Mali- 
nowski's synthesis of psychoanalytic concepts with his 
ethnographic findings. We have previously noted the 
sterility of Rivers' attempts in this direction. Malinowski 
goes further, broaching a new and legitimate problem. 
Granting that the t}q3e of family known in our civiliza- 
tion precipitates repressions and conflicts, may we as- 
sume identical psychological situations as a universal 
human phenomenon? Malinowski answers in the nega- 
tive, pointing out most suggestively that in a matrilineal 
community of the Trobriand type the sentiment directed 
against the father would become an avuncular complex. 
Right or wrong, he has here paved the way for investi- 
gating a possible correlation — the nexus between a type 
of institution and its psychiatric concomitants.^ 

With Malinowski, as with Radcliffe-Brown, an esti- 
mate of achievement must not be warped by restiveness 
over apocalyptic utterances on points of principle. In 
messianic mood Malinowski is forever engaged in two 
favorite pastimes. Either he is battering down wide open 
doors; or he is petulantly deriding work that does not 
personally attract him. From first to last he intermit- 
tently taunts the antiquarian ' ' gloating over isolated and 
outlandish anomalies of hmnan behavior." Yet from Mc- 
Lennan and Tylor to Spier and Birket-Smith not a single 
professional of standing has treated stray items of be- 
havior or craftsmanship as anything but a means to an 
end. In the same spirit Malinowski thumbs his nose at 
technology, flouts distribution studies, sneers at recon- 

8 ' ' Mutterrechtliche Familie und Odipus-Komplex ; eine ethnologisch- 
psychoanalytische Studie," Sonderabdruck aus Imago, X Band, Leipzig, 
Wien, Zurich, 1924. Also, Sex and Repression, 135 sq. 


struction of the past. The only worthy aim is to study 
' ' the part which is played by any one factor of a culture 
within the general scheme." At times the more elusive 
aspects of social life are made to loom large. Malinowski 
goes so far as to suppress his own data from islands 
where a brief sojourn precluded ideal intensiveness of 
investigation. As though a naturalist would fail to re- 
port the existence of the okapi because he had been un- 
able to trace its embryonic development." 

In short, Malinowski 's functionalism is avowedly 
antidistributional, antihistorical, and treats each culture 
as a closed system except insofar as its elements cor- 
respond to vital biological urges. Unhesitatingly reject- 
ing the intransigence of the creed, we accept its main 
positive postulate, but with important reservations al- 
ready indicated in our discussion of Boas (page 142). 

First and foremost, a science of Culture is not 
limited to the study of so many integrated wholes, the 
single cultures. This is doubtless important, but it con- 
stitutes neither the whole nor even the preponderant 
part of the ethnologist's task. A science of culture must, 
in principle, register every item of social tradition, 
correlating it significantly with any other aspect of real- 
ity, whether that lies within the same culture or outside. 
In defiance of the dogma that any one culture forms a 
closed system, we must insist that such a culture is in- 
variably an artificial unit segregated for purposes of 
expediency. Social tradition varies demonstrably from 
village to village, even from family to family. Are we 
to treat as the bearers of such a closed system the chief's 
family in Omarakana, his village, the district of Kiri- 
wina, the Island of Boyowa, the Trobriand archipelago, 
the North Massim province. New Guinea, or perchance 
Melanesia? The attempt to adhere rigorously to any one 

'E.g., Argonauts, 509, 517. Introduction to Hogbin, op. cit., vii. 
Myth in Primitive Psychology, 34 f.. New York, 1926. Coral Gardens, 457. 


of these demarkations precipitates absurdities. There is 
only oue natural unit for the ethnologist — the culture of 
all humanity at all periods and in all places ; only when 
the functionalist has, at least implicitly, defined his par- 
ticular culture within that frame of reference, does he 
know what he is talking about. Why are anthropologists 
forever harping on the Maya system of notation? The 
most significant thing is not that we find it embedded in 
mystical associations. But that these Americans achieved 
the abstract notion of the zero, thus intellectually tri- 
umphing over Greece and Rome — that is indeed a matter 
of moment. Yet in order to assess the fact we must know 
its distribution: imagine the same notation among Ituri 
Pygmies, Central Australians and Andamanese, and its 
significance radically changes. 

But the very spatial arrangement of our data leads 
to factors of another order. Without the physical en- 
vironment in which they are rooted, many of them are 
unintelligible. Here lies the extraordinary significance 
of Nordenskiold's investigations. In short, spurning iso- 
lated facts as superciliously as does Malinowski, we 
shall seek meaningful relations in all directions, not 
within the supposedly watertight compartment of a 
single body of social tradition. 

As already noted, Malinowski does emphasize one 
extracultural set of determinants : each phase of culture 
corresponds to "some fundamental tendency of the hu- 
man organism." Marriage satisfies the sexual need; legal 
parenthood stems from parental emotion; prescribed 
emergency behavior is derived from the urge of self- 
preservation.^" Whether these are truisms or statements 
of deeper significance, the propositions are too vague to 
interpret what we wish to have explained. We take it for 
granted that all forms of marriage have some connection 
with sex. What we want to know is why the Toda practice 

1* Introduction to Hogbin, op. di., xxxvii. 


polyandry, the Bantu polygyny, the Hopi monogamy ; and 
that cannot be explained in terms of generic human tend- 
encies, as Hocart well recognized. 

Concentrating, then, henceforth on the aspect of 
functionalism that has unquestioned explanatory worth, 
we must sharply define what has been proved from what 
has been alleged. It has been demonstrated that magic 
interlocks with economic attitudes, these again with 
social structure, and so on. But neither Malinowski nor 
any other "totalitarian" has ever shown that all the 
hundreds of descriptively separable traits play a definite 
role in tribal life, are all manifestations of one grand 
mystic unity. This, we saw, is now freely conceded by 
Radcliffe-Brown. As soon as functionalism is reduced to 
what it is — a worthy programme for ascertaining what 
intracultural bonds may exist — the neglect of other meth- 
ods appears as solely a matter of personal preference. 
That is to say, Malinowski may legitimately ignore for 
his purposes what he cannot interpret in functional 
terms, but he cannot deny the existence of "accidental 
or fortuitous" complexes. It is nothing but a clever law- 
yer's dodge when he turns the tables on Graebner, con- 
tending that chance connections can be inferred only 
after we have exhausted "all the possibilities of ex- 
plaining form by function and of establishing rela- 
tionships between the various elements of culture." 
Similarly, it is sheer dogmatism to declare : ' ' The better 
a certain type of culture is known, the fewer survivals 
there appear to be in it. ' ' " How many cultures has 
Malinowski ever examined from this point of view? 

Finally, Malinowski 's recurrent emphasis on the in- 
effable aspects of our field of inquiry seems exaggerated. 
Culture certainly includes untranslatable features, but 
so does all human experience. Science cannot deal with 
the incommunicable as such, however explicitly it takes 

""Culture," in Encyclopedia of the Social Sciences, 4:620-645, 1931. 


cognizance of its existence. What if a myth is *'not merely 
a story but a reality lived?" ^' We cannot relive the 
reality, but we can study that textual rendering which 
Malinowski disdains as merely the 'intellectual" aspect 
of the tales divorced from their mystic aura. 

Actually Malinowski 's practice soars above the limi- 
tations of his doctrinaire philosophy, an intuitive sense 
of fitness preserving him from its extreme implications. 
He will not "move one inch from my intransigeant posi- 
tion that the study of technology alone and the f etishistic 
reverence for an object of material culture is scientifi- 
cally sterile." But in the same breath he admits that "a 
knowledge of technology is indispensable as a means of 
approach to economic and sociological activities and to 
what might be adequately called native science. ' ' What is 
more, he gives the customary details about Trobriand 
yam houses; and though much is made of the emotions 
of native mariners towards their boats as "the deepest 
ethnographic reality," empathetic sentiment presently 
yields to a thorough consideration of the craft as an 
adaptation to obvious needs.^^ 

Very far from treating, say, Omarakana village as 
an impermeable entity, Malinowski defines Trobriand 
distributions in the spirit of any competent up-to-date 
ethnographer. He marks off the several economic dis- 
tricts of the archipelago, correlates local emphasis on 
fishing and stone polishing with environmental peculiari- 
ties, points out the regional differences in regard to 
shipbuilding. He even steps outside the magic circle of 
the Trobriands to include Kitava because all Boyowan 
canoe mythology is associated with that island. Indeed, 
the ceremonial trading to which he devotes a major 
treatise could not be described without discussing other 
northern and even some southern Massim tribes. Where, 

^ Myth in Primitive Psychology, 18, 24, New York, 1926. 
13 Coral Gardens, 460. Argonauts, 105-145. 


however, shall the line be drawn? Boyowa is intelligible 
only with reference to Massim, Massim cannot be under- 
stood apart from Melanesia, from Oceania, from all 
culture history as the one and only true whole. 

Better still, this scorner of history himself recon- 
structs the past. He does so self-consciously, mumbling 
the purifying spell that he is ''discounting any undue 
antiquarian or historical bias." Actually, he infers — 
convincingly — that taro preceded several kinds of yams 
in native cultivation because of its preponderance in 
horticultural magic. It thus appears that "the ethnogra- 
pher ought to keep his eyes open for any relevant indica- 
tions of evolutionary lag or historical stratification. ' ' " 
But, if so, there is functional disharmony: the small 
yam is economically more important than taro, hence it 
and not the taro ought to be the primary center of magi- 
cal usages. The "evolutionary lag" of taro magic is 
evidently an instance of what the wicked evolutionists 
called "survival." And if historical stratification is not 
only permitted but prescribed when relevant, the taboo 
against a chronology of the past is mitigated in the 
obvious sense: sound reconstruction is admitted, and 
only fanciful reconstruction remains outside the pale. 

To sum up, Malinowski's practice fortunately does 
not bear out the negative excrescences of his principles. 
He is certainly within his rights in studying most what 
interests him most; he becomes a dogmatist only when 
laying down the same tastes for others; and since he 
absolves himself from orthodoxy when common sense so 
dictates, others will do well to follow what he does rather 
than what he prescribes. 

As for the valid core of his doctrine, we reiterate 
our faith in its importance. In fairness, however, it can- 
not be considered wholly new. Bachofen, Fustel de Cou- 
langes, and Boas, among others, presented "several 

" Coral Gardens, 459. 


aspects (of culture) closely intertwined and influencing 
one another"; probably everywhere scholars have fol- 
lowed the practice intuitively. In his ostensibly pure de- 
scription of Californian Indians, for example, Kroeber 
expounds the integration of Mohave life by dreams, 
songs, and myth. Among the Northwest Californian 
tribes, he finds the caste spirit reflected on etiquette, 
money-madness on class distinctions, caste sentiment on 
marriage forms. In the latter area Du Bois, too, has 
shown how the prestige of wealth lies at the core of tribal 
ideology. Ajnong the Yokuts of central California, A. H. 
Gayton discovers a social utility of sorcery that strictly 
parallels Malinowski's findings in the Trobriands.^^ Rep- 
resenting another school, Kirchhoff has unearthed signifi- 
cant South American relations : matrilocal residence, 
bride-service, and the menial status of the son-in-law 
form a connected whole. Correlated with them is a special 
marriage form, for a husband's lot is mitigated if his 
daughter marries his brother-in-law, who is thus freed 
from the necessity of seeking a mate away from his natal 

In short, Malinowski is hardly the father or the sole 
exponent of functionalism. We gladly hail him as its most 
articulate, its most persuasive herald. Others have either 
preached or practiced the faith; he has done both. 

There is one grave error of omission in Malinowski's 
achievement, the narrowness of his ethnographic ap- 
proach. His rejoinder to critics who reproach him with 
viewing all savages in the light of Trobriand is fair 
enough : His generalizations purport merely to provoke 

1^ A. L. Kroeber, HandhooTc of the Indians of California, 1-99, 754- 
795, Washington, 1925. Cora Du Bois, "The Wealth Concept as an In- 
tegrative Factor in Tolowa-Tututni Culture," ALK, 49-65, 1936. A. H. 
Gayton, "Yokuts-Mono Chiefs and Shamans," UC-PAAE 24:361-420, 

18 Paul Kirchhoff, ' ' Die Verwandtschaf tsorganisation der Urwald- 
stamme Siidamerikas, " ZE, 63:85-193, 1931. 


parallel inquiries in other regions; he has, indeed, ex- 
pressly demanded a * ' fuller testing in the various anthro- 
pological provinces. ' ' " Nevertheless, he remains an 
ethnographic provincial, unable to wean himself from 
the Trobriand, or at most the Australo-Oceanian, scene 
for that massive comparative attack on civilization that 
characterizes a Tylor or a Boas. It is not, of course, that 
he is actually ignorant of other areas, but that he cannot 
bring himself to institute intensive comparisons, such as 
Eadcliffe-Brown, for instance, broaches when discussing 
the South African avunculate. Yet some of Malinowski's 
most important results fairly clamor for precisely this 
sort of checking. Why are not the Trobriands compared 
with other matrilineal regions, with Northwest America, 
Angola, Arizona"? Is it of no interest to ethnology that 
nepotic succession, avuncular authority and paternal de- 
votion in northern British Columbia closely parallel the 
Melanesian picture ? ^* Is it not worth investigating 
whether that soul-stirring conflict of duty and love so 
graphically portrayed by Malinowski would recur in like 
conditions elsewhere? 

The quality of Malinowski's contribution is pre- 
sumably clear from the foregoing remarks. By precept 
and — better — by the example of his superb field work he 
has thrown into relief the importance of uncrystallized 
aspects of native life, of correlating so far as possible its 
several phases instead of separating them in distinct 
rubrics. His intolerance of other approaches, his adoles- 
cent eagerness to shock the ethnological bourgeois — that 
figment of his fancy, the mere technologist and oddity- 
monger — must not blind us to his soundness on problems 
of social organization, his vital ideas on primitive law 
and economics. On the other hand, the brilliance of his 
results must not dazzle us into brushing aside as inferior 

1'^ Introduction to Hogbin, Iviii. Sex and Repression, 139. 

18 E.g. Franz Boas, "Tsimshian Mythology," BAE-E 31:425f., 1916. 


the methods and topics he rarely touches and, in the 
abstract, arbitrarily taboos. 

T H U R N W A L D 

Richard Thurnwald (1869- ), like Malinowski, is 
distinguished both as an observer and as a theorist. His 
first expedition (1906-1909) led him to the Bismarck 
x\rchipelag-o and the Solomon Islands, a second (1913- 
1915) — interrupted by the Great War — to the interior of 
German New Guinea, and in 1933 he resumed work in 
:\Ielanesia. In 1930-1931, Thurnwald, assisted by Mrs. 
Thurnwald, investigated several East African tribes 
under the auspices of the Africa Institute. As a teacher 
he has been active at Halle, at Y'ale, and at his present 
headquarters, Berlin. Internationally known by his ex- 
plorations, his repeated sojourns in the United States, 
and his innumerable publications, Thurnwald became 
one of the foremost liaison ofl&cers of the social sciences 
when he founded the Zeitschrift fur Volkerpsycliologie 
und Soziologie, a journal that gave full scope to his in- 
terests, at once wide and deep, in economics, sociology, 
jurisprudence, and psychology. 

Since Thurnwald and Malinowski worked in the 
same general region, it is worth while comparing their 
results, which actually coincide to a gratifying extent. 
Thurnwald 's earliest reports" anticipate Malinowski 's 
emphasis on the network of mutual services and counter- 
services {Leistungen und Gegenleistungen) as a char- 
acteristic of Melanesian communities. They also set forth 
the barterers' tendency to reckon as equivalent only 
certain definite types of objects, a pig, e.g., being ex- 
changed for a knife, not for a spear or an armlet. Thus 

18 R. Thurnwald, ' ' Im Bismarckarchipel und auf den Salomoinseln 
1906-1909," ZE 98-147, 1910. Idem, " Ermittlungen iiber Eingeborenen- 
rechte der Siidsee," Z vgl R 23:309-364, 1910. Idem, Das Bechtsleben der 
Eingeborenen der deutschen Siidseeinseln, seine geistigen u. wirtsclmftlichen 
Grundlagen, Berlin, 1910. 


they foresliadow Malinowski 's critique of the concepts 
of ''value" or ''money" as applied to aboriginal con- 
ditions. Theoretically significant are the recurring func- 
tionalist sentiments in these early papers. Thurnwald 
connects real-estate law with food-getting activities ; ex- 
plains inheritance rules from the social structure, includ- 
ing the sexual division of labor ; shows how these usages 
are interwoven with mortuary customs and how the penal 
code reflects religion as well as the political organization. 

But the differences are no less striking than the 
resemblances. Less intent on mirroring the native's in- 
wardness, Thurnwald presents a comparatively sober 
picture. He insists on the importance of personal dif- 
ferences, but does not demonstrate them in the flesh. 
Less subject to the hypnotic charm of his material, he 
arranges it under convenient captions derived from ex- 
trinsic knowledge and views it in the light of current 
theory. In Argonauts Malinowski merely refers to the 
Kulturkreis school; Thurnwald proceeds to test its spe- 
cific conclusions. Are bows really associated with pile- 
dwellings? Thurnwald scrutinizes the several types of 
dwellings seen in New Guinea and finds but a partial 
confirmation of Graebner's assumption.^" 

Thurnwald 's mind, moreover, has an encyclopedic 
range that carries it far beyond the ethnographer's 
purlieus. He not only classifies his natives linguistically, 
but studies their somatic traits, noting albinism and 
pygmy statures; and defines their habitat in geographi- 
cal terms. Nor is this trait confined to the sphere of 
observation. In a treatise on Papuan society" a pains- 
taking exposition of marriage and kinship practices 
merges into a theoretical discussion from linguistic, 
psychological, biological, and even ethical angles. And 

20"Vorlaufiger Bericht liber Forschungen im Innern von Deutsch-Neu- 
Guinea in den Jahren 1913-1915," ZE 147-179, 1917. 

^^ Die Gemeinde der Bdnaro; ein Beitrag sur Entstehungsgeschiehte 
von Familie und Staat, Stuttgart, 1921. 


more amazing than the scope of the author's interests is 
the specitic way in which he considers publications in 
these several branches of knowledge — disquisitions on 
the nature of the state, reports of the Eugenics Labora- 
tory, or monographs on American Indian relationship 

Polymathy, infused with an urge to systematize, has 
prompted both topical " and regional ^ surveys such as 
are obviously uncongenial to Boas, Malinowski, and 
Radclilfe-Brown. Schmidt presents possibly the closest 
recent parallel, but with a characteristic difference of 
another order. Schmidt, we found, though by no means 
opposed to a psychological approach, directs his main 
constructive efforts toward historical ends. Thurnwald, 
on the other hand, while professing no hostility to his- 
tory, is primarily attracted by problems of another cate- 
gory. To be sure, he demands historical perspective lest 
the functional picture be distorted; amply cites docu- 
mentary records; accepts and applies the contact of 
peoples as an explanatory factor; and even chides the 
Graebnerians for paying too little attention to specific 
events. So, when he sketches paleolithic society, neither 
his principle of reconstruction nor his essential result 
ostensibly deviates from Schmidt's. Nevertheless, even 
here there is a noteworthy difference. Schmidt strives 
for as particularized a picture as possible, Thurnwald 
is concerned with what is typical either as process or 
sequence. In other words, his orientation is primarily 

This preference looms large whenever Thurnwald 
discusses diffusion. Borrowing, he is never tired of 

^Psychnloriie fte.f primitiven Menschen, in Gustav Kafka, Handbuch 
der vergleichenden Psychologic, 1:147-320, Miinchen, 1922. 

23 Die menschliche Gesellschaft in ihren ethno-soziologischen Grund- 
lagen, 5 vols., Berlin u. Leipzig, 1931-1935. Esp. vol. I, Eeprdsentative 
Lehensbilder von NaturvdlTcern. 

^Ibid., 1:15 f., 91-93, 1931; 2:280, 308, 1932; 4:297, 315, 1935. 


teaching, is not a mechanical phenomenon, but depends 
on the recipients' as well as on the donors' culture. It 
has psychological correlates : not even a paddle or an 
arrow can be taken over without affecting the borrowers ' 
mentality; further, there may be vital alterations in the 
form and meaning of what is received. Again, so far from 
there being an automatic transfer of traits when peoples 
come in contact, different loan-elements are assimilated 
with varying rates of velocity. All this is unquestionably 
sound, though these propositions are no longer peculiar 
to the functionalists. Since Thurnwald, on the other 
hand, is not on principle so intolerant of chronological 
reconstructions, the antithesis between him and the his- 
torical ethnologists reduces largely to a difference in 
emphasis or personal interests. 

Such divergence, however, remains vital, for it af- 
fects what a scholar does and how he responds to a given 
problem. Thurnwald, while not intransigent, is at times 
quite orthodox in his functionalist convictions. Thus, he 
confronts with skepticism evidence that kinship nomen- 
clatures have been disseminated: such diffusion, he 
argues, can take place only under highly favorable cir- 
cumstances and implies modification, nay, a revolution 
in the whole social structure; never could it rest on 
"mere sport or idle mimicry" {blosser Spieler ei oder 
sinnloser Nachdjfung).^^ Yet Tarde has pointed out the 
strength of prestige suggestion; and, apart from that, 
as already emphasized, the doctrine that the elements 
of a particular culture must be organically related is a 
useful heuristic hypothesis but most emphatically not a 
demonstrated proposition. However, the purpose of our 
example is not to criticize but to define: what concerns 
us is that where others joyously explain a given dis- 
tribution as intelligible only by dissemination, Thurn- 
wald remains dissatisfied unless the fact of borrowing 

28 Die Gemeinde der Bdnaro, 176 sq. 


can be harmonized with his general notion of cultural 

Not concerned with the spread of elements as such, 
Thurnwald is willing to recognize independent develop- 
ment of similarities found in widely separated regions. 
He spurns the old unilinear parallelism; sensible of the 
intricacies of cultural growth, he concedes the distorting 
effect various factors might exercise on a '^ normal" 
trend of events. Notwithstanding this, however, there 
are residual regularities, which science must trace to 
their adequate antecedents. This purged evolutionism 
approaches Radcliffe-Brown's and — intermittently and 
malgre lui — even Schmidt's views, but Thurnwald 's is 
the amplest elaboration, perhaps best exemplified in his 
discussion of the state. With due caution as to variants, 
he presents a series of stages, each of which purports 
to be the logical and psychological successor of its im- 
mediate antecedent. They are : 

(1) The origin of fixed groups of famihes (bands, classes, 

(2) Their crystallization around permanent families of 

(3) Differential estimation of families and family groups 
according to their descent and culture (ethnic stratification). 

(4) Mixture and assimilation among the ethnically distinct 
and diversely estimated groups. 

(5) Exceptional status of mixed-breeds and assimilated 
persons coupled with general resentment because of segregation 
and difference in ranking. 

(6) Aristocracy is superseded by dynastic despotism, with 
transvaluation of values according to dynastic relationships 
(castes and guilds, bureaucracy). 

(7) Individualization of society, democracy and plutoc- 

(8) Upstart rulers, whose powers rest on the personal de- 


votion of their military retinue. At this stage there sets in a 
trend toward creating a secondary homogeneity."® 

To this aspect of the doctrine we shall return pres- 

Thurnwald presents a rare blend of field experience, 
ethnographic erudition, theoretical interests, and syste- 
matic thinking. The last-mentioned trait appears to ad- 
vantage in his discussion of primitive economics in 
Volume III of his great work. Malinowski had, indeed, 
advanced our insight by his investigation of selected 
topics, and Mauss had intensively treated the exchange 
of gifts. Thurnwald, however, offers the first competent 
ethnographic survey of the entire field from the socio- 
political point of view, thus supplementing Hahn's dis- 
cussion of farming and animal husbandry. Apart from 
once more throwing into relief the nonrational motives 
in simpler societies, Thurnwald suggestively defines all 
pertinent concepts, such as ''trade," ''money," "de- 
mand," and "capital," in relation to primitive condi- 

An outstanding trait of Thurnwald 's mentality is 
his poise. Except for occasional anti-Graebnerian flings, 
he avoids partisanship, judicially weighing pros and 
cons. We may cite his penetrating comments on Levy- 
Bruhl (page 220), his utterances on the value and the 
limitations of direct testing techniques in estimating 
primitive capacity." He fights for principles, not for 
shibboleths: even "functionalism" fails to hypnotize 
him into unqualified enthusiasm. Like Boas, he warns us 
against catchwords that purport to exhaust reality but 
merely distort it. The couvade, he shows, is only an 
extreme type of natal taboo systems and must not be 
treated by itself. Similarly, "marriage by capture" 

2«Dfe menschliche GesellscJiaft, 1:9 f., 16, 24 f.; 4:24 sq., 236, 290 sq., 
302 ff. 

^"^ Fsychologie des primitiven Mensohen, 174. 


should uot be divorced from groom-abduction, as E. C. 
Parsons had already indicated.^* 

Thurnwald is most felicitous in handling broad gen- 
eral principles. He sets forth admirably the role of 
leadership, the straining and preferential retention of 
definite personality types by definite forms of society. 
With equal clearness he contrasts the irreversibility of 
technological progress — where each step presupposes 
its predecessors — with the "cyclic" sequences not sub- 
ject to hierarchical grading. While the plough, for ex- 
ample, must be preceded by a hoe or a dibble, matrilineal 
descent is merely one of a limited number of possibilities 
and by any objective criteria neither higher nor lower 
than its alternatives.^^ 

But it is a commonplace that men have the defects of 
their virtues. No one could cover the range of Thurn- 
wald 's material without factual lapses and inconsisten- 
cies, which in some volumes of his major treatise were 
unduly increased by his remoteness from libraries at the 
time of printing. On the other hand, the concern with 
empirical data sometimes betrays Thurnwald into over- 
ample description that no longer strengthens any general 
point, being apparently jotted down solely for the 
author's future convenience. 

More keenly than such peccadilloes we feel the fre- 
quency with which Thurnwald propounds correlations 
of the utmost significance without attempt at proof. 
Probably the second volume of Die menschliche Gesell- 
schaft alone would furnish subject matter for two dozen 
doctoral dissertations. Woman's higher status is con- 
nected with her economic independence, negatively cor- 
related with masculine preponderance in food-getting, 
positively with matrilineal descent; the levirate is at- 

28 Die menschliche Gesellschaft, 2:105 sq., 4:248. 
^liid., 4:266 sq., 288 ff. " Sozialpsychische Abliiufe im Volkerleben, " 
ALK 383 sq., 1936. 


tached predominantly to clanless tribes, and so on.^ 
Irrespective of inherent probability, such propositions 
cannot be accepted without the fullest inductive demon- 

Thurnwald's weakness is thus quite different from 
that of many other ethnologists. He is wise and learned, 
he has abundantly proved his skill as an observer, he is 
empirically and yet also theoretically minded. The 
trouble is that he often fails to concentrate his facts on 
the crucial point of the argument. To revert to his out- 
line of political evolution,^^ there are facts galore and 
there is a plausible scheme, but the two are very loosely 
connected. Thurnwald does not use his historical data to 
demonstrate that there actually has been a repeated suc- 
cession of events in independent areas. Here, and in com- 
bining with such evidence that for the correlations 
alleged, but not proved, lies the greatest gap to be filled 
in Thurnwald's thinking. 

30 Die menschliche Gesellschaft, 2:34, 192, 246. 
3i7bM?., 4:23 sq., 251 sq., 302 ff. 



Nothing would be more remote from the truth than 
to conceive ethnologists as ranged in hostile camps. There 
are noteworthy differences of opinion, yet they often re- 
solve themselves into a difference of emphasis or obvious 
misunderstandings. Some insist on disagreeing where 
there is perfect harmony lest they forego the dramatic 
role of the prophet preaching in the wilderness; others 
criticize their fellows not for what they believe but what 
in the opinion of the writer should logically be their be- 

But turning back a generation or two is to become 
aware of general advancement in which virtually all eth- 
nologists share. No one now relapses into the environ- 
mentalism of Klemm; the danger lies in the opposite di- 
rection. To quote one who started as a geographer: 
"Neither the world distributions of the various economies, 
nor their development and relative importance among 
particular peoples, can be regarded as simple functions of 
physical conditions and natural resources. Between the 
physical environment and human activity there is always 
a middle term, a collection of specific objectives and 



values, a body of knowledge and belief: in other words, 
a cultural pattern. ' ' ^ 

Similarly, whatever divergences exist as to the innate 
endowment of races, responsible writers unite in regard- 
ing these native differences as comparatively small; and 
certainly no one nowadays would derive, say, the profu- 
sion of secret societies in Africa and their lack in Siberia 
from simple hereditary mental differences between 
Negroes and Siberians. Again, Levy-Bruhl's provocative 
thesis has been examined and rejected by workers of di- 
verse schools and most varied opportunities for observa- 
tion — by Thurnwald and Boas, Driberg and Spier. As 
Seligman puts it, it is ' ' in contradiction to the experience 
of field workers, who are after all the best qualified to 
judge. " ^ At the same time the potency of irrational deter- 
minants championed by Tarde, Boas, Levy-Bruhl, and 
others is an integral part of modern teaching. 

In short, four simpliste errors — enviro nmental ism, 
racialism, the notion of a prelogical primitiyeness^ and 
that o f primitive intellectualism — are definitely disc arded . 

Turning to the much mooted question of historical 
connection versus independent development, we find 
again substantial progress since Bastian's day. There is 
general agreement that Bastian's *' genetic law," unde- 
fined as it remained, cannot explain specific coincidences. 
While no one ever denied diffusion in toto, its importance 
has been established beyond cavil and what nowadays 
divides scholars is merely the intensity of their concern 
with this principle, the nature of their methodological 
safeguards against error. We have seen that an avowedly 
antihistorical functionalist like Malinowski may sporadi- 
cally turn historian malgre lui. On the other hand, there 
is no unescapable conflict between functionalism and 

1 0. Daryll Forde, Habitat, Economy and Society, 463, London, 1934, 

2 C. G. Seligman, ' ' The Unconscious in Eolation to Anthropology, ' ' 
British Journal of Psychology, 18:373, 1928. Leslie Spier, "Havasupai 
Ethnography," AMNH-P 29:331, 1928. 


atomism. The same student who is not a doctrinaire shifts 
his ground with the nature of his problem. Spier, at one 
time intent on plotting the distribution of isolated ele- 
ments, insists most vehemently that comparisons based on 
the resulting tables remain inadequate. So Nordenskiold, 
whose charts are a thorn in functionalist eyes, gives the 
most intimate picture of native drinking bouts and juve- 
nile games. This is as it should be. As Parsons remarks, 
"Wisdom in ethnology, as in life, lies in having more 
than one method of approach. " ^ 

But harmony extends beyond matters of general ap- 
proach into the field of special interpretations. Within 
the space of half a century a number of questions have 
been settled — so far as we can judge, definitively. No one 
now defends the three-stage theory of economic progress ; 
and Hahn's distinction of plough-farming from hoe- and 
dibble-farming stands unchallenged. That promiscuity 
now exists nowhere and is an unproved hypothesis for 
the past is the view of Radcliffe-Brown, Malinowski, 
Schmidt, Thurnwald, Brenda Z. Seligman, and all Ameri- 
canists. Swanton, Schmidt, Radcliffe-Brown, and Mali- 
nowski — to mention only a few — ^have definitely sup- 
planted the idea of clan priority still held by Durkheim 
with the recognition of the family as a basic social unit. 
Even in comparative religion there is at least far-spread 
convergence toward acceptance of a high-god concept on 
rude levels: what was still unthinkable to Tylor is now 
cheerfully accepted by Americanists like Radin and 
Cooper, Africanists like Baumann.^ 

3E. C. Parsons, Mitla; Town of Souls, 479, Chicago, 1936. Leslie 
Spier, "Havasupai Ethnography," AMNH-P 29:83-392, 1928; idem, 
' ' Cultural Eelations of the Gila River and Lower Colorado Tribes, ' ' Yale 
University Publications in Anthropology, No. 3, New Haven, 1936. Erland 
Nordenskiold, Comparative Ethnographical Studies, 9 vols., Goteborg, 1919- 
1930; idem, Indianerleben, Leipzig, 1912. 

* Paul Eadin, Primitive Man as Philosopher, 342-374, New York, 1927. 
Hermann Baumann, Schopfung und TJrzeit des Menschen im Mythus der 
afrikanischen VolJcer, 5, 164, Berlin, 1936. 


Of course it would be foolish to deny sharp differ- 
ences on any number of special problems. But such diver- 
sity of opinion should be seen in proper perspective. Some 
of it is inevitable and desirable in a live and growing 
branch of knowledge ; some of it is illusory — the figment 
of controversial would-be Messiahs who obscure issues 
by a melodramatic contrast between the elect and the 

In the introductory chapter we saw that sound eth- 
nology was impossible until geographical discovery had 
paved the way for at least a rough cha rting of the gamut 
of social variability. For some time past exploration has 
taken a turn towards intensiveness of survey. Whatever 
may be feasible in practice, our theoretical aim must be 
to know all cultures wit h equal thoroughness. How inten- 
sively a particular problem shall be studied at a definite 
stage varies with the circumstances, precisely as does the 
decimal place to which a physicist shall carry his calcula- 
tions. For certain purposes it sufiices to characterize a 
dwelling as round ; in Samoa a close analysis reveals that 
what is so described is not a true round structure at all, 
but a rectangular house with shortened middle section 
and terminal apses. Any inferences from the occurrence 
of ''round" dwellings in Samoa are therefore fallacious.^ 
If technological and genealogical particulars of forbid- 
ding aspect loom large in modern monographs, they 
should not be interpreted as meaningless trivialities. They 
may be essential for the broaching of new problems, for 
the definition of the observed phenomenon itself. It may 
be a boresome detail whether the fragments of a Peruvian 
fabric were originally of one web, but it is of the utmost 
importance to ascertain how these textiles rank among 

^ Te Rangi Eiroa, Samoan Material Culture, 16 sq., 665 f., Honolulu, 


those of the world; and only technical considerations of 
the textile expert can tell us/ 

Ethnologists are not always sufficiently conscious of 
the assistance rendered to them by techniques and con- 
A cepts extraneous to their own discipline. Yet such de- 
\^ pendence is no cause for abasement. There are no hard 
C^ and fast lines between culture and the rest of reality. For 

^ ^ ^ specific tasks, zoological, botanical, psychological, histori- 

N^^ -Q.'- ^- cal, metallurgical facts may prove more important than 
> (^0- ^«^ other phases of culture. How can we know that the aver- 
'^ sion to incest is not innate? Only if the psychologist as- 

- sures us that no such instinct exists. How could Rivet 

demonstrate the existence of Mexican bronze and its 
affinity with Peruvian metalwork? How could Norden- 
skiold ascertain the efficacy of the Colombian tools made 
from an alloy of gold, silver, and copper ? Only by requisi- 
tioning the services of a metallurgist. We cannot gauge a 
people's utilization of their natural resources without 
knowing the character of the fauna, flora, and topogra- 
phy, i.e., without the help of natural history and geogra- 
phy; and so theoretical a matter as Levy-Bruhl's thesis 
can be settled only in the light of such ecological insight. 
This is the justification for the development of ethno- 
zoology and ethno-botany.'^ 

Again, cultural phenomena vary in time ; and as Rad- 
cliffe-Bro^vn's recent statements show, it is now generally 
recognized that we cannot understand process without a 
grasp of chronolo gical relations. W e have noted the revo- 
lution in Thought, the sudden flood of light that emanated 
from Boucher de Perthes' discoveries. But whence did 

6 LUa M. O 'Neale, ' * Wide-Loom Fabrics of the Early Nazca Period, ' ' 
ALK 215 sq., 1936. 

■^ Paul Kivet et H. Arsandeaux, ' ' Contribution a 1 'etude de la metal- 
lurgie mexicaine, " SAP-J XIII, 1921. E. Nordenskiold, Comparative 
Ethnographical Studies, 9:101-112, 1931. Erna Gunther, "A Preliminary 
Eeport on the Zoological Knowledge of the Makah," ALK 105-117, 1936. 
Walter E. Eoth, North Queensland Ethnography Bulletins, Brisbane, 1901- 


this new perspective come'? Evidently from the geologist's 
technique of stratigraphy. Repeatedly this intrinsically 
extraneous method of procedure has been applied with 
almost equally spectacular success. In the hands of Kid- 
der and Nelson it transformed a jumble of unintelligible 
facts about the Southwestern United States into an or- 
derly system. Nor should we forget that the greatest sub- 
sequent advance in this region is due to an astronomer's 
concern with climatology. By studying the growth of tree 
rings in drought and in moisture, A. E. Douglass found 
that ' ' definite ring patterns recorded specific year groups 
and as a consequence developed a system whereby he can 
tell the year when a log was cut from a living tree. Begin- 
ning with trees whose actual cutting date was known he 
has been able to devise a type ring chart going back to 
about 700 A.D. ' ' Material antedating living trees thus fur- 
nished an archeological time scale: against Douglass' 
charts the investigator checks the rings on beams in his 
ruins and determines the year of the cutting. Though 
timber need not have been used immediately after felling 
of the tree or may have been used more than once, these 
minor deviations can be corrected and hardly detract 
from the accuracy of the dates computed for the erection 
of the buildings.* 

Primitive literature (page 95) could be studied com- 
prehensively only when ethnographers had borrowed the 
phonetician's technique for writing down the hitherto 
orally transmitted tales and poems of illiterate peoples. 
This loan made possible such superb works as Thalbitzer 's 
studies of Eskimo lore and Bunzel's collection of Pueblo 
chants. Similarly, comparative musicology, hampered by 
inferior techniques until quite recent times, rose to a new 

8N. C. Nelson, "Pueblo Euins of the Galisteo Basin," AMNH-P 15, 
1914. Idem, "Chronology of the Tano Ruins," AA 18:159-180, 1916. 
A. V. Kidder, An Introduction to the Study of Southwestern Archaeology, 
New Haven, 1924. F. H. H. Roberts, Jr., "A Survey of Southwestern 
Archaeology," AA 37:1-35, 1935. 



plane with phonograpliic recording, and is promising to 
bear effectively on some of the central problems of eth- 
nology (see page 195).® 

Ethnology thus leans on her sister sciences and ad- 
vances with their progress. Its autonomy no longer re- 
quires the bumptious assertiveness that marks a sense of 
inferiority. There is no danger of absorption in a ''more 
fundamental" discipline, for the reality of social tradi- 
tion as a separate aspect of the universe is no longer 
doubted. What we need is a clearer recognition of how 
cultural phenomena interlock with others; let us then 
examine the relations of ethnology with geography and 
psychology, since these sciences bear upon culture not 
intermittently but inevitably and constantly. For, how- 
ever inelegant may be a people's adaptation to their sur- 
roundings, so7ne solution of the environmental problem 
is a prerequisite for survival, hence a co-determinant of 
every culture studied. And still more pervasive is the 
psychological factor which enters not only every culture 
but every item of culture. For, as Boas long ago pointed 
out, the artifacts of a museum collection differ from the 
inert objects of a mineralogical cabinet because they in- 
variably symbolize a social tradition — the interplay of 


«sf^ Geography supplies the student of culture with a 

>^ technique as well as with certain definite results. The facts 

^ he studies vary in space, and to determine their spatial 

^o relations is the first and most obvious of his tasks. A 

^ » Wm. Thalbitzer, The Ammassalik Eslcimo, 115-559, Copenhagen, 1923. 

Euth Bunzel, "Zuni Ritual Poetry," BAE-R 47:611-835, 1932. Helen H. 
Roberts, "Melodic Composition and Scale Foundations in Primitive 
Music," AA 38:79 sq., 1936; eadem, "Musical Areas in Aboriginal North 
America," Yale University Publications in Anthropology, No. 12, New 
Haven, 1936. George Herzog, "Plains Ghost Dance and Great Basin 
Music," AA 38:79 sq., 1936; eadem, "Musical Styles in North America," 
ICA 23:455-458, 1928. 


distributional approach does not solve all problems but 
it is a first step towards understanding. It is fashionable 
to deride this pedestrian procedure, but its prophylactic 
efficacy remains unchallengeable. Without accurate in- 
formation about distributions we are unable to appraise 
any given achievement. If the zero concept were the com- 
mon property of the Maya and the Australians, its pres- 
ence would bear a wholly different significance. So we 
cannot regard moieties as a general phase of social evolu- ^S- 

tion when they are so conspicuously lacking in Africa. ^^v*^ y 

Because the range of one trait regularly coincides with ^^^ "v^. 
another we suspect a significant bond between them. On ^ s^*^. V"^ 

the other hand, a given distribution may suggest all sorts \\^ 
of historical problems. It was the geographical tabulation 
of mythological episodes within a continuous area that 
enabled Boas to prove diffusion of tales beyond the 
shadow of a doubt. On the other hand, an intermittent 
occurrence raises questions of another type. The hockey 
game of Plains Indians turns up in the Gran Chaco, but 
it is lacking in the intervening territory. Nordenskiold 
plausibly suggests that the game was shared by the ances- 
tral groups which were once in contact but later drifted 
apart. The story of a benevolent and a marplot brother, 
the latter introducing death and labor into the world, oc- 
curs in California and in Tierra del Fuego, while the 
tales about twins in such regions as the Brazilian interior 
are basically different. Hence neither psychic unity nor 
pan-American unity accounts for the facts, which are 
again most easily explained — whether correctly or not — 
by early intercourse.^** 

Naturally cartography is a mechanical instrument, 
not a master. If we plot distributions in ignorance of vital 
facts, sham issues are inevitably raised. Thus, it is one 
thing to imagine that coiled baskets are lacking in South 

i°Erlancl Nordenskiold, Comparative Ethnographical Studicft, 9:90, 


America outside of Fuegia and quite another to discover 
them also in Panama, Colombia, Brazil, Peru, and on the 
Chilean coast. Similarly, the absence of hoes from all of 
South America would be of real interest were it not that 
their undoubted use among the Quichua, past and present, 
eliminates the problems otherwise raised. Naturally, the 
sane cartographer must pay attention to chronological 
differences emerging from his sources. Thus, Norden- 
skiold finds that in the Upper Amazonas the earlier spear 
throwers were supplemented by the blowgun. This, in 
turn, stimulates the question of why such a change took 
place ; and we learn that it hinged on new trade relations 
that brought in curare since without this poison the darts 
blown from a blowgun remain ineffectual missiles in war- 
fare. Naturally, it is important to define one's items so 
that only the same concept receives the same label. Here, 
Nordenskiold, whose procedure is generally most com- 
mendable for the exhaustiveness of his survey, both 
spatially and temporally, is not always convincing. While 
''hockey" is undoubtedly — irrespective of interpretation 
— a clearly defined feature, this no longer holds for the 
rubric "tents of animal skins"; since their shape is ad- 
mittedly quite different from that of the North American 
tipi, no good purpose is served by the common label.^^ 

Such sporadic lapses by students of distribution in no 
way warrant the slurs cast on their endeavors in recent 
times. "Atomistic" investigations, when based on a con- 
ceptually sound typology, on the complete spatial distri- 
bution of traits and on their ascertainable temporal 
modifications, are fully as justified as the aesthetically 
more attractive "totalitarian" approach. Ethnology, we 
repeat, is not merely the science of cultures but of culture 
— of every fragment of the universe pertaining to the 
social heritage of all human groups. 

If the cartographic approach imperceptibly leads to 

11 Nordenskiold, op. cit., 2:24, 168, 1920; 3:59-64, 1924; 9:77-94, 1931. 


questions of deeper import, the consideration of the physi- 
cal conditions confronting a society forms an integral part 
of any clear conception of their culture. It is not at all a 
matter of identifying culture with an automatic response 
to environment — an error long since exposed and buried. 
What is involved is partly again the elimination of sham 
problems. As Nordenskiold has so beautifully shown, cer- 
tain traits are simply barred by environment: south of 
Santa Cruz, Bolivia, bark cloth and fish-drugging dis- 
appear for want of suitable species of trees and narcotic 
plants, respectively. With navigable rivers are eliminated 
boats, with the absolute lack of the raw material, the whole 
art of stonework.^^ 

Discussion of many problems is bound to remain 
sterile without the application of geographical insight. 
Farming looms large in the history of mankind, but the 
problem of its origin and spread involves matters no eth- 
nologist can settle without recourse to principles not de- 
rived from the social heritage alone. Given the general 
equipment of incipient cultivators, there is the matter of 
possible soils. In Northern Mexico settlements were re- 
stricted mainly to flood plains and to rough mountain 
lands to the exclusion of rich clay and clay loam tracts, 
such as primitive dibbles or even bone and stone mattocks 
cannot cope with. The Calif ornian aborigines resisted the 
impulse to adopt farming after the fashion of the Colo- 
rado Eiver peoples partly because of their soils, partly 
because those crops which were available to them could 
not be profitably raised in regions with winter rain. ' ' The 
Pacific Coast of the United States, as a land of Mediter- 
ranean climate, had to wait on the introduction of crops 
from the European Mediterranean." For maize thrives in 
areas that are warm and humid during the initial months 
of growth and not, as some ethnographers have suggested, 
in arid surroundings. More generically, Sauer finds nas- 

12 E. Nordenskiold, Indianer und Weisse, 15-20, Stuttgart, 1922. 


cent farming must be sought in forest lands, not in brush 
or grass areas, which present insurmountable impedi- 
ments to any but advanced forms of tillage/^ 

Investigators of Arctic peoples have consistently- 
stressed environmental conditions not because they deter- 
mine these cultures any more exclusively than elsewhere, 
but because they are so obtrusive as co-determinants. 
Eskimo life is a constant interplay of geographical and 
cultural factors. The Southampton Islanders, lacking the 
soapstone other Eskimo use for their lamps, substitute 
vessels of limestone slabs painfully cemented together, 
yet tenaciously adhering to the traditional form. The im- 
mense length of sledges in certain Eskimo tribes, the sub- 
stitution of wood for bone, the importance of intertribal 
trade relations — all these and a dozen other features de- 
pend directly on local circumstances. Obscure and mooted 
as the origins of Eskimo economy and its relations to that 
of the Indians remain, no sane theory will ever dispense 
with thoroughgoing ecological considerations. Birket- 
Smith, Speck, and others have thrown into relief the con- 
trast between the subarctic Indians dependent on the 
timber lands for winter subsistence and the litoral Eskimo 
capable of dispensing with the shelter of the forest. Hatt 
had previously pointed out the revolutionary importance 
of snowshoes, which enabled the Indians to hunt freely 
over the inland area of Canada, while hitherto they had 
been obliged to hug the lakes and river courses and main- 
tain themselves by fishing there. Thus, a magnificent per- 
spective opens up on the remote past of an as yet 
undifferentiated subarctic culture, which subsequently 
evolves in divergent directions, the adoption of snowshoes 
leading the Northern Indians along one line of evolution, 
while adaptation to hunting from the ice produced the 

13 Carl Sauer, ' ' American Agricultural Origins : A Consideration of 
Nature and Culture," ALK 279-297, 1936; idem, Aztatlan, Ibero-Ameri- 
eana 1:58 sq., Berkeley, 1932. 


distinctively Eskimo economy." These suggestive views 
are all rooted in a geographical orientation. 

Siberian culture is likewise unintelligible without a 
proper understanding of natural conditions. Why is the 
mode of settlement so different in corresponding latitudes 
of America and Asia? Because, Bogoras explains, the 
Asiatic litoral differs : except in the northeast the shallow- 
ness of the ocean and the low swamps of the coast are 
unfavorable to habitation. How, again, can one understand 
the spread of man and his works in Northern Siberia 
without knowing of the network of connected rivers which 
preclude walking but allow travelers to paddle from one 
river system to another? Similarly, our outlook must be 
wholly askew unless we realize the enormous abundance 
of edible animals, the incredible ease with which fish are 

The aversion from geography is historically compre- 
hensible, but none the less unjustifiable. No ethnologist 
now shares the delusion that culture is man's inflexible 
reaction to his physical surroundings, a view repudiated 
by Eatzel himself. Everyone admits the tertium quid de- 
scribed by Forde (page 250), everyone sees that pure 
environmentalism would imply an exclusively rational 
attitude of humanity such as has been amply refuted. But 
when every possible allowance is made, the obvious fact 
still stands that societies have attained solutions of prac- 
tical problems. Further, this adaptation to the conditions 
of life constitutes the outstanding intellectual achieve- 
ment of mankind. Everything, then, that contributes to 
our insight into the conditions offered by nature deepens 

"F. Boas, "The Eskimo of Baffin Land and Hudson Bay," AMNH-B 
15:75, 357, 1901. K. Birket-Smith, The Caribou Eskimos, 1:233; 2:212- 
233, Copenhagen, 1929. Gudmund Hatt, "Moccasins and their Eolations to 
Arctic Footwear," AAA-M 3:151-250, 1916. F. G. Speck, "Culture Prob- 
lems in Northeastern North America, ' ' Proceedings of the American 
Philosophical Society, Vol. LXV, No. 4, Philadelphia, 1926. 

i'''W. G. Bogoras, "Elements of the Culture of the Circumpolar Zone," 
AA 31:579-601, 1929. 


our insight into the character of culture. ''Human geog- 
raphy," says Forde, "demands as much knowledge of 
humanity as of geography. " ^^ He is right, but he is ad- 
dressing geographers; for the student of culture the 
maxim should be reversed. 


In this context we are neither concerned with the 
psychologizing of the man in the street nor with the inter- 
pretative guesses of the closet philosopher. ''Psychology" 
designates the results and — at least, nascent — conc ept s of 
a branch o f learning specializi ng in the inborn atti tudes 
and behavior of human being s. Precisely because psychol- 
ogy is in principle concerned with what is not culture, its 
interests and those of ethnology must overlap in practice. 
For no one knows intuitively what is and what is not part 
of man's "original nature," everyone's judgment being 
warped by his personal experience. Positive facts ascer- 
tained by either science thus prove a corrective for the 
other, helping to delimit both fields. 

The general trend of research has apparently been 
to narrow the psychologist 's sphere of influence. It is not 
"natural" to point with the index finger, for Pueblo, 
Basin and Plains Indians commonly do so by protruding 
their lips. African and Oceanian observation suggests 
that ' ' even at the crawling and toddling stage the primi- 
tive child can seldom be left alone and must spend a large 
part of its time balanced on its mother's hip." But con- 
tact with American Indians limits the generalization and 
exposes the practice as conventional, Tylor, though rec- 
ognizing handshaking and kissing as far from inevitable 
human responses, still believed in a gesture-language that 
was "essentially one and the same in all times and all 
countries." But the signs of Queenslanders and Sioux 

1^ C. D. Forde, Habitat, Economy and Society, 465. 


lend little support to psychic unity. Again, an American 
native draws a knife towards his body, the African Negro 
whittles in the opposite direction. Sometimes even neigh- 
boring groups of the same race display striking diversity 
in simple everyday matters : the Pueblos carry loads on 
their heads, the Havasupai never do ; most American na- 
tives chip stone tools, but the coastal tribes of the North- 
west only peck and polish them. Emotional expression 
varies as much as do motor habits. On the Plains of North 
America black betokens not grief but victory; in many 
tribes weeping reflects not sorrow but a ceremonial duty ; 
and so forth. Boas thus sums up an immense range of 
observations when he declares that the plasticity of the 
human o rganism makes it follo w the cultural pattern with 
which it has become identified." 

At first blush such conclusions seem to make the co- 
operation of psychology and ethnology a one-sided affair. 
Men all live in society, and if society so deeply affects their 
outlook on behavior, all the illumination seems to come 
from the cultural side, with nothing given in return. Thus, 
Wissler argues that individual differences of motor habit 
are not a cause of relevant tribal differences in basket- 
work: "Culture differentiation and psychological differ- 
entiation . . . run in relatively independent cycles, ' ' the 
former explicable only in historical terms.^* But even if 
we fully accept such a dictum, it would not prove the 
futility of psychology for our purposes. While the hier- 
archical scale of the sciences does not properly express 
their relations, matters are not mended by simply revers- 

" Leslie Spier, "Havasupai Ethnography," AMNH-P 29:329, 1928. 
Brenda Z. Seligman, "The Incest Barrier: Its Rolo in Social Organiza- 
tion," British Journal of Psychology, 22:259, 1932. E. B. Tyler, Re- 
searches . . . , 45, 53 f . F. Boas, * ' The Effects of American Environment 
on Immigrants and Their Descendants," Science, 84:522 sq., 1936; idem, 
Primitive Art, 145 sq. A. L. Koeber, "The Arapaho," AMNH-B 18:417, 
1907. Otto Klineberg, "Notes on the Huichol," AA 36:459, 1934, 

18 Clark Wissler, "Material Cultures of the North American Indians," 
AA 16:501, 1914. 


iiig the ladder nor do we progress by clinging to a single 
rung. Actually the analogy is misleading because the in- 
terrelations of the sciences are of a quite different char- 
acter. Of course culture cannot be subsumed under mental 
processes, otherwise there would be no ethnology at all — 
no more than there would be a biology if organisms grew 
and bred merely by the laws of gravitation and of chemi- 
cal affinity. When the autonomy of our subject is once 
granted, however, declarations of independence grow 
repetitious. Our position towards psychology should cor- 
respond to that assumed towards geography : we cannot 
explain all of our phenomena through it, but neither can 
we explain them fully without it. 

In the first place, then, psychology, working from 
the opposite end, rules out interpretations that lure and 
ensnare the unwary ethnologist. Can exogamy be reduced 
to an innate incest sentiment ? The modern critique of the 
instinct concept explodes the theory and restricts the 
problem to its proper cultural sphere. On the other hand, 
mental manifestations bear constructively on such peren- 
nial themes as the diffusion controversy. How did the fig- 
ments of mythology take form? Dreams parallel them and 
provide a possible source. That they themselves in part 
reflect the regnant folklore offers no ultimate explanation. 
But their content recurs in remote areas with analogous 
interpretations yet in conditions virtually precluding dis- 
persal from tribe to tribe. Such distribution, then, sup- 
ports the possibility of independent origin.^^ This is not 
a relapse into reliance on a vague "genetic law": specific 
ideas of mythology are connected with specific mental 
phenomena or their antecedents. Naturally, what we need 
for clearer insight is fuller information on savage dreams. 

However, a general human explanation breaks down 
when diversity, not likeness, is to be explained. Here psy- 

15 C. G. Seligman, ' ' The Unconscious in Eelation to Anthropology, ' ' 
British Journal of Psychology, 18:377 sq., 1928. 


chology might still conceivably render service by tracing 
group differentiae to original nature, but to nature ra- 
cially circumscribed. Actually, to be sure, any differences 
of endowment that may exist are not congruous with cul- 
tural differences. This, however, is far from maintaining 
that the races are alike. To quote Thomas' reasonable 
conclusion: ''It is to be emphasized . . . that there are 
no proofs that the mind is of precisely the same quality 
in all races and populations, and no such claim is made 
by anthropologists. It is not improbable that there is a 
somewhat different distribution of special abilities, such 
as mathematics, music, etc. ' ' ^'^ Here is the rub. Such varia- 
tions may even jointly account for only a small fraction 
of culture, but insofar as they exist they are verae causae. 
The findings of the Cambridge Expedition to Torres 
Straits, while largely inconclusive, do suggest a higher 
threshold of pain for the Papuans than for the Caucasians 
— a result later paralleled among the Vedda. More fully 
corroborated, such inferences would help account not in- 
deed for the mutilations to which many savage societies 
subject their initiates, but for the fact that such disfigure- 
ment could ever have been conceived and could ever have 
succeeded in persisting. Again, while the development of 
African as against native American music may be largely 
due to borrowing from Egypt and India, we cannot a 
priori exclude a higher innate aesthetic sensitiveness. 
Everything urged against overenthusiastic testers may 
be granted. If native ability is to be determined, the 
groups compared must be relatively pure; we must dis- 
count for environment in choosing tests; affective and 
motor factors may be at least as potent as the intellect. 
But when all this is not merely conceded but emphasized, 
specific racial differences remain a possibility with poten- 
tial consequences for cultural differentiation. The only 
science equipped to deal with this question is psychology ; 

20 W. I. Thomas, Primitive Behavior, 799, New York, 1937, 


and its controlled results are equally important to eth- 
nology, whether they culminate in a proof or in a refuta- 
tion of racial faculties and disabilities. In the one case 
we acquire a potential explanation of empirical differ- 
ences; contrariwise we are thrown back upon historical 
accident as the sole determinant. 

Original nature may differ individually as well as 
racially, and here psychology offers a positive contribu- 
tion of incalculable importance. Galton's notion of con- 
genital individual variability, seized by Boas and others, 
has revolutionized the outlook on savage life. What had 
hitherto been a static phenomenon now appeared instinct 
with the germs of change; automata obedient to custom 
gave way to human beings paralleling the gamut of emo- 
tional and intellectual values familiar in civilization. 
These deviations, moreover, are sometimes demonstrably 
significant for society, and it is the joint task of psychol- 
ogy and ethnology to define the interplay of personal and 
social determinants. 

One clear-cut result has been attained. There is a 
social selection of personalities, discus sed by Thurnwald 
under the caption *'Siebung" (sifting) and independently 
by several British and American authors.^ The madcap 
hero of a horde of warriors is the ruffianly bravado of a 
more staid society; the musing sage of one group is a 
maladjusted milksop in a mining camp. Indeed, as Selig- 
man remarks, savages invest with prestige persons we 
should clap into an asylum for the insane. Since this is a 
general human process, the records of literate peoples 
might well be scrutinized from this angle. Savage society 
presents no more striking case than the ascendancy of 

21 E. Thurnwald, article "Siebung" in M. Ebert's Reallexikon der 
Vorgeschichte, Berlin, 1929; idem, Die menschliche Gesellschaft, 4:264f., 
309 sq., Berlin u. Leipzig, 1935. Barbara Aitken, "Temperament in Na- 
tive American Eeligion," JRAI 60:363-387, 1930. Ruth Benedict, "Con- 
figurations of Culture," AA 34:1-27, 1932. Ralph Linton, The Study of 
Man, 443-487, New York, 1936. C. G. Seligman, op. cit., 374. 


Samuel Johnson in British life contemporaneously not 
only with Lessing and Voltaire on the Continent, but with 
Adam Smith, Hume and Gibbon in Scotland and England. 
For the British strainer, to use Thurnwald's analogy, 
these towering intellects were mere dregs. 

Social sifting may effect biological selection. Unre- 
mitting preference for preferred traits, say skill in the 
chase, might culminate in the elimination of the ''unfit," 
hence lay the foundation of hereditary group differences. 
If so, these would be demonstrable only by psychological 
techniques. In the light of present knowledge, meagre as 
it is, an alternative result seems more plausible, viz., 
the survival of the ''inferior" types through protective 
mimicry. They assume naturally lacking virtues, persist 
in ordinary circumstances, and fail only when unmasked 
by a major crisis. 

Stimulated by psychology, then, the ethnologist has 
shown that indivi dual va,riation exists in rude societies; 
and that societies, re cognizing diversity, respond dif- 
ferently to the same variation, one group exalting the 
very deviation that is derided by its neighbors. Further, 
the oc cu rrenc e of deviants accounts for the undoubted 
fa ct of change. As Morgan remarks, personal experi- 
ences — rooted in mental idiosyncrasies — may "deeply 
condition the individual, sometimes so deeply that if the 
experience is at variance with a tribal . . . belief, the in- 
dividual will retain his own variation. There can be no 
doubt that this is a very significant means of modifying 
a culture." Whether such an alteration is achieved de- 
pends obviously on the strength of the aberrant person- 
ality, with the social receptiveness as the co-determinant. 
The ethnographer takes cognizance of the deviation re- 
gardless of whether it actually leads to a new tradition. 
Thus, in northwestern California, where women are the 
shamans, supernatural revelations normally exclude the 
military tenor so dominant on the Plains. Yet a sporadic 


gallant will sock a lonely spot, acquiring as a spirit's 
blessing strength as well as wealth. Because of the tribal 
ideology, "most men's ambition did not lie in this di- 
rection"; yet even the rare presence of the impulses 
typically expressed in Eastern visions has cultural mean- 
ing, because here lies the possible germ of a traditional 
change of shamanistic purpose and routine. Other cases 
are less hypothetical. Among South American aborigines 
the Guarani are noted for their periodic migrations in 
search of an earthly paradise. The prophets mimicked 
the behavior of mythical heroes and to that extent were 
conditioned by their environment. But why did such 
characters arise only now and then? And what made 
their tribesmen welcome these ''veritables acteurs d'un 
drame mythique"? There were evidently leaders and fol- 
lowers; further, the leaders themselves differed in the 
specific nature of their message, which was thus not 
traditional. More clearly, though not by any means 
clearly enough, we recognize the force of qualitative 
variations in the personalities of the North American 
"messiahs" who periodically arose soon after the first 
contact with civilization.^^ 

Why does our insight into such events remain im- 
perfect? Evidently because our reports are inadequate 
on the psychological side. The biographical details, as a 
rule too sketchy even for a layman, hardly ever sufiice 
for a ** clinical" picture of the actors. Unfortunately the 
scientific understanding of personality is as yet inade- 
quate. The field has been largely cultivated by psychia- 
trists, often phenomenally penetrating in their intuitive 
grasp of a patient's needs, but weak in conceptualization 

22 Wm. Morgan, ' ' Human Wolves among the Navaho, ' ' Yale University 
Publications in Anthropology, 11:40, New Haven, 1936. A. L. Kroeber, 
Handiook of the Indians of California, 67 f. Curt Nimuendajti, "Die 
Sagen von der Erschaffung und Vernichtung der Welt als Grundlagen der 
Religion der Apapoctiva-Guarani, " ZE 46:284-403, 1914. A Metraux, 
La religion des Tupinambd, 217, Paris, 1928. 


and critical judgment. Their classifications, possibly of 
great therapeutic and even theoretically of heuristic 
value, must be treated with reserve. Thus, we cannot fol- 
low Seligman in his wholehearted use of the psycho- 
analytic distinction between "extravert" and '4ntra- 
vert" types for explaining shamanism. In the first place, 
for cautious psychologists these categories strictly apply 
only to extreme forms of mentality, normal personality 
displaying the commingling of extravert and intravert 
elements. Secondly, ethnographic experience does not 
bear out the contention that savage peoples are predomi- 
nantly extravert but rather suggests the very same dual 
character among them as among ourselves. To quote one 
of our keenest observers about the Maricopa of Arizona : 
''Like all Indians, these people can sit endlessly saying- 
nothing and looking fixedly into space . . . On the other 
hand, they can become excessively talkative and are at 
all times ready to joke and laugh, "^^ Brooding and 
boisterous self-expression are thus not at all mutually 

But if the psychiatrists are precipitate and the 
academic students of personality overdilatory, is not the 
ethnographer left in the lurch? Yet the situation is less 
desperate than it seems. Once freed from the fallacy 
that psychology is to explain culture without residue, 
the ethnographer profits, first of all, by the psycholo- 
gist's ''case method." He must not forget that the cul- 
ture he investigates is a living reality only as mirrored 
in its bearers ; the two are as inseparable as the sides and 
the angles of a triangle. In other words, the culture by 
itself is an abstraction; the reality is adequately de- 
scribed by exhibiting samples of personality responding 
to the social setting. The correct procedure is to give an 
adequate definition of both. The better observers have — 
sometimes quite intuitively — groped towards such char- 

23 Leslie Spior, Yuman Tribes of the Gila River, 327, Chicago, 1933. 


actorization. Rasmussen's demonstration of an Eskimo 
skeptic helps delimit shamanism as a cultural phenome- 
non, as does Bogoras' account of Siberian inverts as 
shamans. Radin's biographies of Winnebago Indians 
serve the same purpose. Hallowell's frankly program- 
matic study of Western Ojibwa consciously applies 
modern clinical concepts to individual cases. In one of 
his instances a powerful wonder-worker satisfies his in- 
cestuous cravings and justifies them in terms familiar 
to his group ; he is merely carrying out the dictates of a 
tutelary spirit, which cannot be disobeyed without a 
sense of sin. His rationalization has the strongest efficacy 
from the aboriginal viewpoint, yet it is interesting that 
even in this exceptionally favorable instance the flouting 
of custom was unable to quell some sense of guilt.^* 

Here, as usual, the picture is blurred by the meagre 
description of the actor's personal traits. The merit of 
Hallowell's approach lies in its concentrating on the is- 
sues involved. Addressing psychiatrists, he emphasizes 
the significance of ethnographic data when garnered by 
an observer "aided in formulating his problems by the 
psychiatrist sensitive to the implications of culture." 
Our point is that such formulation may be profitable for 
the student of culture provided the psychiatry is sound. 
This means that it must progressively fight shy of catch- 
words, substituting empirically definable descriptive 
traits. Radin cogently argues for the reality of an in- 
tellectual class among primitive groups, and on this 
assumption plausibly deduces the possibility of mono- 
theism as an individually recurring creed.^^ But the 
psychological identification of such believers is less satis- 

24 W. Bogoras, "The Chukchee," AMNH-M 11:415, 426 sq., 441, 450 
sq., Leiden, 1909. Knud Easmussen, Intellectual Culture of the Caribou 
Esliimo, 58 sq., Copenhagen, 1930. A. I. Hallowell, "Psychic Stresses and 
Culture Patterns," American Journal of Psychiatry, 92:1291-1310, 1936. 

25 Paul Eadin, Primitive Man as a Philosopher, 342 sq., 366, New York 
and London, 1927. 


factory. They are supposed to combine an eminently 
religious temperament with speculative capacity; but 
what precisely is eminent religiousness? Do the historic 
figures commonly credited with this characteristic con- 
form to a single mental type? What is the common de- 
nominator of the saintly recluse and the propagandist, 
the bold inventor and the zealous preserver? The psy- 
chology of the future will, we hope, supplant vague 
terms drawn from vulgar experience with the subtler 
concepts of a more refined observation. 

Although personality stands in the foreground of 
current discussion, the contacts with psychology may 
be viewed from another angle. Psychology may help 
even when it fails to shed direct light on an ethno- 
graphic phenomenon. According to Wissler we saw that 
cultural diversity has a solely historical basis. But the 
subject Wissler discusses is the direction of movement 
in basketwork, in other words, motor processes, a con- 
cept ethnographically significant because cultures vary 
with regard to such processes, yet obviously one trans- 
ferred from psychology. Further, Wissler cites evidence 
to show the difficulty of an individual change from sew- 
ing in a clockwise to counterclockwise direction or vice 
versa, even though the initial choice of either alternative 
be accidental. He is thus invoking another psychologi- 
cal concept, habit; and habit is not without pertinence 
to our cultural problems. According to Weltfish, all 
modern American Indian basketry weaves occur prehis- 
torically within the same areas. This amazing stability 
is surely to be connected with the persistence of habits 
once formed, as Boas suggests. In other words, the vary- 
ing processes in basketry may well have a historical 
basis and yet be related to vital facts of psychology .^° 

28 Gene Weltfish, "Problems in the Study of Ancient and Modern 
Basket Makers," AA 34:108-117, 1932; eadem, "Prehistoric North Ameri- 
can Basketry Techniques and Modem Distributions," AA 32:454-495, 1930. 


III this connection may be cited Fechner's ideas on 
experimental aesthetics." He may be quite wrong in 
postiUating his '^ golden section" — an ideally pleasing 
ratio of the sides of a rectangle. But rectangles appear 
as a distinct feature in Plains Indian decoration; we can 
measure their sides, and determine tribal preferences. 
What if Fechner's ratio turns out to be fallacious 1 
Wliat if the observed preferences are due to historical 
chance? If real, they aid towards a finer discrimination 
between regional geometrical styles: the psychological 
stimulus proves ethnographically profitable. 

Psychological factors loom large in primitive reli- 
gion even apart from the obvious phenomena of leader- 
ship. Dreams are of vital relevance, but again mental 
variations of definable character may assume ethno- 
graphic importance. The dream reports of Colorado 
River Indians are steeped in a mythological atmos- 
phere ; those of their kinsfolk on the Gila describe actual 
sleep imagery. The latter can no more be disregarded 
than the former when the tribal contrast hinges pre- 
cisely on this distinction. But we are confronted not 
merely with free sleep experiences and those tradition- 
ally patterned. There are visions as well as dreams, and 
whole areas differ in whether they stress one or the 
other form of experience. As a matter of fact, ''visions" 
are sometimes either nothing of the sort but hallucina- 
tions of an auditory nature, or at least are associated 
with nonvisual components. This is not a trivial point 
considering the incredible frequency of sacred, i.e., re- 
vealed, songs. Among the Navaho, again, the capacity 
to pass into a trance characterizes the culturally recog- 
nized class of diagnosticians. Those aspirants fail who 
cannot ''withdraw sufficiently from conscious awareness 

F. Boas, Primitive Art, 149. L. Spier, "Havasupai Ethnography," 
AMNH-P 29:136, 1928. 

2" Gnstav Theorlor Fechner, Vorsehule der AesthetiTc, Leipzig, 1876. 

Ch. Lalo, L'esthetique experiment ale, Paris, 1908. 


to allow any stream of unassorted ideas to pass through 
their mind, or for a picture to form itself. ' ' The ethnog- 
rapher cannot well make shift without some recourse to 

But the psychology should be accurate. The failure 
to attach a precise and accepted meaning to the terms 
** trance," "frenzy," ''orgy," largely vitiates Benedict's 
interesting contrast of non-Pueblo and Pueblo cultures 
of North America. The supposedly Dionysian worship 
of the former, rooted in their "ecstatic" visions, consti- 
tutes a major differentia of the scheme. But the revela- 
tions in question conform to this pattern only to a mod- 
erate degree. The Crow Indians do employ the word 
for "intoxication" when describing the hypnotic state 
of a Sun Dancer after protractedly riveting his gaze on 
a sacred effigy. It also embraces the condition of a per- 
son whose indwelling supernatural threatens to pass out 
of the host 's mouth. But the term is not extended to the 
more customary supranormal phenomena, and the nar- 
ratives of such experiences only rarely imply ecstasy. 
Still less is the ecstatic timbre discernible when an 
Ojibwa boy seven years of age retires to fast at his 
elders' prompting and after meticulous surveillance ac- 
quires the sort of guardian spirit his instructors permit 
him to accept.^^ 

The following principles, then, hold. Neither science 
can be reduced to the other, but cross-fertilization is 
practicable and helpful. Ethnology enlarges the psychol- 
ogist's ken, demonstrating the scope of social pattern- 
ing in individual behavior. The ethnologist, inevitably 
bound to use terms relating to the mind, uses the de- 

28 A. L. Kroeber, Handboolc of the Indians of California, 753, Wash- 
ington, 1925. Leslie Spier, Yuman Tribes of the Gila Biver, 238, 257, 
Chicago, 1933. William Morgan, "Navaho Treatment of Sickness: Diag- 
nosticians," AA 33:390-402, 1931. 

29 Ruth Benedict, ' ' Psychological Types in Southwestern Culture, ' ' 
ICA 23:572-581, New York, 1930; eadem, Patterns of Culture, New York, 
1935. Paul Radin, "Ojibwa and Ottawa Puberty Dreams, ALK 233-264." 


terminations of scientific psychology prophylactically 
against the snares of vulgar psychology; and construc- 
tively in refining his analysis of regional differences and 
of particular processes. 

M eta-Ethnography 

The anthropological expeditions that followed on 
the heels of the Tylor period yielded publications which 
differed notably from earlier accounts. Addressed to a 
scientific public, they generally lacked literary quality, 
which indeed could hardly be preserved because of the 
technical detail that demanded registry. Since, further- 
more, the aim was to render the material available for 
comparative work by scholars, a stereotyped arrange- 
ment evolved to facilitate reference. To take at random 
two representative monographs on very different 
peoples, the Maidu (California) and the Yuchi (South- 
eastern United States), both authors begin with certain 
generalities as to habitat, demography, and history; 
then proceed to material culture, art, social organiza- 
tion, and the life cycle ; and close with religion and my- 
thology.^" It was both natural and essential that field 
workers, apart from any chance observations they could 
make, should inquire into matters emphasized by the 
great theorists of the preceding period. The results in- 
evitably varied with the writer 's ability : born observers 
were not hampered by the fashionable procedure, which 
in the hands of others still yielded worth-while though 
uninspiring fruits. Its dangers are being vastly over- 
rated at the present time: the true creative spirit is 
never cramped by formal restrictions and the dullard 
remains petty whether he devotes himself to the un- 
crystallized phases of culture or to free verse. 

soEoland B. Dixon, "The Northern Maidu," AMNH-B 17:119-346, 
1905. Frank G. Speck, "Ethnology of the Yuchi," University of Pennsyl- 
vania, Anthropological Publications of the University Museum, 1:1-154, 


With growing insight into savage existence it be- 
came clear that the traits obvious either through the 
natives' or the theorists' emphasis did not cover the 
whole of social life. Boas early drew his students' at- 
tention to such amateur records as the Lapp Turi's 
reminiscences and Rasmussen's first book. In 1922 some 
twenty Americanists under the leadership of E. C. Par- 
sons, recognizing the deficiency of current monographs, 
deliberately set out to correct this by fictitious biograph- 
ical and impressionistic sketches based on observation 
of their favorite tribes. In her preface the editor thus 
criticized contemporary technical literature: "The com- 
monplaces of behavior are overlooked, the amount of 
^common sense' is underrated and the proportion of 
knowledge to credulity is greatly underestimated." In 
the same year Malinowski proclaimed the need for going 
beyond ''the collection of crystallized, ethnographic 
data." When, therefore, a decade later Margaret Mead 
issues a clarion call for recognition of ''all parts of a 
culture, and not merely those which present the super- 
ficial appearance of having greatest form," this mani- 
festo is somewhat belated. The technique she prescribes 
for such researches, viz., the systematic observation of 
countless concrete instances, is excellent; but her goal 
is not new and she seems strangely unaware of previous 
striving towards it. Our nonhistorically minded younger i 
generation often rediscover America, and it is perhaps / 
cruel to disturb their illusions.^^ 

While Mead t hrows into^ r^?^_?i?P®^s_ that escape 
notice because janslresse d by the n atives, Ben edict calls 
attention to the ' ' dominant drives' ' of c ultu res. Certain 
integrative principles shape the raw material of cus- 
tom; no more mysterious than art styles, they "are as 

81 E. C. Parsons (editor), American Indian Life, 2, New York, 1922. ' •«' 
Br. Malinowski, Argonauts . . . , 20, 1922. Margaret Mead, "More Com- 
prehensive Field Methods," AA 35:1-15, 1933. 


characteristic for individual areas as are house forms 
or the regulations of inheritance,"^^ hence demand rec- 

This view has both philosophical and ethnological 
affinities. In the history of psychology its cognate is the 
critique of associationism, with its insistence that the 
whole is more than a summation of its parts. In the study 
of civilization Benedict recognizes such forerunners as 
Dilthey and Spengler, but urges that the higher cultures 
they have investigated are too complex to reveal their 
essence. Ethnographically, she must be grouped with all 
who study the connection between elements rather than 
the discrete elements. Thus, contemporaneously with 
Benedict's earliest essay in this field an article by Bar- 
bara Aitken distinguished Pueblo from Eastern Wood- 
land religion not in point of content but for its difference 
in emphasis. The Easterners had allowed the individual- 
istic temperament to set the tone, while the Pueblos 
yielded precedence to the social temperament.^^ While 
this gives a psychological character to the cultural dif- 
ferentiation, several earlier writers had demonstrated 
patterns especially in ceremonialism without recourse 
to basic mental facts. Boas' pupil Haeberlin, for exam- 
ple, showed that Pueblo religion was characterized by 
the ever-recurring idea of fertility, which sharply set 
off the identical ritual as performed by Pueblo and 
Navaho, respectively. 

Benedict's aim, however, transcends her predeces- 
sors' in envisaging the totality of culture. This would 
seem to bring her into the functionalist camp, but she 
insists on marching ahead of its vanguard. True, Malin- 
owski has shown each element playing its part in the 
whole ; but what manner of whole is it within which the 

32 Ruth Benedict, "Configurations of Culture," AA 34-1-27, 1932. 

33 Barbara Aitken, "Temperament in Native American Religion," 
JRAI 60:363-387, 1930. 


traits function? And while he has shown reciprocity as 
a ** basic behavior trait" of Melanesians, Benedict would 
like him to disclose the ''fundamental attitudes" of 
which such behavior is the outward symbol. 

The goal, then, is to fix the stylistic peculiarities of 

c ultures and to expres^them m psychological terms^ 
Benedict disarms criticism by admitting tha t many cul - 
tures have failed to achieve a thoroughgoing integration ; 
and by narrowing the scope of her principle so as to 
exclude technology. She thus seems to vacillate between 
a consistent totalitarianism and the recognition of ' ' dom- 
inant drives." Moreover, to date she has fully discussed 
only three peoples from this point of view.^* All this 
makes it peculiarly difficult to assess her contribution. 

Against the totalitarian version of the doctrine we 
repeat the criticism already advanced against Malinow- 
ski and Radcliffe-Brown. Even if te chnology is barred, 
it remains an unproved hypothesis that all the residual 

items are integrated. On the other hand, the search for 
dominant drives is a fruitful idea ; and the picture Bene- 
dict herself offers of a Pueblo people is both vivid and 
suggestive. The subject has not, indeed, been so uni- 
formly neglected as she implies: the wealth obsession 
of northwestern California, the lure of military glory 
on the Plains, are ethnographic commonplaces. But Bene- 
dict's gospel will make us seek leitmotifs in more recon- 
dite spots and should prompt scrutiny beyond their gross 
manifestations. In any case, her discussion concentrates 
thought on several worth-while questions. In what cir- 
cumstances do people achieve a distinct ideology? Are 
configurations, for instance, to be found on the Fuegian- 
Basin — Negrito level? And what is their total influence 
on the several cultures that disclose it? 

Patterns, then, if not all-pervasive, may be accepted 
as potentially extending over major cultural blocks. How 

•^* Patterns of Culture, New York, 1935. 


are they to be defined? Here we feel Benedict has been 
somewhat kicking in caution. The reality she seeks, which 
she aptly compares to an art style, shares with such 
styles a comparative elusiveness. The danger of impres- 
sionism is averted only by the ample and concurrent testi- 
mony of several good witnesses. That is why the Eskimo, 
described in detail by so many first-rate investigators, 
would form an ideal subject for configurational treat- 
ment, one whose omission is hardly intelligible. That is 
why Benedict's account of Zuni, resting as it does on 
most ample documentation by herself and other observ- 
ers, is so satisfactory; why the contrast between Pueblo 
and Eastern Indians, sensed as it is by Aitken, Kroeber, 
and others, ranks as a fact, not mere fancy. But it also 
indicates why Benedict erred in choosing for one of her 
three major subjects the Melanesian Dobu, whose sup- 
posed pattern is derived from a single source with un- 
impressive documentation. 

There is another pitfall. In setting off configurations 
against one another there is an inescapable te ndency to 
ov erweigiit differences ; the writer, intent on distinguish- 
ing, distorts the total picture in favor of his specific 
criteria. This is one of the errors Parsons' symposium 
was designed to correct. As she wrote in her preface: 
"Commonly the interesting aspects are those which differ 
markedly from our own culture or those in which we see 
relations to other foreign cultures we have studied." 
A recent Chinese visitor to Zuni offers the same com- 
ment: he finds that Benedict's differentiae are correct 
but that their emphasis throws the picture out of focus.^'* 
Yet if anything should give a true perspective it is the 
pattern that animates the whole. 

Yet Benedict's description of Zuni has found favor 
with the best judges. How is the contradiction to be 

^"Li An-che, Zuni: Some Observations and Queries," AA 39:62-76, 


reconciled? The point is that Li An-che's stricture 
touches not the picture as given but Benedict's epitome 
thereof. She resembles a physicist seeking enlightenment 
about electricity not from the empirical relations of 
electrical phenomena but from a concise dictionary defi- 
nition. Borrowing, for example, a Nietzschean antithesis, 
Benedict defines the Pueblos as ''Apollonians" who insti- 
tutionalize ''sobriety and restraint in behavior," while 
the "Dionysians" of the Plains favor ''abandon and 
emotional excesses." Hence, the contrast of ritual formal- 
ism with personal shamanism; of subdued and frenzied 
mourning ceremonial; of diffident hiding from public 
notice and boastful competitiveness. 

Now this is, indeed, to oversimplify a real antithesis. 
We have already repudiated the notion that the Eastern 
Indian's vision conforms to the pattern of orgiastic 
excess. Groing further, we find in much of Plains religion 
a ritual formalism that attains Apollonian degree. The 
opening of a Blackfoot sacred bundle has no element of 
frenzy, let alone orgy. The celebrants assume positions 
not to be altered before there is ritual dispensation. They 
burn incense, shake rattles, and sing interminable chants. 
Even the mildly dramatic imitation of buffalo is highly 
stereotyped. The performance is as solemn and sober as 

Our judgment, then, culminates in qualified approval. 
There are cultural leitmotifs, and their scope should be 
ascertained regionally and within particular cultures. 

But an adequate definition of these patterns is as yet a 
thing for future research. 


Being a science, ethnology implies an orderly ar- 
rangement of its data, the verifiability of its findi ngs, a 

36 Clark Wissler, ' ' Ceremonial Bundles of the Blackfoot Indians, ' ' 
AMNH-P 7:204 sq., 1912. 

loe:ical basis for its conclusions. But in conforming to 

the canons of all science, it must not adopt the particular 
Techniques of physics, biology, or geolog y ex cept where 
the cultural data as su ch demand such recourse. 

Scientific procedures are not gadgets preconceived 
and thrust into reality in the hope of a catch. They evolve 
and are applied spontaneously as problems arise in the 
mind of a thinker saturated with his theme; and then 
no arbitrary boundary will stop his sally into the un- 
known. When Laufer concentrates on porcelain, it is not 
in the spirit of sinological antiquarianism. "Porcelain" 
is conceptually defined as a pottery type with distinctive 
glaze. Its history, then, is that of Chinese glaze and 
earthenware. The differentia can be studied only by the 
aid of chemistry, physics, mineralogy, and the history of 
the Near Orient ; and it is thus that its rise is determined. 
Curiosity is not yet exhausted, and as a by-product we 
get a dissertation on the potter 's wheel and its sociologi- 
cal correlates. No one could have foreseen the questions 
opened by the incipient inquiry nor the techniques re- 
quired for answering them.^^ 

The student of culture, then, will be unable to predict 
what he may need, let alone what his colleagues may 
require for their purposes; but his inescapable duty is 
to define his concepts and arrange them in a spatial, 
temporal, and causal context. 

Concepts must be clear and rigid, rising above the 
fluidity and vagueness of the raw phenomena. Here lies 
a major difficulty for all thinking. "Bronze" is copper 
deliberately alloyed with tin to enhance the hardness of 
the artifact. How sure, however, can we be in a given 
case that the percentage of tin is not due to chance? 
"Tanning" is a process involving chemical as well as 
mechanical changes in the skin. But what if excessive fat 

3T Berthold Laufer, ' ' The Beginning of Porcelain in China, ' ' FMNA- 
PAS 15:79, 183, Chicago, 1917. 


stimulates an unplanned chemical action? And, if so, who 
can be sure that the natives fail to recognize the effect of 
this factor? A ''moiety" is etymologically one of comple- 
mentary halves; most commonly, one of the two exoga- 
mous clans of a tribe, but the term has been extended, 
e.g., to two complementary ceremonial groups not affect- 
ing marriage. How can the concept do service among the 
Timbira of Brazil, who bisect all the tribe on two distinct 
principles and its male constituency on still other lines? 
Again, in the Southeastern United States one tribe has 
exogamous halves, while its neighbor's moieties do not 
regulate marriage, yet otherwise share the same func- 
tions. The "moiety" label does not matter, but the as- 
sociated concept is all-important. For an exogamous 
moiety is a species of the genus "clan." We may well 
ask, then, whether a coexisting moiety and clan system 
are genetically related. But to link nonexogamous with 
exogamous moieties is warranted only if extraneous con- 
siderations suggest the connection; otherwise the common 
name deceives us into assuming real unity. 

The clarification of concepts, then, directly gauges 
scientific progress. Morgan's tilt against MacLennan 
brought into relief the nonlocal character of the clan as 
then known. Later, in California, E. W. Gifford and Wm. 
D. Strong discovered landowning and politically auton- 
omous groups reckoning descent through the father and 
imposing exogamy. If the emphasis is put on unilateral 
descent and exogamy, these "lineages" are clans; yet 
evidently such independent units differ from their non- 
localized Iroquois counterparts. Along lines already an- 
ticipated by MacLennan, Gifford and Strong explained 
the ordinary form of clan by the fusion of lineages. In 
Australia Radcliffe-Brown found similar patrilinear 
"hordes" coexisting with a clan system of the usual type. 
Steward, recognizing the affinity of Ona bands with Cali- 
fornian lineages and Australian hordes, propounds the 


question of a common underlying antecedent. He visu- 
alizes these ** bands" as one subtype of the general ''uni- 
lateral" category, the ''clan" representing another.^^ 
Goldenweiser injected another point: An Iroquois clan 
includes both actual and assumed descendants of one an- 
cestress through females; and ethnologists had been 
wont to speak of inheritance following clan lines. With 
nice discrimination Goldenweiser showed that the Iro- 
quois distinguished the actual kin within a clan, privi- 
leges being primarily transmitted from blood relative 
to blood relative. In other words, the Iroquois "clan" 
comprises several matrilineal groups of true kindred, 
and these are the rightful claimants to title and office. 
This notable contribution to our stock of sociological 
ideas has proved applicable to Pueblo society, where the 
same distinction is vital.^^ 

But this once more raises a problem. Goldenweiser 's 
"maternal families" are the matrilineal equivalent of 
Steward's patrilineal bands inasmuch as both are exoga- 
mous units reckoning descent unilaterally. But whether 
on this basis they are best classed together as "lineages" 
or are to be distinguished by virtue of autonomy is a de- 
batable question. Thus, new knowledge offers an ever- 
recurrent challenge to the ethnologist's acumen. His 
data, complex and oscillating, constantly menace re- 
ceived categories. Precisely for that reason it is one of 
his noblest tasks to bring order into this hodgepodge, 
barring sham problems and smoothing the path for real 

38 E. W. Gifford, * ' Miwok Lineages and the Political Unit in Aborigi- 
nal California," AA 22:389-401, 1926. W. D. Strong, "An Analysis of 
Southeastern Society," AA 29:1-61, 1927. A. R. Eadcliffe-Brown, "The 
Social Organization of Australian Tribes," 29. J. H. Steward, "The 
Economic and Social Basis of Primitive Bands," ALK, 331-347. 

3^ A. A. Goldenweiser, On Iroquois WorTc; Summary Report of the 
Geological Survey, Ottawa, Canada, 464-475, 1912; 365-373, 1913. Elsie 
Clews Parsons, Hopi Journal of Alexander M. Stephens, l:xxxiv, New 
York, 1936. 


Apt concepts, then, are a primary goal. Without 
them even a simple charting of distributions is invali- 
dated. But from another angle the spatial extent of a 
phenomenon defines the concept itself. Our notion of the 
Trobriand Islanders differs from our notion of the Lili- 
putians or the Yahoos because the first are at least im- 
plicitly thought of as rooted in a definite part of the 
globe; and what is true of whole peoples applies to the 
single items of their social existence. What any one of 
them signifies in the total course of human history de- 
pends on where it occurs and does not occur. Bronze, 
stone masonry, a position system of arithmetical notation 
— all these would denote utterly different values if typi- 
cal of the Andamanese, Fuegians, and Tasmanians. 

And as distribution influences appraisal, so it leads 
directly to questions of causality. In California, the Pai- 
ute of Owens Lake prize the pinenut as their staple food, 
while their brethren of the Oregon boundary line con- 
sider it of minor importance. Why this difference be- 
tween subdivisions of a tribe? Because in the north 
Pinus monophylla is very nearly lacking.** Is the expla- 
nation trivial? Well, for a philosophical inquirer it in- 
volves an attack on the riddles of the universe. Here are 
two groups linguistically so close that they separated 
hardly more than several centuries ago. To compare 
them is thus to measure the rate of cultural differentia- 
tion. Further, according to a popular theory, economic 
factors primarily determine the whole of social life. Here, 
then, we can directly test the effects of a shift from or 
to reliance on the pinenut. 

Knowledge of distribution regularly merges in his- 
torical interpretations. The American Indians are es- 
sentially a single race that gradually came to occupy all 

*o J. H. Steward, ' ' Ethnography of the Owens Valley Paiute, " UC 33 : 
241, 1933. I. T. Kelly, "Ethnography of the Surprise Valley Paiute," 
UC 31:99, 1932. 


of the New World. From British Columbia to Cape 
Horn countless modern tribes prescribe sticks as head- 
scratchers in ceremonial situations varying in detail but 
without overthrowing the basic unity of the concept. It 
is, then, no perverse historical mania but a normal im- 
pulse to give temporal meaning to the distribution. Is 
it due to recent propagandist fervor? That would be 
psychologically curious since the reasons for such spread 
are not at all obvious. Or are we — more probably — deal- 
ing with an extremely ancient trait that has persisted in 
many places? If so, here is evidence of the incredible 
tenacity with which inconspicuous elements may be re- 

We have seen that scholars differ widely in their 
chronological tastes. Some are content with vague indi- 
cations; others insist on precise time fixation; still 
others profess reliance on only written records, or even 
disdain history on principle. But even the extremists 
smuggle a chronology into their systems for the simple 
reason that the time category is inescapable, hence the 
need for making our chronology as accurate as we can. 
There is a further reason. Conceptual, spatial, temporal, 
and causal aspects of culture are not so many distinct 
realities ; insight into any one of them enhances our com- 
prehension of the rest. Here, if anywhere, the functional- 
ist point of view should be applied. 

The point is important enough to warrant ample 
illustration. Let us consider, then, how our understand- 
ing progresses with our knowledge of space and time 
relations. To Ratzel the plate armor of Bering Sea na- 
tives was a copy in bone or hide of Japanese metal 
laminae, first, because he ignored the wide occurrence 
of plate armor in Asia; secondly, because lacking faith 
in the originality of primitive peoples he barred them 
from a creative part. Laufer corrected the distributional 
statement: plate armor occurs in China, western Asia, 


Turkestan, and ancient Siberia, hence even on Ratzel's 
psychological doctrine Japan need no longer figure as 
the focus. Further, there is written history which Ratzel 
neglects. Chinese annals prove that the maritime Su- 
shen wore bone plate armor in 262 a.d. — from six to 
eight centuries earlier than the hypothetical Japanese 
prototype, which is thus eliminated. Let us further drop 
the psychological dogma, and the relation of the cruder 
armor to that of any higher civilization appears in a 
new light. The primitive peoples may have borrowed 
greater regularity for their laminae, but such imitation 
does not preclude the independent invention of armor 
as such." Note here the functional query that entered by 
a side door : In what technological setting can armor be 
invented? But corresponding issues go with any chrono- 
logical determination. Conspicuous in Pueblo ritual is 
the masquerading in its rain cult, which a naive observer 
accepts as an integral part of native religion. Parsons, 
starting from other Spanish influences on Pueblo life, 
at first traced the masks to the white intruders. Forthwith 
the concept of aboriginal Pueblo ceremonial is modified. 
But another metamorphosis ensues: an archeological 
site, dating back to the fourteenth or thirteenth century, 
harbors an unmistakable drawing of a Pueblo mask, evi- 
dence supported by Sahagun's account of an Aztec rain 
priest's costume. The pre-Columbian antiquity of the 
masks is re-established ; only their efflorescence, not their 
origin, can be derived from the invaders.^^ But if certain 
traits have cohered at any time, they are at least not 
mutually exclusive. Thus, our functional advances with 
our historical insight. 

Again, Beals' ransacking of early chronicles proves 

41 B. Laufer, "Chinese Clay Figures," FMNH-PAS 13:258 sq., 
Chicago, 1914. 

*2E. C. Parsons, "Some Aztec and Pueblo Parallels," AA 35:611 sq., 
1933. E. C. Parsons and R. L. Beals, "The Sacred Qowns of the Pueblo 
and Mayo-Yaqui Indians," A A 36:510, 1934. 


Tamaulipas an enclave of farmers surrounded by Mexi- 
can nomads, but of farmers lacking Aztec traits beyond 
earthenware and agriculture. The archeological evidence 
Beals cites, however, including temple mounds, stone 
structures, and metallurgy, leaves no doubt of southern 
affinities at an earlier period. Similarly, in the Plains 
Strong and Wedel find a succession of cultures, very 
early hunters receding before horticulturists, who are 
superseded by post-Columbian equestrian hunters." 
These temporal data have an infinitely deeper signifi- 
cance than appears on the surface. What are the condi- 
tions that lead to decadence? Wliat factors specifically 
determine relapse into nomadism? And what is the func- 
tional significance of the horse in the Plains? Is it per- 
chance comparable to that of the snowshoe in subarctic 

Every scheme of developments in the New "World 
bears on vital matters of theory that go beyond mere 
dating. According to Perry and Elliot Smith, until 
several centuries ago the Indians north of the Rio Grande 
lived in more or less simian fashion, only acquiring their 
customs and beliefs along with maize from Southern 
Mexico. Here we have the psychological dogma of human 
uninventiveness ; the functional dogma that practically 
no culture is possible ^vdthout farming; the diffusionist 
dogma that traits are dispersed in large blocks from a 
single center ; and by implication a second psychological 
dogma that maize, once injected into the life of a people 
hitherto ape-like, produces a frenzied burst of creative- 
ness by which they forthwith grow the plant in com- 
pletely novel ways and originate usages and beliefs 

^ R. L. Beals, ' ' The Comparative Ethnology of Northern Mexico be- 
fore 1750," Ihero- Americana, 2:136, 143, 149, Berkeley, 1932. W. D. 
Strong, "The Plains Culture Area in the Light of Archaeology," A A 35: 
271-287, 1933. Idem, "An Introduction to Nebraska Archaeology," 
Smithsonian Miscellaneous Collections, 93: No. 10, Washington, 1935. 
W. E. Wedel, "An Introduction to Pawnee Archaeology," BAE-B 112:94- 
102, Washington, 1936. 


never known in Mexico. To single out one phase of the 
scheme, the simple archeological determination of pot- 
tery and farming in southern Arizona about the begin- 
ning of our era explodes the incredible rapidity de- 
manded by the theory. Substituting Kidder's minimum 
dates of 400 b.c. for the crystallization of Maya civiliza- 
tion, and 1000 b.c. for incipient American agriculture, 
we arrive at a quite different and more accurate notion 
of the process of diffusion. 

Again, there are the stone tools of Folsom, New 
Mexico. Even as recent artifacts they would arouse in- 
terest, but Penck's estimate of an antiquity of 9,000 
years establishes primitive hunting as a concomitant of 
extraordinary stonework. If, on the other hand, the age 
were geologically proved several times as great, the 
superiority of the settlers over their Eurasiatic con- 
temporaries would be of interest. 

Even intelligent description is inseparable from a 
time perspective. As Sapir convincingly pointed out, 
the several culture areas into which Americanists divide 
their natives are by no means classificatory equivalents. 
The Plains, e.g., differ far less from the Eastern Wood- 
lands than either region differs from British Columbia. 
And a proper weighting of these units implies a relative 
chronology. There can, indeed, be few more fundamental 
questions than those concerned with the factors making 
respectively for stability and for modification ; and with- 
out a time scale this matter remains inconclusive. Settle- 
ment of the New World implies, geographically, adapta- 
tion to at least six distinct zones. According to Penck, 
even 25,000 years would be inadequate for the succes- 
sive acclimatizations consequent on migrating from 
Alaska to Tierra del Fuego. This may be sheer conjec- 
ture, but any definitive confirmation or qualification 
would teach us how rapidly a culture can be created. 
Verily, the time factor is not a thing adventitiously 


linked with culture, but is always present and defines 
the very essence of our phenomena.** 

"We have repeatedly mentioned correlations. They 
are, in our opinion, the closest approximation ethnology 
is likely to achieve to the ideals of exact science. Some 
scholars, indeed, postulate ''laws," feeling that otherwise 
ethnology would lapse to an inferior status. This is aping 
ideals applicable only to a part of physics and in no way 
incmnbent on other branches of learning. Every science 
formulates its data according to their nature and may 
even differ in its procedure and results from problem to 
problem. Biology uses mathematics sparingly and for 
limited purposes; astronomers predict eclipses but are 
unable to calculate the gravitational pull exerted by all 
bodies on all others. The ethnologist, correspondingly, 
may never discover laws ; yet his scientific respectability 
remains unimpaired so long as he co-ordinates with a 
maximum of attainable efficiency, the particular phenom- 
ena he studies. 

The reality of correlations on the pattern of Tylor's 
adhesions is unshaken. However, we must beware of 
sham regularities. The proper method is to note empiri- 
cal associations that force themselves on our attention 
and then test whether they are random or organic. 
Lesser has admirably set forth and exemplified the logic 
of the procedure: ''Interrelation in an area cannot es- 
tablish causal connection by itself, but if this interrela- 
tion occurs elsewhere independently, it becomes probable 
that a functional relation is present which can be used 
as a working hypothesis." The most cursory inspection 
of our literature shows that ethnologists are forever 
postulating functional relations without vouchsafing any 

•** Edward Sapir, ' ' Time Perspective in Aboriginal American Culture : 
A study in Method," Canada Geological Survey, Memoir 90:45, Ottawa, 
1916. Albrecht Penek, Wann Teamen die Indianer nach Nordamerilca, ICA 
23:25, 28 f., New York, 1930. A. V. Kidder, "Speculations on New World 
Prehistory," ALK, 143-151. 


reason for their conjectures. Progress will consist 
largely in substituting demonstrated or demonstrable 
correlations for idle allegations.*^ 

Given the complexity of social phenomena, we cer- 
tainly cannot as a rule expect a simple causal relation: 
any established correlation must be treated as over- 
lapping others; or, to put it differently, our correlates 
do not fully determine one another, the presence of some 
simply renders the others more probable than would 
otherwise be the case. That is the simple meaning of 
the much misunderstood relationship between matrilineal 
descent and certain kinship classifications ; the latter are 
to be expected rather with than without matrilineal clans ; 
no one asserts that the rule of descent automatically 
evokes the classification. 

Such correlations justify a greatly tempered paral- 
lelism or neo-evolutionism. If four or five features are 
organically linked, it is probable and sometimes certain 
that they have not always synchronized. That is, one of 
them may be not merely an equivalent concomitant but 
a significant antecedent. If so, its presence anywhere 
favors the appearance of the same consequents. Thus, 
Steward finds a common ecological cause underlying 
patrilineal bands in the four quarters of the globe. 
Definite decorative patterns recur with definite basketry 
techniques, and certainly the angularity of life forms in 
woven fabrics is not the cause but the effect of weaving. 
Among the Pueblos, woman's importance in ritual 
varies directly with the absoluteness of feminine house 
ownership and the strength of the clan. Parsons' analy- 
sis indicates not merely a functional tie but an irreversi- 
ble sequence: women as house owners; matrilocal resi- 
dence; matrilineal lineages; matrilineal clans. The 
causal role of matrilocal residence agrees with Tylor's 

*^ Alexander Lesser, ' ' Kinship Origins in the Light of Our Distribu- 
tions," AA 31:716, 1929. 


earlier ideas and might well be checked in all compara- 
ble regions.** 

Naturally, correlations can be established only if 
the several regions concerned are historically independ- 
ent of one another. In other words, their determination 
is inconsistent with an extreme diffusionism. But the 
messianic ideas precipitated by European contacts 
among North American, South American, and African 
natives clearly prove that similar notions can be evoked 
by similar antecedents where there is demonstrably no 
dijffusion. In principle, then, the possibility of independ- 
ent repetition is vindicated. 

What, then, is the prospect of ethnology? Its hope 
lies in maintaining the universalist and the objective 
approach of Tylor and Boas: whatever preferences the 
individual worker may gratify, our science as a whole 
can neglect no aspect of social life as intrinsically in- 
ferior to the rest. Specifically, material objects must be 
studied as embodiments of their makers' craftsmanship, 
aesthetic taste or spiritual aspirations. Subjective at- 
titudes and personality must also be investigated as so- 
cial symbols; what is obscure must be made clear, and 
results can be reckoned scientifically valid only if verifi- 
able by subsequent investigators. This topical breadth 
will be matched by the massiveness of the regional ap- 
proach, which must include all past and present cultures 
from the highest to the lowest. Breasted 's Egyptian re- 
searches and Laufer's ransacking of the sinologue's 
treasure-trove are an earnest of what may be expected 
from a thorough examination of written sources by 
scholars steeped in the ethnologist's point of view. This 
regional and topical universalism implies an ever- 
widening and deepening culture history which will pro- 
vide matter for an increasing number of significant cor- 

*«J. H. Steward, op. cit. E. C. Parsons, "The House-Clan Complex 
of the Pueblos," ALK, 229-231, 1936. 


relations in Tylor's fashion, an ever-active epistemo- 
logical critique of concepts in the spirit of Boas. 

The discipline we have described is bound to rank 
co-ordinate with other sciences insofar as it continues 
to investigate objectively and intensively that segment 
of reality which falls to its province. Spatially it will 
arrange its data after the fashion of geography; chron- 
ologically, it will use — according to particular exi- 
gencies — the logic and techniques of geology, historical 
astronomy, political history ; causally it will establish an 
indefinite number of valid correlations, thereby attaining 
the degree of generalization compatible with its own 
section of the universe. 


Agriculture, 165, 259 (see also Eco- 
nomics, comparative) 

Aitken, Barbara, 266, 276, 278 

Andree, Eichard, 73 f., 80 

Animism, 83 f., 110, 163, 205 (see 
also Eeligion, comparative) 

Ankermann, B., 157 

Art, 93-95, 134, 139, 272 

Associations, 58, 99 f. 

Avunculate, 26, 234, 241 

Bachofen, J. J., 40-43, 51, 53, 61, 
114, 119, 142, 189, 230, 239 

Balfour, Henry, 28, 35, 69 

Bastian, Adolf, 29, 30-38, 58, 72, 73, 
105, 122, 129, 157, 251 

Beals, Ealph L., 285, 286 

Benedict, Euth, 134, 266, 273, 275- 

Birket-Smith, K., 192, 234, 260 

Boas, Franz, vii, 17, 25 f., 37, 73, 
88, 105, 109, 119, 128-155, 157, 
158, 159, 172, 178, 184, 188, 189, 
193, 202, 203, 207, 211, 212, 217, 
219, 223, 230, 231, 232, 235, 239, 
241, 244, 247, 251, 256, 257, 263, 
266, 271, 275, 276, 290, 291 

Bogoras, Waldemar, 261, 270 

Boucher de Perthes, J,, 7 ff., 21, 
196, 254 

Bride-capture, 44 sq., 48 f,, 247 f. 

Brinton, D. G., 91, 148 

Bronze Age, 21, 87, 280 

Buckland, A. W., 90, 92, 114 

Catchwords, 23, 160, 165 f., 184, 
187, 202, 247, 253, 258, 281 


Ceremonial (see Eeligion, compara- 

Cooper, John M., viii, 183, 184, 194, 

Correlations, 79 sq., 225 f., 234, 
240, 248 f., 288-290 (see also 

Couvade, 42, 247 

Criteria of antiquity, 147, 154 

Criteria of diffusion, 76, 126, 148 f., 
158, 160, 161, 166 

Cultivated plants, 92 (see also Agri- 
culture; Economics, comparative) 

Culture, defined, 3, 10, 12, 199, 235 

Culture areas, 36, 37, 125, 287 

Darwin, Charles; Darwinism 19, 24, 

31 f., 67, 125 
Degeneration (cultural) 20, 161 sq., 

168 f., 173, 286 
Diffusion, diffusionists, 27-29, 52, 

59 f., 72 sq., 78, 90 f., 108, 118, 

122 sq., 150 f., 156-195, 244 f., 

Dixon, Eoland B., 274 
Domesticated animals, 59, 92, 112 

sq. (see also Economics, compara- 
tive; Hahn, Eduard) 
Dreams and visions, 78, 207, 264, 

272, 273, 279 
Du Bois, Cora, viii, 240 
Durkheim, fimile, 197-212, 215, 

216 f., 218, 222, 226, 228, 230, 


Economic determinism, 189 f. 
Economics, comparative, 91-93, 112- 
119, 214 ff., 233, 242 f., 247, 252 



Elementargcdanken (see Psychic 
unity; Parallelism) 

Endogamy, 45 

Environmentalism (see Geography) 

Evolution, evolutionism 12 f., 19-29, 
41 sq., 56 sq., 80, 178, 189 f., 201, 
206, 218, 226 f., 246 f. (see also 
Bastian; Morgan; Tylor) 

Exogamy 45 f., 62, 72, 79, 96, 171, 

Field work, 5 f., 88 sq., 131-136, 

170, 196, 221 f., 231 f. 
Forde, C. Daryll, 250, 261 f. 
Frazer, J. G., 69, 101-104, 212 
Frobenius, Leo, 159 
Functionalism, 8, 41, 42, 53, 111, 

118, 142-144, 188, 189, 190 f., 

202, 223, 227, 230-249, 251, 258, 

260, 284, 286 
Fustel de Coulanges, N. D., 39 f., 

41, 43, 53, 142, 198, 230, 239 

Galton, Francis, 68, 92, 105, 134, 

Games, 76, 91, 172, 257 

Gayton, A. H., 240 

Geography, geographical determi- 
nants, 15, 17, 36, 58 f., 120 f., 
144, 167, 179, 236, 238, 250, 256- 
262, 283 

Gifford, E. W., 172, 281 

Goldenweiser, A. A., 102, 141, 188, 
198 f., 208, 210, 220, 282 

Graebner, Fr., 35, 76, 157, 159 f., 
177 sq., 237, 243, 247 

Grosse, Ernst, 95, 189, 192 

Gusinde, Martin, 192 f, 

Gynecoeracy (see Matriarchate) 

Haddon, A. C. vii, 37, 69, 89, 94, 

Haeberlin, Hermann 142, 276 
Hahn, Eduard 92, 112-119, 127, 173, 

247, 252 
Hallowell, Irving, 270 
Hatt, Gudmund, 116, 178, 184, 260 
Hatt, Mrs. Gudmund, 135 
High-gods 82 f., 211 (see also Eeli- 

gion, comparative) 

Historical schools, 156-195 

History, 144 sq., 175 f., 195, 198, 
201, 222 f., 226 f., 227 f., 230, 
234 f., 238, 239, 244, 251 f., 254 f., 
257, 283 f., 286 (see also Diffu- 
sion; Evolution) 

Hocart, A. M., 199, 237 

Holmes, Wm. H., 94, 139 

Hough, Walter, 71, 87 

Independent origin (see Parallel- 

Individual variation, 105, 134, 189, 
198, 206 f., 212 f., 220, 222, 227, 
231 f., 266 ff. 

Kidder, A. V., 165, 255, 287 

Kirchhoff, Paul, 240 

Klemm, Gustav, 11-16, 18, 24, 70, 

71, 189, 250 
Koppers, Wm., 27, 73, 113, 116, 178, 

Kroeber, A. L., viii, 64 f., 100, 129, 

140, 153, 171, 175, 192, 194, 200, 

232, 240, 278 
KuUurJcreise, 157, 177-195, 243 

Lang, Andrew, 69, 82 f. 

Laufer, Berthold, 5, 12, 59, 119, 

157, 192, 280, 284 f., 290 
Law, comparative, 39-67 (see also 

Social Organization) 
Laws (of cultural growth), 23, 

26 ff., 36, 51, 77, 80 f., 145, 202 f., 

218, 222-225, 227, 251, 264, 288 
Lesser, Alexander, 288 
Letourneau, Ch., 20, 84 
L6vy-Bruhl, Lucien, 136, 197, 200, 

216-221, 247, 251, 254 
Linguistics, 69 f., 89 f., 95, 130, 

132, 172 
Linton, Ralph, 266 
Literature, primitive, 95, 255 
Lubbock, Sir John, 20, 24, 25, 51, 

61, 66, 69, 84, 106, 186 

Magic, 102 f,, 204, 209 f., 212 ff., 
(see also Religion, comparative) 

Maine, H. J. S., 49-53, 55, 58, 95, 
96, 106 

Malinowski, Br., 37, 216, 223, 230- 
243, 244, 247, 251, 252, 275, 276 



Marett, E. E., 69, 102, 109-111, 212 
Marginal peoples, 126, 150, 179 
Matriarchate, 40 sq. 
Matrilineal and patrilineal descent, 

40 sq., 52, 59, 81 f., 145, 150, 181, 

190, 203 
Matrilocal and patrilocal residence, 

46 f., 79, 200 
Mauss, Marcel, 197, 212-216, 221, 

McLennan, J. F., 43-49, 50, 51, 52, 

53, 55, 56, 61, 234, 281 
Mead, Margaret, 134, 275 
Meiners, C, 5, 10-11 
Meta-ethnography, 274-279 
Metranx, A., 6, 268 
Michelson, Tr. 133, 135 
Montandon, George, 185-187 
Morgan, Lewis H., vii, 40, 45, 48, 

49, 51, 52, 54-67, 70, 79, 83, 88, 

95, 96, 119, 150, 157, 171, 190, 

226, 281 
Morgan, Wm., 267, 272 f. 
Motor habits, 262 f, 
Musicology, 95, 195, 255 f. 
Mythology, 140 f., 148 f., 194 f., 

257, 264 

Nelson, N. C, 255 
Neolithic, 21 f., 23, 70 
Nimuendajti, Curt, 6, 268 
Nordenskiold, Erland, 178, 236, 
252, 254, 257, 258, 259 

O'Neale, Leila, 254 

Paleolithic, 21 f,, 23, 70 
Parallelism, 28 f., 44, 48, 51, 58 

sq., 72 sq., 79 sq., 91, 95, 102, 

146, 148, 158, 166, 168, 184, 201, 

203, 246, 264 (see also Laws; 

Psychic unity) 
Parsons, E. C, 49, 101, 106, 134, 

135, 183, 228, 248, 252, 275, 278, 

282, 284, 289 f. 
"Participation," 218 
Patterns, cultural, 109, 117, 211, 

Perry, W. J., 159, 163, 164, 169, 170, 

176, 184, 286 

Pitt-Eivers, A. Lane-Fox, 19 f., 
20 f., 28, 60, 69, 74, 91, 93, 122 

Polyandry, 46, 48, 62 

Prehistory, 19-20, 75, 86 f., 90, 91, 
196, 255 (see also Boucher de 

Progress, 12 f., 18, 202 

Psychic unity, 29, 35 f., 72, 76, 78 f., 
123, 257 (see also Parallelism) 

Psychoanalysis, 172, 234 

Psychology, 36, 43, 52 f., 96, 102, 
104-111, 112 sq., 120, 131, 136 sq., 
188 f., 199 f., 202, 206, 217 sq., 
222, 254, 262-274 (see also 
Individual variation ; Psychoa- 
nalysis; Eace psychology) 

Eace psychology, 13-15, 16-18, 24, 
84, 104 f., 121, 136 f., 170, 202, 
251, 265 

EadcUffe-Brown, A. E., 27, 37, 45, 

66, 188, 210, 221-229, 230 f., 234, 
241, 244, 246, 252, 254, 277, 281 

Eadin, Paul, 135, 143, 252, 270 
Eationalism, 105 sq., 110 f., 112 sq., 

137 f., 207, 216, 217 sq., 233, 251 
Eationalization, 138 sq., 202, 211 
Eatzel, Ft., 72, 74, 99, 113, 119-127, 

144, 157, 158, 159, 178, 184, 193, 

195, 261, 284 f. 
Eelationship terms, 25, 47, 54, 60- 

67, 79, 89, 170, 173, 225 f., 245 
Eeligion, comparative, 70, 88, 102 sq., 

109 ff., 191 f., 198 f., 204 sq., 

Eetrogression (see Degeneration) 
Eites, transitional, 100 f. 
Eivers, W. H. E., 36, 62, 66, 69 £., 

89, 104, 169-176, 178, 184, 188, 

199, 221, 234 
Eivet, Paul, 197, 254 

Sapir, Edward, 160, 287 

Sauer, Carl 0., 163, 259 f. 

Schmidt, Wm., 27, 73, 83, 116, 159, 
177 sq., 188-195, 198, 207, 212, 
220, 227, 232, 244, 246, 252 

Schurtz, Heinrich, 58, 99-100 

Secondary interpretation, 138 sq., 
218 (see also Eationalization) 

Selection, social, 266 f. 



Seligraan, C. G., 89, 251, 266, 269 
Smith, Q. Elliot, 72 ff., 84, 90, 146 f., 

159, 160-169, 176, 178, 179, 

184, 187, 193, 286 
Social orgauization, 45, 60, 95-101, 

171, 202 f., 232, 240, 252, 281 

(see also Law, comparative; 

Maine, H. J, S. ; McLennan, J. F. ; 

Morgan, L. H.) 
Sociology, French, 196-229 
Speck, F. G., 184, 192, 260, 274 
Spier, Leslie, 64, 234, 251, 252, 263, 

Steward, J. H., 281 f., 289 
Stolpe, Hj., 93 f. 
Stone Age, 7 sq., 13, 21 f., 23, 28, 

72, 75, 86 f., 108 
Strong, Wm, D., 281, 286 
Subjectivism, 24 f., 51, 58, 82 f., 

83, 98, 107, 137, 186, 189, 201 
Survivals, 25, 42, 44 sq., 81 f., Ill, 

202, 237 
Swanton, J. E., 82, 145 f., 232, 252 

Tarda, G. de, 106-109, 117, 137, 217, 

245, 251 
Technology, 16, 21 sq., 75, 87, 253 f. 
Thurnwald, Hilde, 134 f., 207, 242 
Thurnwald, Richard, 27, 37, 48, 188, 

216, 220 f., 223, 227, 242-249, 

251, 252, 266, 267 

Totemism, 47 f., 89, 127, 141, 152, 
180 sq., 205 sq. 

Tylor, Edward B., 12, 15 f., 20, 34, 
44, 48, 61, 66, 68-85, 87, 88, 91, 
95, 97 f., 99, 100, 102, 105, 106, 
122, 129, 136, 137, 141, 145, 154, 
157, 158, 159, 171, 172, 173, 174, 
197, 199, 200, 207, 217, 234, 241, 
262, 288, 289, 290, 291 

Uninventiveness, human, 74, 120, 
122, 158, 161, 167 f., 176, 284, 

Van Gennep, Arnold, 100 f. 
Virchow, Rudolf, 7, 13, 30, 31, 34, 

85, 125, 129, 154 
Von den Steinen, Karl, 6, 23, 31, 

33, 37, 89, 94, 138 
Von Hornbostel, Erich, 95, 192, 195 
Von Luschan, Felix, 160 

Waitz, Th., 16-18, 84 f., 121, 137, 

Warner, Wm, Lloyd, 210 
Webster, Button, 100 
Weltfish, Gene, 271 
Westermarck, Edward, 95-99, 232 
Wissler, Clark, 71, 140, 154, 178, 

194, 199 f., 263, 271 


Los Angeles ' 

This book is DUE on the last date stamped below. 



B Am 19 Ml 

MfiD ? ■> 195; 


' \f mi n * ,^ 

QLJA,li,l 3 1991 
1 7 199/ 


, (^CNOV0 719<M 



L 005 829 454 7