1
Ústav
religionistiky
Jonathan Z. Smith: Adde Parvum Parvo Magnus Acervus Erit, in:
Jonathan Z. Smith, Map is not Territory, The University of Chicago
Press: Chicago 1993, s. 240 - 264
FRVŠ
1890/2005
Téma: Religionistická klasifikace: monotetický a polytetický
Theprocess of comparison is a fundamental characteristic of human intelligence.
Whether revealed in the logical grouping of classes, in poetic similes, in mimesis, or other like
activities-comparison, the bringing together of two or more objects for the purpose of noting
either similarity or dissimilarity, is the omnipresent substructure of human thought. Without
it, we could not speak, perceive, learn, or reason. Instances of comparison are presumably as
old as mankind itself and may be found in our earliest literary documents. It is unfortunate
that relatively little work has been done on the history of the use of comparison in scholarship
and that deeper questions of method and the underlying philosophical implications of
comparison have been ignored by many disciplines including History of Religions. That
comparison has, at times, led us astray there can be no doubt; that comparison remains the
method of scholarship is likewise beyond question. S Samuel Butler recorded in his notebook:
"Though analogy is often misleading, it is the least misleading thing we have."
Robert Redfield, in his important essay, "Primitive World View," suggested that a
world view of any people consisted essentially of two pairs of binary oppositions: MAN/NOTMAN
and WE/THEY. Both of these pairs imply comparison; it is the latter which underlies
most forms of mltllral comparisol1. When one encounters another, a place must be found for
the other within or without one's cosmos. Four specifications of the WE/THEY duality have
been employed: (1) They are LIKE-US, (2) They are NOT-LIKE-US, (3) They are TOOMUCH-LIKE-US
(Robert Frost's "Good fences make good neighbors"), or (4) We are NOTLIKE-THEY
(expressions and polemics concerning "uniqueness"). This primary sense of
comparison depends upon the living contact between two peoples, whether this contact be
through travel, invasion, or trade. It is the attempt to "place" one another.
A second major class of comparison is that of historical comparison, whether of the
popular example type or in more complex hermeneutic discussions by scientific and
philosophical historians. Here the question is not communication between two living representatives
of differing cultures and their mutual positioning, but between the present and
the past. The question has become: are the individuals and cultures of the past LIKE-US or
NOT-LIKE-US? If they were LIKE-US why should we want to know about them? If they
were NOT-LIKE-US how can they be comprehended? These (questions are to the fore in
Dilthey's description of the hermeneutic task: "Interpretation would be impossible if [past]
expressions of life were completely strange. It would be unnecessary if nothing strange were
in them. It lies, therefore, between these two extremes." History in general and historical
study of religion in particular, is caught between these tensions. Comparison, the existence of
similarity, is the inescapable presupposition of historical research. On the one hand the
2
historian recognizes the truth of Toynbee's remark that "this word 'unique' is a negative
term signifying what is mentally inapprehensible. The absolutely unique is, by definition,
indescribable." On the other hand, he recognizes that he is dealing with a subject so value
laden (whether it be an individual's biography or religion) as to elicit from his subject a selfconsciousness
of importance and uniqueness. In a shrewd passage, which I recall each time I
hear the historian of religions insist on the sui generis character of religions, William James
put the matter with precision: "The first thing the intellect docs with an object is to class it
along with something else. But any object that is infinitely important to us and awakens our
devotion feels to us also as if it must be sui generis and unique. Probably a crab would be
filled with a sense of personal outrage if it could hear us class it without ado or apology as a
crustacean, and thus dispose of it. 'I am no such thing,' it would say: 'I am myself, myself
alone.'" This tension between the subject's sense of the unique and the methodological
requirement of the analogous generates both the excitement and the problematics of
historical research.
The third great class of comparison is, in part, a combination of the first two. It is, at
one and the same time, both observation and theory. It is the noting of similarity (LIKE-US)
and the accounting for this similarity in terms of a process of assimilation, diffusion, ` (i.e.,
they are LIKE-US because they ARE-US, they have COME-FROM-US, we are RELATEDTO-THEIR).
It is to introduce a temporal, historical framework with high value placed on
priority in time. While not always displaying the hostility of early Christian theories of
"demonic plagiarism," there is a clear sense of higher value and authenticity attached to the
source and a sense of second-handedness, of imitation, and even of fraud attached to the
alleged borrower. Furthermore, there is frequently a strong sense of in- and out-groups, of
peoples from whom it is all right to have borrowed and peoples from whom one ought not.
The pedigree, being "of good stock," is all-important. Such diffusionist theories have
remained the prime mode of accounting for similarity since the writings of Herodotus.
The fourth great class of comparison is comparison as a hermeneutic device. Here the
meaning and function of a particular motif, symbol, or custom in one culture may be used as
a key to interpret a similar motif, symbol, or custom in another culture by moving from what
is known to what is unknown. The rationale given for this procedure may take three forms:
an atemporal argument that both elements are manifestations of a common archetype or
reflections of the psychic unity of mankind; a broadly temporal argument for the validity of
comparing similar stages of human development (e.g., the comparisons between the
Australian aborigines and certain Paleolithic religious structures); or an argument that
presupposes some temporal sequence (e.g., the hermeneutic use of the notion of survivals).
It will be my contention in this paper that these four great classes of comparison, when
applied to religious and cultural data, have involved four modes or styles of comparison; the
ethnographic, the encyclopedic, the morphological, and the evolutionary. The remainder of
this paper will be concerned with a brief characterization, history, and evaluation of these in
the belief that most of the procedures and specific instances of comparison may best be
understood as an application of one or more of these styles or modes.
THE ETHNOGRAPHIC
3
The earliest extant example of comparative religions occurs in the well-known
fragment of Xenophanes (ca. 570-475 B.C.); "The Ethiopians say that their gods are snub
nosed and black; the Thracians say that their gods have light blue eyes and red hair."
However, the text is too brief to permit one to determine the sorts of traditions which lie
behind this observation or to permit one to elucidate the method of comparison being
employed. The development of procedures of comparison between religions is first displayed
in the writings of Herodotus (484-424 B.C.) who, in addition to his usual title, Pater historiae,
may be accurately termed the Father of Anthropology and the Father of History of Religions.
Herodotus sums up the ethnography of the Ionian School of historians, and his work and
procedures remain the model for the ethnographic style Central to this enterprise was the use
of comparison. As T. Hodgen has noted: "the most significant bequest made by Herodotus to
subsequent thought was his use of comparison, and his recognition, even in the fifth century
B.C., of some of the problems which have emerged whenever the comparison of cultures has
revealed either similarities or differenccs." A hint as to the basis for cultural comparative
religion." comparison may be found in Herodotus' report of the reply of the Athenians to the
Spartans when they refused to break with the common Greek cause. Here we are given the
four criteria Herodotus had for determining a cultural unity: common descent, common
language, common religious practices, common customs and world view (VIII.144). It will be
the presence of one or more of these features in common that will allow Herodotus to make
cross-cultural comparisons.
In Ionian ethnography as represented by Herodotus we find histo6a in the most literal
sense of the word, "a narrative of what one has learned by inquiry," and it would appear that
there was a fixed form of inquiry, a questionnaire, employed by Herodotus in his researches.
The topics may be discerned in the stereotyped form Ionian ethnography followed for the
presentation of its data.
I. The 1,-and: (a) borders, measurement, shape; (b) nature and character of the land; (c)
rivers; (d) climate; (e) animals, trees etc.
II. People-History and Customs: (a) number of population; (b) their antiquity; (c) their
habits (e.g., dress, food); (d) customs (nomoi). The subdivisions of this category may be best
illustrated by the Scythian material in Book IV (1) gods (theoi, IV.59); (2) sacrifices ( I
V.60); (3) war customs ( IV.64); (4) oracles (V.67); (5) oaths (, IV.70); (6) burial practices
and tombs ( 1\'.71); (7) foreign customs (1V.76).
III. Wonders and Marlels.
What is being collected under these headings for some fifty cultures is basically a set of
traveler's impressions. Something other has been encountered, and it is surprising either in its
similarity or dissimilarity to what is familiar "back home." In such a context, features,
political structures, customs, and religious practices-these are at one and the same time the
most ethnically distinct and humanly similar features that would attract the eye of the
traveler. Only rarely is the truly unique met and recorded (e.g., "The Magi are indeed a
unique race different entirely from the Egyptian priests and indeed from all other men"
[1.140]); more often the differences are relativized by means of comparison.
4
This reliance on the traveler's eye relates Herodotus and the Ionian ethnographic
tradition to ethnography as it is practiced today. Not only does the outline of topoi look very
much like a table of contents for a contemporary field worker's report, but other
characteristics are shared as well. Without pressing the point too far (and thus continuing the
fallacious tradition in popular works on classics for finding Greek antecedents for
contemporary theories such as evolution and atomic structures) there is a marked functionalism
to these ancient accounts. As each culture encountered is presented largely in its own
terms (the interpretatio graeca notwithstanding), the individual items are given in the
complexity of their interrelationships and not excised from their context. They are given as
they appear to the observer's vision without apparent analytical mediation. In the case of
religious material, it is striking how rarely Herodotus introduces a description of a people's
god or their mythology for its own sake. Of the thirteen peoples whose religion is discussed, in
only one case, that of the Scythians, do the gods occupy first place in Herodotus's account. In
the other twelve, priority is given to cultic institutions, and their gods or theology are only
introduced insofar as it is required to comprehend the cult which has been seen by the
observer. Thus in the Egyptian material, the fullest presentation of religion in Herodotus and
one which clearly derives from his own travel experience, the material is not arranged as in
present-day handbooks, by deities, but by animals sacred to the gods or utilized in their cult.
The individual gods are only introduced as excurses under the heading of the appropriate
animals. The same pattern may be noted in Herodotus's use of myth. He only narrates seven
myths, and these he tells with extreme economy. Two of them (the births of Herakles [11.43-
45] and of Pan and Dionysos [II.142146]) are told for their importance in establishing a
chronological detail; five are cult aetiologies 111.42, 63, 146; Vl1.26; VIlI.55). Consistently,
for Herodotus, it is the descriptions of the cultic institutions as they are seen which calls forth
a discussion of the gods and not vice versa.
Herodotus's cultural comparisons are likewise dependent upon what happens to have
been seen and what catches his fancy:
Oaths are taken by these people the same way as by the Greeks except that they make a slight
flesh wound in their arms.(1.74)
The Lydians have very nearly the same customs as the Greeks except that they do not raise their
girls in the same way. (1.94)
[The Babylonian's] shoes arc peculiar; not unlike those worn by the Boetians. (1.195)
In their dress and mode of living, the Massagetae resemble the Scythians. (1.215)
The Thracians, Scyths, Persians, Lydians and almost all other barbarians hold the citizens who
practice trades in no less repute than the rest .... These ideas prevail also in Greece, particularly
among the Lacedaemonians. Corinth is the place where handicrafts arc least despised. (II.167)
The difficulties with these comparisons are the same difficulties encountered in specific
cultural ethnography today (as Levi-Strauss has brilliantly argued with respect to Boas and
Malinowski). Such comparisons are idiosyncratic, depending upon intuition, a chance
association, or the knowledge one happens to have at the moment of another culture. There is
nothing systematic to such comparisons, they lack any basis, and so, in the end, they strike us
as uninteresting, petty, and unrevealing. In Lévi-Strauss's critique, such comparison "loses
5
the means of distinguishing between the general truths to which it aspires and the trivialities
with which it must be satisfied." Finally, the ethnographer is driven to generalizations which
are platitudes (such as Herodotus's famous dictum, quoting Pindar, "custom is king over all"
[III.38]) or to the tautologies of functionalism.
Herodotus has bequeathed to us the rich tradition of culture specific ethnography; he
has bequeathed to us also the problems of such a procedure. Whether in Greco-Roman texts
or in the more recent arguments concerning Kulturkreislehre, historicism, fieldwork, limited
comparison, or functionalism-the ethnographer cannot escape generalization and comparison
and thus cannot escape the problem of the basis for such generalization and comparison, its
criteria, its .extent, and meaning.
THE ENCYCLOPEDIC
The ethnographic tradition begun by the Ionian school continues with varying fortune
to the present day. In Greco-Roman writers such as Ctesias on the Persians and Indians,
Megasthenes on the Indians, Poseidonius on the Celts, Tacitus on the Germans, or the
nationalistic ethnologies compiled by figures such as Alexander Polyhistor, the device of
presenting a set group of to poi concerning the culture of a specific people remained a major
scholarly option. Dependent as it was upon travel and first-hand acquaintance, it flourished
in periods of travel and trade and declined in periods of isolation. In the West, with only an
occasional pilgrim's iti1lerarilllll as an exception, there is little new ethnography until the
Ystoria Alo11galort/m of John of Pian del Carpine in the thirteenth century.
From this point on there has been, with few gaps, a relatively unbroken tradition of
ethnographies. The encyclopaedic tradition was not limited by external circumstances.
Rather than the presentation of topoi for a single culture which has been encountered by the
author as was characteristic for the ethnographic mode, the encyclopaedic style offered a
topical arrangement of cross-cultural material (arranged either by subject matter or
alphabetically) culled from reading.30 It is the style of the "armchair" anthropologist rather
than the field worker.
The tradition began with Hellanicus in the fifth century. He produced a variety of
ethnographies reminding us of the works of the Ionian school (Aegyptiaca, Persica, Scythica)
but distinguished from them by being entirely derived from reading other men's works. He
also produced a set of general, topical collections with titles such as On Peoples, On the
Foundings of Cities and Peoples, and Foreign Customs. If one stream of the encyclopaedic
tradition descends from these sort of collections, the other stream focuses on the contents of
these collections. The third group of data given in the Ionian scheme was that of wonders.
While frequently these take the form of points of interest which the traveler must see, at times
they express what is different about a culture, how they are NOT-LIKE-US. Thus Herodotus
provides, in 11.35-37, an antipodal account of Egypt: "Not only is the climate different from
6
that of the rest of the world, and the rivers unlike any other rivers, but the people also, in
most of their manners and customs, exactly reverse the common practice of mankind." A
similar description is given of the Scyths (IV.28). Occasionally other sorts of marvels are
narrated (e.g., the Phoenix [II.73]; flying snakes [II. 75], or the ants who dig up gold r III.102,
105]) but Herodotus is careful to insist that he has never seen these things but was "told"
about them.
With the development of the paradoxographical tradition associated in its origin with
the poet Callimachus (fl. 270 B.C.), these materials. began to be collected in works
reminiscent of more recent books such as Charles Fort's Book of the Damned or Robert
Ripley's series, Believe It or Not. The encyclopaedic tradition became inextricably 'wedded to
the quest for the exotic, the marvelous, the anomalous, the strange. Callimachus's worksCostumes
of Foreign Peoples, Curiosities Collected All Over the World according to Place as
well as those of his descendants - the pseudo-Aristotelian (?), De Mirabilibus, Antigonus of
Carystas, Collection of Wonderful Stories, Phlegon of Tralles, On Marvels and Long Lived
Peoples - stand in direct relationship to such later works as Frazer's The Golden Bough. The
encyclopaedic tradition, fused with paradoxographical and teratological traditions, became
cabinets de curiosités, contextless lists of strange things clone by strange peoples in strange
lands. Characteristic of this entire tradition is relatively little interest in explicit comparison.
The material is NOT-LIKE-US, it is inhuman, monstrous, primitive, exotic. It possesses, by its
very nature, no comparability with civilized or cultured men. Nor, with its interest in the
unique, does the encyclopaedia compare the data in its lists. They simply cohabit within some
category, inviting comparison by their very coexistence; but providing no explicit clues as to
how this comparison may be undertaken.
The sheer density of the "factual" material presented-whether it be Pliny's claim to
have read two thousand books by four hundred and seventy three authors and to have
excerpted some twenty thousand items for his Historia Naturalis XXXVII or the even grander
enterprise of Frazer-the almost epic quality of the lists tends to dull one's critical faculties.
The data are accepted as being overwhelmingly and massively there. Interpretation and
comparison are simply not asked for. Similarly it is the aesthetic of the catalogue or list to
present largely a surface appearance. Depth, the problematics, are eliminated (much as in a
Robbe-Grillet novel) in favor of the "hard" enumeration of things. When one does begin to
ask depth questions, when one inquires into the context of the material, the principles of
internal order governing the lists, or asks for some evaluation of the significance of the
material (other than that it is exotic and hence intrinsically interesting), the surface cracks
apart. Franz Steiner, in his important book Taboo, has caught a classic instance in Frazer:
"Burial grounds were taboo; and in New Zealand a canoe which had carried a corpse was
never afterwards used, but was drawn on shore and painted red. Red was the taboo colour in
New Zealand; in Hawaii, Tahiti, Tonga and Samoa it was white. In the Marquesas a man who
had slain an enemy was taboo for ten days: he might have no intercourse with his wife and
might not meddle with fire; he had to get some one to cook for him. A woman engaged in the
preparation of cocoa-nut oil was taboo for five days or more." Extending Steiner's analysis, I
would suggest that five different verbal associations make up this extract: (1) General thesisburial
grounds are taboo, that is, the place where the corpse was laid. This leads to the
7
example of the New Zealand canoe in which a corpse was laid because it was taboo. (2) This
canoe was painted red; this leads to the report, irrelevant to the subject of burial taboos that
in New Zealand red is a taboo colour. (3) On the order of the familiar test question, "If I say
black, what do you think of?" Frazer continues a chain of association. If red is the taboo color
on one Pacific island, white is taboo for four other Pacific islands. (4) Having noted color
taboos on Pacific islands, Frazer recalls the taboos on Marquesa, another Pacific island,
regarding a man who has slain his enemy. This report, while having nothing to do with colors
or burial grounds, has at least the virtue of having something to do with death. (5) One of
these Marquesan taboos involves cooking, which leads Frazer to speak of another cooking
taboo in some unnamed location (presumably another Pacific island) having to do with the
preparation of cooking oil. And so, on and on. After carefully working through a passage like
this, which is by no means unique to Frazer or any other encyclopaedist, one can only echo
Steiner's conclusion: "This is the rhetoric of association. The more clearly it stands out, the
less trustworthy the scholarship of the author appears."
The unfortunate link of the use of contextless lists held together by mere surface
associations rather than careful, specific, and meaningful comparisons with the interest in
exotic content has plagued the encyclopaedic tradition until the present time. Malinowski's
charge to his anthropological colleagues likewise remains in force: "We can only plead for the
speedy and complete disappearance from the records of fieldwork of the piecemeal items of
information, of customs, beliefs and rules of conduct floating in the air. ... With this
[disappearance] the theoretical arguments of Anthropology will be able to drop the lengthy
litanies of threaded statement which make us anthropologists feel silly and the savage look
ridiculous."
THE MORPHOLOGICAL
In his seminal essay on the sociology of knowledge, Karl Mannheim introduced a valuable
distinction which bears directly on the question of the logic of comparison as well as having
profound implications for discussions of reductionism. Mannheim distinguished between
"right-wing" and "left-wing" methodologies: "Early nineteenth century German
conservatism ... and contemporary conservatism too, for that matter, tend to use
morphological categories which do not break up the concrete totality of the data of
experience, but seek rather to preserve it in all its uniqueness. As opposed to the morphological
approach, the analytical approach characteristic of the parties of the left, broke down
every concrete totality in order to arrive at smaller, more general, units which might then be
recombined through the category of causality or functional integration." Extending
Mannheim's observation, the following methodological clusters emerge as possibilities:
"Right Wing": Morphological/Synthetic / Structural / Synchronic / Phenomenological.
"Left Wing": Evolutionary / Analytic / Functional/Diachronic / Historical.
8
Each cluster has its legitimate types of tasks, questions, and procedures. Reductionism is an
inherent danger for either group (although the conservative approach, by its very nature,
appears less likely to be reductionistic than the radical; an observation not sufficiently
recognized in many discussions of reductionism). A more serious problem arises when, to
borrow a pun from a recent anthropological symposium, there is a muddle instead of a model,
when elements from the two clusters are carelessly combined. It is this muddle, as I shall
discuss below, which produced the dismay with late nineteenth and early twentieth-century
use of The Comparative Method.
The rise of the "science" of Comparative Religions cannot be separated from
nineteenth-century scientific thought in general. When, in 1905, Louis H. Jordan defended
the legitimacy of Comparative Religion to claim for itself the title of a "science," he pointed to
analogies in the field of Comparative Anatomy and Comparative Philology. It is often
forgotten how decisive a role the successes of Comparative Anatomy played in the
development of the various comparative disciplines. Beginning with Buffon and continued by
his great disciple, Cuvier, and as an independent development the more speculative
researches of Goethe and St. Hilaire, the anatomists consistently insisted that comparison was
the scientific method. Buffon declared that "comparison was the sovereign key to the
discovery of general laws in zoology"; Goethe insisted that "natural history is basically
comparison."
As is well known, in early literature the term "anthropology" was almost exclusively
limited to what is today termed "physical anthropology," and most of the important early
cultural anthropologists were themselves medical men (e.g., A. Bastian, D. Livingstone, K.
von den Steinen, W. H. R. Rivers, C. G. Seligman, P. Rivet). Thus it is not surprising that
Anthropology was swift to adopt the comparative (anatomical) method both fur its treatment
of physical and cultural data. Likewise in Linguistics, the second great nourishing stream of
the Comparative Religions method, Jacob Grimm, Friedrich Schlegel, and William von
Humboldt all confessed themselves to be deeply influenced by Comparative Anatomy. The
same was the case for the equally influential philosophical sociologists, A. Comte and H.
Spencer.
The discipline of Comparative Anatomy in the eighteenth and nineteenth centuries
underwent two main stages of development: first, the morphological approach; then, the
evolutionary. Comparative Religions, naively reflecting this development, was basically an
illegitimate and unreflective combination of the older morphological approach to the
problems of comparison with the newer evolutionary frame of reference.
Linnaeus had proposed that his taxonomy represented a closed immutable system
which reflected the world of nature conceived as a fixed order of permanent structures.
Goethe, who first coined the word "morphology," while recognizing variability, nevertheless
arranged his material serially in a hierarchy of increased organization and complexity. One
series illustrated the gradations from, the simplest herb to the most complex tree; another
series illustrated the sequence of individual organs within the same plant from cotyledon to
leaf to corolla to fruit. In no case was a temporal schema employed. The progression was one
of logical-formal sequence, not growth and development In time. Goethe's scheme, despite the
claims of some of his later admirers, was never evolutionary; it was a typological series that
9
was fundamentally ahistorical while employing the most complex dialectic between the
universal and the particular, between ideal and experience, between idea and appearance,
between Being and history.
The constellation of terms used by Goethe to describe the status of the Urpflanze ­ it is
a concept (Begriff), supersensible, beyond the vicissitudes of time, symbolic, a self-revealed
mystery (das offenbare Geheimnis), visible to the mind alone-has led some interpreters to
overemphasize the idealistic character of the morphology and to fail to take into account the
dialectic between the type and history. The type is by definition ahistorical, yet it stands in a
complex relationship to the historical. "Time is governed by the oscillations of a pendulum;
the moral and scientific worlds, by oscillations between ideas and experience." This
"oscillation" has profound implications for the comparative enterprise. Meditation on
concrete plants or animals may reveal the archetype which they manifest; meditation upon
the type will reveal the vitality of the individual variations. The morphological enterprise is
"to recognize living forms as such, to see in context their visible and tangible parts, to
perceive them as manifestations of something within and thus to master them to a certain
extent in their wholeness through a concrete vision (Anschauung). (I note parenthetically that
this seems closer to what historians of religion mean when they invoke phenomenology than
the technical procedures of the philosophical phenomenological movement.) But, as in Plato,
the upward path requires the downward for its fulfillment. "The particular always underlies
the universal; the universal must forever submit to the particular." Comparison may thus
occur between the individual and the archetype; comparison may also occur between
analogous members of an atemporal series (X is homologous to Y). In either case, neither
temporal nor causal conclusions may be drawn from such comparison, that is, X is simpler
than Y, it is logically and formally prior but not necessarily prior in time.
In cultural studies, this morphological perspective might be illustrated by J. F.
McLennan in 1896, when he advocated the study of the "least-developed races" which he
defined as "the lowest and simplest," reminding his readers that "in the science of history old
means not old in chronology but in structure." It is massively and masterfully illustrated in
Mircea Eliade's Patterns in Comparative Religion. Thus when I read in Eliade his numerous
and central references to un systeme coherent behind the various manifestations and hierophanies;
when he arranges a series of hierophanies from the most elementary to the most
complex; when he insists that this systeme nowhere exists but that the archetypes preexist any
particular manifestation, that the systeme "manifest more clearly, more fully and with greater
coherence what the hierophanies manifest in an individual, local and successive fashion"-I am
reminded of such classic statements of the morphological enterprise as Goethe's letter to
Herder (Naples, May 17, 1787). The failure to recognize this ambience has been responsible
for significant misinterpretation of the role of history and the status of the patterns in Eliade's
work.
Such a morphology docs have one crucial presupposition, the scientific analogue to
which was clearly stated by Kant in his Critique of Judgement: "All such comparison
presupposes this: that in respect to her empirical laws nature has observed a certain economy
that to our judgement appears fitting, and an understandable uniformity, and this
presupposition as a principle of a priori judgement must precede all comparison." Whether
10
in the strict morphology of Eliade (or the looser works of Jung or Goodenough on symbols)
this "fitting economy" is assumed. There can only be a relatively limited number of systems
or archetypes, though there may be an infinite number of manifestations. Comparison, while
global in scope, nevertheless remains strictly limited in procedure. One may only compare
within the system or between the pattern and a particular manifestation. Comparisons within
the system do not take time or history into account; comparisons between the pattern and
manifestation are comparisons as to the degree of manifestation and its intelligibility and do
not take historical, linear development into account. If a series is proposed, it will deal with
the movement from the simple to the complex, from the perfect manifestation to the
fractured--in Goethe's term, with metamorphosis (compare Eliade on degradation,
infantilism, etc., in Patterns, chapter XIII).
THE EVOLUTIONARY
In what I have termed the second stage of development of scientific theory, the
comparative anatomists introduced the temporal dimension. Buflon, Cuvier, Lamarck, and
Darwin drew the implication that an increased complexity reflected a process of growth,
adaptation, and evolution. In so doing, they abandoned the atemporal presuppositions of their
predecessors and concentrated more and more on the variability, the internal history of
individual species. One might claim that with the introduction of a temporal frame of
reference, the comparative method, in the strict sense, was abandoned. As I have suggested,
Comparative Religion as it was practiced in the nineteenth and early twentieth centuries was
essentially an illegitimate combination of the older morphological approach to comparison
and the newer evolutionary frame of reference. As among the older comparative
morphologists, the comparative religionists held that structures were to be studied and
compared without regard to chronology and geography (or speciation, for that matter); but
rather as to an increased order of complexity. However this was linked by the anthropologists
and comparative religionists to an overall evolutionary frame of reference. The analyses of
structure and the taxonomic classifications were employed to demonstrate a chronological
chain of religious and cultural evolution on a global scale. Paradoxically (and impossibly), the
newer evolutionary and temporal perspective was linked to the older ahistorical methodology.
The detection of simplicity now yielded historical origins (e.g., "the primitive") rather than
"logical" Urformen.
The cultural evolutionists wedded this hybrid method to the encyclopaedic style of presenting
data. The dynamics of this mode (as discussed above), its contextless lists, its lack of depth, its
"hard" surface made this style ideal for avoiding the implications arising from this
combination of methods. As was the case in the encyclopaedic style, a careful examination of
the works of the cultural evolutionists will reveal no principles of organization or comparison.
The meticulous arrangement of series, the micromovements and variations so essential to the
biological evolutionists' work, were totally lacking in the studies of their cultural
counterparts.
11
The problems inherent in this combination are displayed with 'clarity in the introduction to
E. B. Tylor's classic work, Primitive Culture (1871). Tylor begins with a rather confused
attempt to apply the Linnaean category of species to cultural phenomena:
"What the ethnographer's task is like may be almost perfectly illustrated by
comparing these details of culture with the species of plants and animals as studied by the
naturalist. To the ethnographer the bow and arrow is a species, the habit of flattening
children's skulls is a species, and the practice of reckoning numbers by tens is a species. The
geographical distribution of these things and I their transmission from region to region, have
to be studied as the (naturalist studies the geography of his botanical and zoological species."
This attempt to answer the question of what in culture / corresponds to the morphologist's
organism or the evolutionist's (species remains unconvincing in Tylor as well as in later
theorists. As Levi-Strauss shrewdly observed, "The historical validity of the naturalist's
reconstructions is guaranteed, in the final analysis, by the biological link of reproduction. An
ax, on the contrary, does not generate another ax."
As Tylor continued to articulate his methodological presupposition, it became clear
that he was dependent upon the combination of evolution with an ahistorical methodology.
The purpose of his scheme of classification is that of demonstrating "how the phenomena of
culture may be classified and arranged, stage by stage, in a probable order of evolution." But
in so doing Tylor claimed, as had the older comparative morphologists, the privilege of
ignoring chronology and geography: "Little respect need be had in such comparisons for date
in history or for place on the map; the ancient Swiss lake-dweller may be set beside the
Medieval Aztec, and the Ojibwa of North America beside the Zulu of South Africa. As Dr.
Johnson contemptuously said when he had read about the Patagonians and South Sea
Islanders in Hawkesworth's Voyages, 'one set of savages is like another.' How true a
generalization this really is, any Ethnological Museum may show." Comparison and the
recognition of parallels, in such a program, might be divorced from historical studies and
considerations. Similarities were to be explained on the basis of their reflecting similar
"logical stages" of human development and cultural evolution (the Urmensch corollary to the
Urpflanze) though, again impossibly, chronological implications were to be drawn. This
hypothesis and its relationship to theories drawn from comparative anatomy was clearly
formulated by J. G. Frazer in the "Preface" to the concluding part of The Golden Bough: "If
there is one general conclusion which seems to emerge from this mass of particulars, I
venture to think that it is the essential similarity in the working of the less developed human
mind among all races, which corresponds to the essential similarity in their bodily frame
revealed by comparative anatomy."
These two legacies from biological theories-the linkage of the comparative method to
evolutionary presuppositions and its ahistorical method-combined in a motley hybrid, when
brought over into the cultural realm and religious scholarship came swiftly under attack.
Perhaps the most significant critic was the doyen of American anthropology, Franz Boas, in
his article first published in 1896 entitled "The Limitations of the Comparative Method of
Anthropology." In this paper (which had great impact when published, even though its
argument seems slight today), Boas insisted that "the same phenomena may develop in a
multitude of ways" and "demanded" that "comparisons be restricted to those phenomena
12
which have been proved to be the effects of the same cause"- thus attempting to introduce an
analogue of "the biological link of reproduction" (Levi-Strauss). It needs to be stressed that
Boas did not reject comparison per se, as this point was immediately misunderstood by some
of Boas's American and British colleagues. Indeed, as late as 1938, he called for "a cultural
morphology founded on comparative studies of similar forms in different parts of the
world."64 He sought rather to improve the method, to tighten up the criteria by which
similarities were determined and on the basis of which comparisons might be made. He did so
by introducing the temporal dimension so important to the biological researches of his time.
Boas suggested the employment of what he termed "the historical method"; "Historical
inquiry must be considered the critical test that science must require before admitting facts as
evidence. By means of it, the comparability of the collected material must be tested, and
uniformity of processes must be demanded as proof of comparability. "
The results of this article were unfortunate. Boas had linked comparison to naive
evolutionary theory and had juxtaposed what he termed historical method to this
combination, thus setting off a conflict which continues to the present day (Boas's historical
method, it may be noted, seems no less naive today). With the general loss of faith in social
Darwinism and theories of progress, with the rising dominance of historicistic and positivistic
thinking in the social sciences, Boas's article could be and was used as the excuse for
jettisoning the comparative enterprise and for purging one's work of all but the most limited
and specific comparisons. As Paul Mercier noted in his recent history of Anthropology, Boas's
article represented the end of an era. The problem for History of Religions today is that a new
era has not yet arrived to take its place. And thus we need to reflect on our own history, for:
"If anthropology returns to the comparative method, it will certainly not forget what it has
learned meanwhile in general and what it has learned about the limitations of the method in
particular. It will return only in that spiral like movement, so characteristic of scientific
thought, arriving after half a century at the same point but at a higher level. It will know
better how and what to compare than it knew fifty years ago." This applies to our discipline
as well.
AFTER WORD
The discussion of general methodological principles for comparison has not advanced since
the original publication of this essay. See, among others, the excellent reviews of the state of
the question by Gopala Sarana, The Methodology of Anthropological Comparisons: An
Analysis of Comparative Methods in Social and Cultural Anthropology)' ([Viking Fund
Publications in Anthropology, Number VIII Tucson, 1975) and Ulrich Weisstein, Comparative
Literature and Literary Theory: Survey and Introduction (Bloomington, 1973) both of
which contain rich bibliographies.
Note should be taken of the careful specification of eight modes of comparison within a
limited area by Morton Smith, Tannaitic Parallels to the Gospels ([Journal of Biblical
Literature Monograph Series, Volume VI] Philadelphia, 1951) and its effective application by
Jacob Neusner, Aphrahat and Judaism: The Christian-Jewish Argument in Fourth-Century
Iran ([Studia Post-Biblica, Vol. XIX] Leiden, 1971) esp. pp. 187-195. If latitude be still
13
permitted on the ,,wider cross-cultural comparisons, surely this much ought to be required of
any historian of religions working within a particular cultural context.