Empiricism in Babylonian Omen Texts and the Classification of Mesopotamian Divination as Science Author(s): Francesca Rochberg Source: Journal of the American Oriental Society, Vol. 119, No. 4 (Oct. – Dec., 1999), pp. 559- 569 Published by: American Oriental Society Stable URL: http://www.jstor.org/stable/604834 . Accessed: 03/01/2011 11:59

Your use of the JSTOR archive indicates your acceptance of JSTOR’s Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR’s Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=aos. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

American Oriental Society is collaborating with JSTOR to digitize, preserve and extend access to Journal of the American Oriental Society.

http://www.jstor.org

EMPIRICISM IN BABYLONIAN OMEN TEXTS AND THE CLASSIFICATION OF MESOPOTAMIAN DIVINATION AS SCIENCE

FRANCESCA ROCHBERG

UNIVERSITY OF CALIFORNIA, RIVERSIDE

This paper reevaluates the empirical content of Babylonian omen protases in the light of more recent discussions among philosophers of science of the relation between observation and theory, and argues against separating observationally derived phenomena, understood as physical objects of ordinary sense perception, from those derived by use of schematic symmetries. The goal of this paper is to ascertain the criteria of observation implied by omen texts in order to evaluate the “empirical” nature of Mesopotamian divination in the wider framework of the history of science.

To the memory of John A. Phillips

INTRODUCTION: THE CLASSIFICATION OF

BABYLONIAN DIVINATION AS SCIENCE

THE STUDY OF ANCIENT MESOPOTAMIAN SCIENCE has

been largely devoted to the reading and analysis of the

many astronomical records, both observational and com-

putational, found in the southern Mesopotamian cities of

Babylon and Uruk, dating mostly from the period after 500 B.C.] Earlier texts of astronomical interest, found in

Assyrian sites such as Nineveh and Assur, provide evi- dence of the incorporation of astronomical events within a vast system of divination that predicted the future on the basis of natural and other events of many kinds.2 Such events were viewed as signs produced by the gods by

Portions of this paper were presented at the 208th meeting of the American Oriental Society in New Orleans, April 1998. I want to thank Sir Geoffrey Lloyd and Prof. Ernan McMullin for their insightful and encouraging readings of the paper.

1 See 0. Neugebauer, Astronomical Cuneiform Texts, 3 vols. (London: Lund Humphries, 1955); idem, A History of Ancient Mathematical Astronomy, 3 vols. (Berlin: Springer Verlag, 1975), with bibliography.

2 See S. Parpola, Letters from Assyrian Scholars to the Kings Esarhaddon and Assurbanipal, vols. I and II, Alter Orient und Altes Testament, vol. 5.2 (Kevelaer: Verlag Butzon & Berker, 1970 and 1983); H. Hunger, Astrological Reports to Assyrian Kings, State Archives of Assyria, vol. 8 (Helsinki: Helsinki Univ. Press, 1992); and S. Parpola, Letters from Assyrian and Babylonian Scholars, State Archives of Assyria, vol. 10 (Hel- sinki: Helsinki Univ. Press, 1993).

means of which humans were forewarned of future events.

Foreknowledge could therefore be obtained by systematic consideration and interpretation of the omens.3

Assyriologists have considered the omen texts a form of science in Mesopotamia primarily because many of the phenomena of interest in these texts are of the phys- ical, natural, world.4 Thus many of the omen protases of

3 For general studies of Mesopotamian divination, see A. L.

Oppenheim, Ancient Mesopotamia: Portrait of a Dead Civiliza- tion, rev. ed. E. Reiner (Chicago: Univ. of Chicago Press, 1977); and J. Bott6ro, Mesopotamia: Writing, Reasoning and the Gods, tr. Z. Bahrani and M. van de Meiroop (Chicago: Univ. of Chicago Press, 1992). Editions of Babylonian omen texts may be found in E. Leichty, The Omen Series Summa Izbu, Texts from Cune- iform Sources, vol. 4 (Locust Valley: J. J. Augustin Publisher, 1970); Erica Reiner and David Pingree, Babylonian Planetary Omens, vols. I-II, Bibliotheca Mesopotamica, vols. 2.1 and 2.2 (Malibu: Undena Publications, 1975 and 1981); F Rochberg- Halton, Aspects of Babylonian Celestial Divination: The Lunar Eclipse Tablets of Enuma Anu Enlil, Archiv fur Orientforschung, Beiheft 22 (Horn: Verlag Ferdinand Berger & Sohne, 1988); Wilfred H. van Soldt, Solar Omens of Enama Anu Enlil: Tablets 23 (24)-29 (30) (Leiden: Nederlands Historisch-Archaeologisch Instituut te Istanbul, 1995); A Leo Oppenheim, The Interpreta- tion of Dreams in the Ancient Near East, Transactions, vol. 46.3 (Philadelphia: American Philosophical Society, 1956); F. R. Kraus, Texte zur babylonischen Physiognomatik, Archiv fur Orientfor- schung, Beiheft 3 (Osnabruck: Biblio-Verlag, 1939).

4 Note the discussion of omens under the rubric “science” in the general overview of Mesopotamian culture by W. von Soden,

559

Journal of the American Oriental Society 119.4 (1999)

the celestial omen series Enuma Anu Enlil, parts of Summa alu dealing with fauna, of Summa izbu focusing on anomalous animal and human births, of Alamdimmu, that deal with the variable forms of the human anatomy, and even parts of the Ziqiqu dreambook, have come to be

inspected as sources for understanding the Mesopotamian attempt to grasp the workings of nature. Because the diverse systems of Mesopotamian divination all stemmed from a belief in the gods’ involvement in the physical, as well as the social worlds, and because of the close rela-

tionship of divination to apotropaic ritual magic, the body of knowledge represented by the omen texts has not al-

ways been classified as science, particularly by historians of science who prefer to see in this material a form of pre- or proto-science.5

Beginning in the 1960s, however, philosophers and

anthropologists have argued about the similarities and dif- ferences that relate or distinguish traditional (religious/ magical) and modem (scientific) thinking. As well, they have also discussed the implications of accepting a rela- tivism of “modes of thought” for defining both science and the criteria of scientific truth.6 Those disposed to- ward relativism extend the term science to divination

The Ancient Orient: An Introduction to the Study of the Ancient Near East, tr. Donald G. Schley (Grand Rapids, Mich.: William B. Eerdmans, 1994), 153-57; celestial omens under the heading “Astronomy in Mesopotamia,” in H. W. F Saggs, Civilization

Before Greece and Rome (New Haven: Yale Univ. Press, 1989), 236-38; and the chapter entitled “Divination and the Scientific

Spirit,” in J. Bottero, Mesopotamia, 125-37. 5 A. Aaboe, “Scientific Astronomy in Antiquity,” in The Place

of Astronomy in the Ancient World, ed. F R. Hodson (London: Oxford Univ. Press for The British Academy, 1974), 21-42; O. Pedersen, Early Physics and Astronomy: A Historical Intro-

duction, rev. ed. (Cambridge Univ. Press, 1993): see ch. 1 “Science Before the Greeks,” under the subheading, “The Myth- ological Explanation of Nature,” pp. 7-9.

6 See Robin Horton, “African Traditional Thought and West- ern Science,” Africa 37 (1967), reprinted as ch. 7 of Patterns of Thought in Africa and the West: Essays on Magic, Religion, and Science (Cambridge: Cambridge Univ. Press, 1993); and Stanley Jeyaraja Tambiah, Magic, Science, Religion, and the Scope of Rationality (Cambridge: Cambridge Univ. Press, 1990), ch. 6:

“Rationality, Relativism, the Translation and Commensurability of Cultures.” See also the collection of papers edited by Bryan R.

Wilson, Rationality (Oxford: Basil Blackwell, 1970); the now classic collection edited by R. Horton and Ruth Finnegan, Modes

of Thought (London: Faber and Faber, 1973); and the more re-

cently edited collection by David R. Olson and Nancy Torrance, Modes of Thought: Exploration in Culture and Cognition (Cam-

bridge: Cambridge Univ. Press, 1996).

and magic. This discussion becomes relevant to the study of magical, alchemical, and astrological sources in the

history of Western science, with the result that the cri- teria defining science “in general” established by Boyle and the Royal Society of London in the seventeenth cen-

tury have at long last been rejected, as Barnes, Bloor, and Henry put it, “not least because philosophers and historians have now demonstrated repeatedly that the contents of the accepted, authentic history of science are not capable of being demarcated by this criterion, or indeed by any other.”7

For historians in the current post-positivistic climate, science has ceased to be the exclusively logical and em-

pirical inquiry it once was, clearly and cleanly separable from theology, metaphysics, and other speculative or

“mythic” forms of thought.8 The impact of this histori-

ography is such that many philosophers of science no

longer exclude all but “matters of fact and ratiocination” from science and have even come so far as to call into

question the old demarcation game itself.9 Historical con- siderations aside, on purely epistemological grounds some have argued that “there is apparently no epistemic feature or set of such features which all and only the ‘sciences’ exhibit…. It is time we abandoned that lingering ‘scien- tistic’ prejudice which holds that ‘the sciences’ and sound

knowledge are coextensive; they are not.”‘0

7 Barry Barnes, David Bloor, and John Henry, Scientific

Knowledge: A Sociological Analysis (Chicago: Univ. of Chicago Press, 1996), 149.

8 The following are merely a suggestion of what is now an enormous literature: Ron Millen, “The Manifestation of Occult

Qualities in the Scientific Revolution,” in Religion, Science and Worldview: Essays in Honor of Richard S. Westfall, ed. M. J. Osler and P. L. Farber (Cambridge: Cambridge Univ. Press, 1985), 185-216; the collection of papers edited by David C.

Lindberg and Robert S. Westman, Reappraisals of the Scientific Revolution (Cambridge: Cambridge Univ. Press, 1990); also the

papers of Keith Hutchinson, “What Happened to Occult Quali- ties in the Scientific Revolution,” B. J. T Dobbs, “Newton’s Al-

chemy and His Theory of Matter,” and other papers collected in The Scientific Enterprise in Early Modern Europe: Readings from Isis, ed. Peter Dear (Chicago: Univ. of Chicago Press, 1997).

9 See “Introductory Remarks” of Marx W. Wartofsky in Sci-

ence, Pseudo-Science and Society, ed. Marsha P Hanen, Margaret J. Osler, and Robert G. Weyant (Waterloo, Ont.: Wilfred Laurier Univ. Press, 1979), 1-9. See also Barnes, Bloor, and Henry, Scientific Knowledge, ch. 6, “Drawing Boundaries,”140-68.

10 L. Laudan, Beyond Positivism and Relativism: Theory, Method, and Evidence (Boulder, Colo.: Westview Press, 1996), 85-86.

560

ROCHBERG: Empiricism in Babylonian Omen Texts

Where ancient Mesopotamian traditions are concerned, it would seem that the “scientistic prejudice” does linger and old demarcations prevail. Otherwise, why do ideas

persist such as that science begins only with the Greeks and continues to evolve to the present day?11 Or that

genuine science in Babylonia begins with the mathemat- ical astronomy at the end of the sixth century B.C.? The reason for this surely has to do with the fact that the cuneiform “scientific texts,” whatever these are taken to include and however they are defined, are our earliest known historical sources for science; and so, inquiring into Mesopotamian science carries the extra burden of

inquiring into the origins of science. To raise the question of when science begins already implies a demarcation between science and pre-science, or non-science, but the “scientistic prejudice” becomes explicit when, as von Staden said, “the quest for the ‘origins’ of science often is tacitly accompanied by a search for ancient motivations that resemble modern scientific ones.”‘2

If, however, classification of the omen texts is not to be based on an argument from affinity with modern or other known sciences, on what set of criteria is it to be

11 Despite the sizable body of work by O. Neugebauer and A. Aaboe showing the debt to Mesopotamia of Greek astro- nomical science, statements from the wider field of history and

philosophy of science still frequently assume, as Philip Kitcher

does, that “scientists in the tradition that extends beyond the seventeenth century to the ancient Greeks have been moved by the impersonal epistemic aim of fathoming the structure of the world.” See P. Kitcher, The Advancement of Science: Science without Legend, Objectivity without Illusions (Oxford: Oxford Univ. Press, 1993), 94.

12 Heinrich von Staden, “Affinities and Elisions: Helen and

Hellenocentrism,” Isis 83 (1992): 588. This tendency can be found in serious scholarship which carries over into less sophis- ticated work, where it does even more damage because of its wider and less knowledgeable readership. One example (there are many) is the following from a collection of readings in an- cient history where James Breasted is quoted as saying about the Edwin Smith surgical papyrus: “Here we find the first scientific observer known to us, and in this papyrus we have the earliest known scientific document.” Breasted’s comment was typical in

1930, but what follows from the editor ought now not to be: “The unknown author, who lived sometime during the Old Kingdom [sic], has written a treatise on surgery in which he inductively draws conclusions from a body of observed facts.” He goes on to

point out that magic was only used in one of the forty-eight cases described and he asserts that “a truly scientific attitude” is exem-

plified for the first time by this document. See Readings in An- cient History, ed. Nels M. Bailkey, 3rd ed. (Heath, 1987), 37.

based? We may strive not to distort ancient systems of

thought by the imposition of our own definitions and criteria and may try to determine the content, aims, and methods of such systems “from within.” But surely if the

discipline of scholarly divination bears no relation to the

discipline we have defined and determined by our own social and cultural concensus to be science, why do we seek to classify the native discipline of omens as “sci- ence” at all? Without attributing any necessary universal criteria to science, I think a simple answer is that, cor-

rectly or incorrectly, we recognize in this Mesopotamian tradition aspects of what we term science in our, i.e., the Western tradition. However carefully we may try to re- construct the terms of an “alien” system on the basis of primary texts, the meaning of the term science is not entirely recreated in every scholarly historical investi- gation. Though our classification of Mesopotamian divi- nation as “science” probably would make no sense to a

Babylonian, it serves to make comprehensible to us some aspects of this ancient intellectual tradition by connoting a number of features: among them, empiricism and sys- tematization of knowledge. Within the wider framework of assessing cuneiform sources in terms of the criteria by which Mesopotamian divination might be classified as science, the present paper tackles only one such criterion, namely, the empirical character of the omens.

THE EMPIRICAL CHARACTER OF OBJECTS OF

(MESOPOTAMIAN) SCIENTIFIC KNOWLEDGE

Most would agree that the desire to comprehend natu- ral phenomena is the common denominator for science

regardless of its cultural manifestation. However, in ref- erence to the Mesopotamian omen texts, to equate omens with an inquiry into such phenomena does not fairly represent these sources and seems to lose sight of the fact that Mesopotamian omen texts concerned signs of many kinds, of which natural phenomena formed but one. That Enima Anu Enlil and its companion piece, MUL.APIN, have generally been understood as the chief sources for Babylonian physical science before 500 B.C. takes these astronomical sources out of their broader in- tellectual context just to satisfy modern Western tastes. In an effort then to appreciate the full range of interests comprising the discipline developed for the study of phenomena, it seems important not to limit the general discussion of Mesopotamian divination as science to those parts of it, such as the astronomical omens, that have the greatest similarity to more familiar sciences. The “Babylonian” approach seemingly makes no episte- mological or methodological distinction between astro- nomical omens and other items of scholarly divination,

561

Journal of the American Oriental Society 119.4 (1999)

as we see clearly in the letters and reports of the Neo-

Assyrian diviner-scholars.13

Beginning with Aristotle’s Posterior Analytics, the dis- cussion of scientific knowledge has taken sense perception as fundamental to science and the basis for generaliza- tions about the natural world:

… we cannot employ induction (irmayoyri) if we lack

sense-perception, because it is sense-perception that

apprehends particulars. It is impossible to gain scientific

knowledge ( ntcriuTil) of them, since they can neither be apprehended from universals without induction, nor

through induction apart from sense-perception. (Posterior Analytics, I.xviii)

Speaking strictly of the protases, the arrangement of

subjects into categorical groups within the various lists of omens seems to point toward an empirical foundation for the lists in general, since any sort of classification of sub-

jects would be difficult to imagine without such a founda- tion. The study of signs, in the form of Babylonian omen

series, however, does not exhibit the same empirical constraints as are found in the study of some natural

phenomena, particularly astronomical phenomena that be- have in accordance with certain limited parameters. The

organization of tablets in the series Summa alu, for exam-

ple, assembles and classifies phenomena of widely dispar- ate subjects. The omens that deal with human phenomena would seem to be endlessly and unsystematically variable, as in the series Summa alu, which defines its interests rather broadly; Alamdimma, which focuses on the physi- ognomic characteristics of people; and SA.GIG, which studies the symptoms of the sick. Clearly there is some

overlap in what is of interest from series to series, but each series of so-called unprovoked (or non-impetrated) omens establishes a field of phenomena deemed appropriate for

study within its particular confines. The scope of the series Summa alu encompasses things of “real life” relating to

13 In a letter from the celestial divination expert Marduk-

sapik-zeri to king Assurbanipal, the scribe describes the breadth of his learning: “I fully master my father’s profession, the dis-

cipline of lamentation; I have studied and chanted the Series. I am competent in [.. .], ‘mouth-washing’ and purification of the

palace […] I have examined healthy and sick flesh. I have read the (astrological omen series) Enuma Anu Enlil [. ..] and made astronomical observations. I have read the (anomaly series) Summa izbu, the (physiognomical works) [Kataduqqui, Alam- dimmti and Nigdimdimmu, . . . and the (terrestrial omen series) Sum]ma alu.” Translation of S. Parpola, Letters from Assyrian and Babylonian Scholars, 122, no. 160: 36-42.

cities and houses, flora, fauna, water, fire, lights, or to an individual’s thoughts, prayers, actions of daily life (sex, sleep, family quarrels), and his perception of demons and

ghosts. This last subject, contained in tablets 19 and 21, raises

our primary question about the empirical nature of the omens. What does it mean to state that a demon was seen in a house?14 Or to construct omens from footprints of

gods or from cries of a ghost?15 The omens about super- natural entities are constructed in precisely the same way as are other physical appearances in omen protases, i.e., as subjects of the verb amdru (the basic meaning of which is “to see”) in the N-stem (nanmuru). In translations of omen protases, the passive forms of amaru, meaning “to be seen,” are often rendered as though active, “to appear,” as in the example: summa ina bit ameli hallulaya innamir “if a hallulaya-demon appears (literally, “is seen”) in

somebody’s house….”16 When phenomena we do not

regard as “observable” in the normal way, i.e., by ordinary sensory perception, are included in the list of omen pro- tases, such protases necessitate a reexamination of the cri- teria underlying observables in the Mesopotamian view.

And what of the omen protases that refer not to “super- natural” but to purportedly natural phenomena that cannot occur in nature? Unless we have totally misunderstood what is said in the following examples, these few omens

may serve to illustrate such non-occurring or unobservable

phenomena:

If the sun comes out in the night and the country sees its light everywhere: there will be disorder in the coun-

try everywhere (EAE 25 I 1).17

If the sun comes out in the night and lasts until the

morning: Enlil [.. .] the rumor of [.. .] if Erra speaks the people of the land will be diminished, the entire

country will not [ . .] rain (EAE 25 I 2).

If the sun comes out during the evening watch: an

uprising in the land [. ..] (followed by omens for the sun

coming out in the middle and morning watches, as well as the sun rising during various watches with other as- tral bodies “standing in front” of it) (EAE 25 I 5-10).

14 [DIg ina E NA MA]SKIM GIM UZ IGI E.BI BIR-ah “If a

goat-like demon appears in a man’s house: that house will be

dispersed” (CT 38 25). 15 See Summa alu tablet 19: 16-22 and 36-47, respectively;

see also CT 38 25 and KAR 396 dupl. lines 36-41. 16 CT 38 25b: 6 = Summa alu 19 6, also lines 7-12; see Sally

Moren, “The Omen Series ‘Summa Alu’: A Preliminary Inves-

tigation” (Ph.D. diss., Univ. of Pennsylvania, 1978), 71. 17 Wilfred H. van Soldt, Solar Omens of Enuma Anu Enlil, 32.

562

ROCHBERG: Empiricism in Babylonian Omen Texts

If we assume that the criterion for observability is access by ordinary sensory perception, then omens are problematic. The editor of the above text has translated, “If the ‘sun’. .. ,” in each case, implying that Samas, here, wr. with the normal logogram 20, does not refer to the sun, but perhaps something else we have not properly identi- fied. This demands that the something else we have not identified be observable by the criterion of ordinary see- ing. But he also notes that these omens continue the topic of the previous tablet, in which the protases of the final section deal with the unexpected appearance of the sun.18

Another example may be found in the omens that provide predictions for lunar eclipses “occurring” on days 17, 18, 19, 20, 21, and even istu UD.1.KAM adi UD.30. KAM” (on any day) from the first to the thirtieth (of the month).”19 These omens indicate a schema for eclipses on days of the month beginning with possible days of opposition, 14, 15, and 16, and continuing until con- junction at the end of the month. If we think the omen schemata refer only to what we take to be empirically observable phenomena, such a schema, setting out days 14-30 for “observing” a lunar eclipse, does not “make sense,” but may be excused if we assume it was not yet known that in the lunar calendar, days 17-30 (as well as days 1-13) are excluded as possible lunar eclipse days. Even if such limits on eclipse occurrences were not known when the Enuma Anu Enlil series was first com- piled, certainly by the Neo-Assyrian period the scribes knew well what were the possible days for the occurrence of a lunar eclipse. Many copies of Enuma Anu Enlil were made during this period, yet the faulty omens were still not deleted from the series.

If our goal as interpreters of this body of texts is to pre- serve the practical rationality of the Babylonians, we may choose to interpret these “eclipses” in accordance with Neo-Assyrian commentaries that gloss the term AN.KUlo “eclipse” with an explanation that clouds darken the moon.20 But given that the same term, attali sin, seems to be applied to astronomical, meteorological, and unob- servable eclipses, can the idea of a universal and objective empirical description really be a criterion for deciding when the text means an eclipse “occurring” on a physi- cally impossible day, or an eclipse caused by a cloud?

18 Ibid.. 32. 19 See my Aspects of Babylonian Celestial Divination, chs.

7-9, 11-12 (in EAE 17-18, 19, 21-22). 20 E.g., summa sin adirma istenis irim /I attalu ina erpeti sa-

limti raqqati izzizma “if the moon is dark and is totally eclipsed (literally: covered)” (text writes salimtu raqqatu syllabically, without indicating the genitive); see my Aspects of Babylonian Celestial Divination, 285, appendix 2.4 rev. 3-4.

The question of such veracity was raised by Erle Leichty in his treatment of the series Izbu. His determi- nation was that, indeed, most of the birth anomalies de- scribed in the protases could be identified with attested birth abnormalities, but that some could not. He adds, “from this we do not wish to argue that all the omens in the series were actually observed. This is simply not true. In addition to the cases where omens were obviously added in an attempt to make the series all-inclusive, there are also occasional omens where the anomaly is naturally impossible.”21 Leichty makes the point, and others (my- self included) have as well, that in the expansion and redaction of omen collections, additional omens were introduced, not on an empirical basis, but on the basis of the requirements of formal schemata into which phe- nomena were arranged.22 Most recently, John Britton and Christopher Walker have said that a method of “extrapo- lation from observed experience ad absurdum”23 charac- terizes Mesopotamian divination, citing as an example the series of lunar eclipse omens for days 14, 15, 16, 19, 20, 21, of which only the fourteenth to sixteenth days are possible in a lunar calendar.24

This approach, to quote Sahlins, is “a common or gar- den variety of the classic Western sensory epistemology: the mind as mirror of nature.”25 It presumes the existence of a distinction between fact and fiction, “real” knowl- edge and fantasy, in the omen protases. Accordingly, the compilers of the series worked from an “empirical” core of actual observations of phenomena, but included phe- nomena of their own conceptual devising for the sake of completeness or the symmetry of certain scholastic sche- mata. Although omen protases have been analyzed this way, no Assyriologist has suggested that the Babylonians could not tell the difference between “true” and “made- up” phenomena. On the contrary, the emphasis has been that the fabricated phenomena were later introductions into lists of empirically established signs and that the late introductions served scholastic purposes of comprehen- siveness.26 Such an interpretation salvages Babylonian

21 E. Leichty, The Omen Series Summa Izbu, 20. 22 See J. Bott6ro, Mesopotamia, 134-35. 23 J. Britton and C. Walker, “Astronomy and Astrology in

Mesopotamia,” in Astronomy before the Telescope, ed. C. Walker (London: British Museum Press, 1996), 42.

24 See my Aspects of Babylonian Celestial Divination, 38-40. 25

Sahlins, How “Natives” Think: About Captain Cook, For Example (Chicago: Univ. of Chicago Press, 1995), 6.

26 Britton and Walker, “Astronomy and Astrology in Mesopo- tamia,” 42-43, have suggested that the absurdities in the lunar eclipse sequence could have stemmed from a scribal error which was then perpetuated in the canonical text.

563

Journal of the American Oriental Society 119.4 (1999)

practical rationality, but leaves open the question of why, if the scribes indeed thought that some omens were plausible and others not, all the omens were formulated identically “if x, then y,” and why all the protases appear to be signs of equal mantic validity. The omen texts made no distinctions between empirically true signs and “con-

ceptually fabricated” signs. As pointed out above, even

supernatural entities are formulated as “if x is observed (innamir).”

It strikes me that to designate some omen protases as ‘absurd” or “impossible” both begs the question of what it was the diviners were trying to observe, and reduces the criterion of empiricism in Babylonian divination (as in all sciences) to simple sense data presupposed as hav-

ing a basis in a reality apart from the observer. Not only can we not make this claim on the basis of any evidence, but such a reduction reflects an approach characteristic of a simple empirical inductivism that is no longer credi- ble.27 Yet many of our translations of omens and our com- ments on their meaning reflect just such an approach.28 To speak of the criterion of observability as one of untem-

27 The following represents but a small selection from the dis- cussion of the interaction of theory and observation, ongoing since the 1950s: N. R. Hanson, Patterns of Discovery: An Inquiry into the Conceptual Foundations of Science (Cambridge: Cam-

bridge Univ. Press, 1958), ch. 5; K. R. Popper, The Logic of Sci-

entific Discovery (London: Hutchinson, 1968), appendix 10; idem, Objective Knowledge, 341-61; T S. Kuhn, The Structure of

Scientific Revolutions (Chicago: Univ. of Chicago Press, 1970), ch. 10; Mary Hesse, “Is There an Independent Observation-

Language?” in The Nature of Scientific Theories, ed. R. Colodny

(Pittsburgh: Pittsburgh Univ. Press, 1970), 35-77; the collection

of papers in Images of Science: Essays on Realism and Empiri- cism, ed. Paul M. Churchland and Clifford A. Hooker (Chicago: Univ. of Chicago Press, 1985); and Philip Kitcher, The Advance-

ment of Science: Science without Legend, Objectivity without Il-

lusions (Oxford: Oxford Univ. Press, 1993), ch. 7 . 28 A parallel may be seen in the criticism offered by Robin

Horton in Patterns of Thought in Africa and the West: Essays on

Magic, Religion and Science (Cambridge: Cambridge Univ. Press,

1993), 99, who points out that moder social anthropologists’ views (he names L. Levy-Bruhl, B. Malinowski, E. E. Evans-

Pritchard, M. Gluckman, R. W. Firth, E. Leach, and J. Beattie) assume a positivistic tone when it comes to defining science as

against magical-religious thought. These anthropologists, he says, “have made it plain that they regard science as an extension of

common sense; as based on induction from observables [defined as ‘occurrences in the visible, tangible world’ (p. 98)]: and as

limiting itself to questions of how these observables behave. Once

such a position has been taken up with respect to science, it is

inevitable that the magico-religious thinking of the traditional

cultures should be seen as radically contrasted with it.”

pered sensory perception, mainly of “seeing,” takes no account of the role of cognition in perception and observa- tion,29 nor the relativity of culture, history, or language in the expression of what is observed.30 It also creates a prob- lem by applying to Babylonian omen protases a modern (though derived from a classical Greek) epistemological distinction between “empirical” and “rational” phenom- ena, a distinction that does not seem to be made in the texts. All the signs entered in the omen lists were ostensi-

bly accepted as having equivalent mantic significance. To have an interest in the observable effects of unob-

served phenomena is certainly not strange in science, but in the omens, the “observable effect” refers to the event predicted in the apodosis and has as its only con- nection to the unobserved phenomenon the fact that it is associated with that phenomenon in the particular line of the text. Still, the interest in phenomena for what they indicate about future events in the world of human

enterprise creates a context in which a range of non-

occurring, hence unobserved, phenomena can be included. Some examples stem from a physically correct under-

standing of the phenomenon in question, for example, a lunar eclipse that was known to occur but was not visi- ble from the scribes’ geographical reference point, such as one occurring elsewhere, but not visible in Nineveh. Others stem from an incorrect understanding of the

physical behavior of phenomena, for example, a lunar

eclipse in which the shadow travels across the moon the

wrong way, from west to east, a behavior extrapolated schematically.

All omens refer to “observables,” in the sense of being phenomena of interest to the diviners, because in the event of the actual observation of any one of them, their mantic meaning would be known. The things formulated as “observed” depended in part upon a conceptual (the- oretical) framework, and reciprocally theoretical schemes which applied to the phenomena depended, in part, upon observations. A category of “observables,” therefore, is

posited that is not limited to material things seen in the world, but includes unobserved entities that nonetheless could be imagined by extension or extrapolation from observed phenomena (such as eclipses on days of the month other than the days of syzygy, as well as demons that look like goats). What counted as empirically valid in the series of omens, i.e., what could or could not be an

29 Note, however, the argument (against, as he puts it “Han-

son, Kuhn, Churchland, Goodman and Co.”) for the neutrality of observation with respect to theory as well as the neutrality of

perception with respect to cognition in J. Fodor, “Observation

Reconsidered,” Philosophy of Science 51 (1984): 23-43. 30 Valerie Gray Hardcastle, “The Image of Observables,”

British Journal of Philosophy of Science 45 (1994): 585-97.

564

ROCHBERG: Empiricism in Babylonian Omen Texts

event that presaged, stemmed from the presuppositions of divination.

That the system of divination itself determined what the diviner should attend to in the world of phenomena, natural and otherwise, illustrates the by now commonly held notion that observation is “theory-laden.”3′ Here, “theory” refers to the conceptual schemes evidenced in the omen texts that organize and systematize the diversity of phenomena. In view of the interdependence between observation and theory, the omen schemata become clues to the empirical aspect of divination. In the absence of sources making explicit what the Babylonian scribes’ cri- teria of observation were, the schemata themselves must indicate to us what was deemed “observable.” Indeed, div- ination itself provided the framework within which pro- tases were expressed in a certain observation language, functioned as observation statements, and determined what was observable. The mere inclusion of phenomena within omen series, and the regular use of the verb “to observe” (amdru) in the protases, defines the items of the protases as observable objects, and objects of knowl- edge, regardless of their physical status.

The use of an ancient divinatory language, whose mean- ing reflects accepted notions of what was observable, parallels other scientific observation languages attested in various periods and pertaining to different “conceptual frameworks.”32 “The observable,” as Wartofsky put it, is “the index of the whole framework of science, or of the standard beliefs of the scientific community … if the

31 See Norwood R. Hanson, Patterns of Discovery, ch. 1, for a definition of theory-ladenness. The idea has deep roots in the

history of philosophy. The logical positivist H. Reichenbach said in 1924: “Every factual statement, even the simplest one, contains more than an immediate perceptual experience; it is

already an interpretation and therefore a theory…. We shall have to make use of the scientific theory itself in order to inter- pret the indications of our measuring instruments. Thus we shall not say, ‘a pointer is moving,’ but, ‘the electric current is increas- ing.’ The most elementary factual statement, therefore, contains some measure of theory” (apud Michael Friedman, “Philosophy and the Exact Sciences: Logical Positivism as a Case Study,” in Inference, Explanation, and Other Frustrations: Essays in the Philosophy of Science, ed. John Earman [Berkeley: Univ. of California Press, 1992], 86).

32 For the construct “conceptual scheme,” see Donald David- son, “On the very idea of a conceptual scheme,” Proceedings and Addresses of the American Philosophical Association 47 (1974): 5-20; and the discussion in Ian Hacking, Why Does Language Matter to Philosophy? (Cambridge: Cambridge Univ. Press, 1975), ch. 12. See also his “Language Truth and Reason,” in Rationality and Relativism, ed. M. Hollis and S. Lukes (Cambridge, Mass.: MIT Press, 1982), 48-66.

sense of observable shifts from one to another frame- work, it may also be seen that the frameworks of science also shift, historically, so that the standard observables of one period are either augmented or replaced by those of another.”33 It follows from this notion that if we study what it was that Babylonian diviners considered observ- able, we will understand something of the “conceptual framework” of the system of divination. The evidence for such a study must be the omen protases, which either state plainly that something was “observed,” or which are formulated as observation predicates, i.e., protases in the form “if x is red,” or “if x is like a goat,” or “if x stands in front of Jupiter.”

OMENS ARE NOT “OBSERVATION STATEMENTS”

Although formulated frequently as observation predi- cates, phenomena recorded in omen protases were nec- essarily only potentially observable, because in no case do the omens function as a record of observations of identifiable (i.e., datable) instances.34 They represent ab- stractions from experience and cognition, (mere) even- tualities, “ifs.”35 The diviner watched for occurrences of

33 Marx W. Wartofsky, Conceptual Foundations of Scientific Thought: An Introduction to the Philosophy of Science (London: MacMillan, 1968), 120.

34 See P. Huber, “Dating by Lunar Eclipse Omina with Specu- lations on the Birth of Omen Astrology,” in From Ancient Omens to Statistical Mechanics: Essays on the Exact Sciences Presented to Asger Aaboe, ed. J. L. Berggren and B. R. Goldstein (Copen- hagen: University Library, 1987), 3-13. Huber demonstrates a statistically probable correlation between some lunar eclipses described in omen protases and actual datable historical eclipses. He argues that textual “unreliability” due to transmission or scribal manipulation will not matter if one holds as an initial hy- pothesis that no such correlations exist between omen eclipses and historical eclipses. If a fit between the description of the eclipse in the protasis and a historical datable eclipse is “implau- sibly good,” that hypothesis must be rejected and correspond- ingly, the claim that some omen phenomena refer to specific historical and datable events is supported. Huber’s analysis poses a historical dilemma in that the eclipses he has determined as “real” date to the third millennium (2301, 2264, and 2236), long before the occurrence of any extant textual evidence of celestial divination or astronomy in Mesopotamia.

35 Note the occurrence of the Sumerogram SUM.MA.rME?l “the ifs,” in the summary to a catalogue of SA.GIG ND 4358 + :93, published by I. Finkel, “Adad-apla-iddina, Esagil-kin-apli, and the Series SA.GIG,” in A Scientific Humanist: Studies in Memory of Abraham Sachs, ed. E. Leichty, M. deJ. Ellis, and P. Gerardi, Occasional Publications of the Samuel Noah Kramer Fund, no. 9 (Philadelphia: The University Museum, 1988), 152 and note 82.

565

Journal of the American Oriental Society 119.4 (1999)

phenomena, but the written omens in his handbook stood for general cases of such phenomena whenever such occur.

The difference between phenomena recorded in omen protases and observations of phenomena is made clear in the so-called reports of the Babylonian and Assyrian astronomers of the Neo-Assyrian kings. This corpus spans the period from 708 to 648 B.c. and documents the systematic observation and reporting of heavenly phe- nomena together with the appropriate omens for the ob- served phenomena.36 In the following report, dated to Apr. 15, 656 B.c., a solar eclipse was observed and re- ported, and “its interpretation (pisersu),” was included, i.e., what such an eclipse portended according to omens of the Enuma Anu Enlil series:

On the twenty-eighth day, at two and one-half ‘double- hou[rs’ of the day….] in the west […. ] it also cover[ed ….] two fingers towards [….] it made [an eclipse], the east wind [….] the north wind ble[w. This is its inter-

pretation]: If the day [becomes covered] with clouds on the north side: [famine for the king of Elam]. If the day be[comes covered] with clouds on the south side: [famine for the king of Akkad]. If the day is dark and r[ides] the north wind: [devouring by Nergal; herds will diminish]. If [there is an eclip]se in Nisan (I) on the twenty-eighth day: [the king of that land will fall ill but recover]; in his stead, a daughter of the king, [an entu-priestess, will die]; variant: in [that] ye[ar, there will be an attack of the enemy, and] the land will panic [….].37

As is clear in this report, an entry in an omen text records an event which-if or when such was “observed” (using this term to refer to all “observables” in the Babylonian framework)-warned of another event. The relationship between the observable phenomenon and its predicted event was expressed in the form “if x, then y” (summa x, y). In the astrological reports, the observa- tions are stated declaratively, as in the following: “This

night, the moon was surrounded by a halo, [and] Jupiter and Scorpius [stood] in [it],” which is then followed im-

mediately by a series of relevant omens, viz., “If the moon is surrounded by a halo, and Jupiter stands in it: the king of Akkad will be shut up. If the moon is sur- rounded by a halo, and Neberu stands in it: fall of cattle and wild animals.”38

However their connection came to be made, the sign and associated event, once recorded as an omen, were related

36 See Hunger, Astrological Reports to Assyrian Kings. 37 Hunger, ibid., no. 104: 1-rev. 1. 38 Hunger, ibid., no. 147: 1-6.

to one another in an invariant correlation.39 The omen, expressed by “if (or, ‘whenever’) x then y,” recorded the recurrent invariant association of one thing with another. Whether the omens reflect a conception of invariant se- quence, “if x occurs first then y occurs next,” or, alterna- tively, a conception of invariant coincidence, “whenever x occurs y also occurs simultaneously,” is difficult to deter- mine. Regardless of whether x and y were sequential or coincident, the important element is their subsequent re- currence, i.e., particular instances of x and y were assumed to recur thereafter together. This must be the case because the essence of omens is their predictive aspect. Y may be predicted on the basis of x because the two have been des- ignated as invariantly associated.

If the omens recorded in the scholarly lists are inter- preted as recurrent invariant associations, they cease to be observations as such, becoming rather abstractions resting upon empirical as well as theoretical foundations. And if it is true that Babylonian diviners determined that x and y were recurrently and invariantly associated, then those associations were products of something like in- ductive (or empirical) generalization, or were extrapola- tions from such empirical generalizations. On this basis the omen lists would represent the arrangement and codi- fication of a wealth of past experience in which the occur- rences making up the omen protases were “observed” to be associated with certain human events.40 The formu- lation “if/whenever x, then y” points to the necessity of recurrence.

General formulations “if/whenever x then y” are a spe- cies of causal statements. In the case of the omens, I view this causal connection between protasis and apodo- sis as a function merely of the relationship created by their invariant association. Such a causal relationship is like the statement “if/whenever the teapot whistles, the water is boiling,” without entertaining the question of what causes the teapot to whistle or the water to boil. The formulation itself gives the omens a lawlike appearance, especially when it is further evident that predictions derivable from the relation of x to y are the goal of the

inquiry into the set of x that bear predictive possibilities. If we regard the omens as lawlike abstract statements,

39 My discussion here is influenced by M. Wartofsky, Con- ceptual Foundations of Scientific Thought, ch. 11 (on causality), pp. 291-315, especially sub (a), 293-95.

40 This was termed “circumstantial association” by E. A.

Speiser, “Ancient Mesopotamia,” in The Idea of History in the

Ancient Near East, ed. R. C. Dentan, 4th ed. (New Haven: Yale Univ. Press, 1967), 61; and see also J. J. Finkelstein, “Mesopota- mian Historiography,” PAPS 107 (1963): 463ff.

566

ROCHBERG: Empiricism in Babylonian Omen Texts

however, it must be said that any actual instances of such “laws” are limited to the unique, and as such trivial, co- occurrence of just that particular x and y stated in the omen. Although a formal similarity may be seen between the omen statements and so-called scientific statements of empirical generalization, both of which enable predic- tion of phenomena, there is something fundamentally different in the extreme limitation of the domain of the omens as predictive statements.

Also to be taken into account is an aspect of the method whereby a sign (omen protasis) was paired up with an event (omen apodosis), which in fact undermines the idea that the link between the sign and its prediction necessarily involved an observation of the simultaneous or sequential occurrence of those two elements.41 It seems that associations between the sign and the predicted event could be purely linguistic and independent of any obser- vation, as, for example, in cases of paronomasia, where the apodosis plays upon the sound of a word in the pro- tasis. In Ziqiqu, a dream in which a man eats a raven (arbu) portends income (irbu) for that man. Similarly, a dream which presents fir wood (mihru) portends no rival (mdhiru) for the dreamer. Such a method appears to be a result of scribal imagination, presumably when the omens were being compiled and set down. In Summa alu an omen from tablet 15 (line 16) concerning spilled water links the appearance of the puddle-if it looks like “someone holding his heart”-to a prediction that the person “will experience heartache.” This indicates rather persuasively that while empiricism was a factor in the formation of omen protases it does not always come into play in the decision to associate certain apodoses with certain protases.

Another common method used in the correlation of protasis to apodosis was that of analogy. For example: “If someone’s firstborn is short: his house will be short (lived).” Simple analogies from form and appearance are found in the physiognomic omens, where, for example, a short face means a short life, and its opposite, a long face means a long life. A clear example from Enima Anu En- lil is the omen which correlates the “entering” (usurpa- tion) of the king’s throne by the crown prince to the “entering” of the plant Venus within the moon, an ex- pression used to describe the occultation of the planet by the lunar disk.

41 This has been discussed by I. Starr, The Rituals of the Di- viner (Malibu, Calif.: Undena Publications, 1983), 8-12, who jux- taposes “theoretical” methods of relating protasis to apodosis against “empirical.” In Starr’s analysis, the empirical method of divination is equivalent to Speiser’s “circumstantial association,” for which, see note 40.

Even where the connection between protasis and apo- dosis stems from an empirical consideration, symbolic as- sociation by analogy could be used to determine whether the sign was to have a favorable or unfavorable predic- tion. The lion, for example, was regarded as good, the dog as bad. Or, for example, the predictions for the first two omens of the series Summa alu are “if a city is sit- uated on a high place: the inhabitants of that city will not prosper,” and “if a city is situated on low ground: the in- habitants of that city will prosper,” suggesting that from a Mesopotamian point of view, high ground was bad, low ground good. This may, however, be an example of per contrariam prediction, like that in Ziqiqu tablet IX col.i: “if (a man) ascends to heaven: his days will be short,” and “if he descends to the netherworld: his days will be long.” Other forms of symbolic association can be identified, such as the simple distinction of good and bad with the common polarity of right and left. Something on the right indicates a good outcome while left indicates a bad, and bad is sometimes signified by the outcome being good for one’s enemy. Examples from the anomalous birth omens of Summa Izbu are illustrative: the presence on the mal- formed newborn of a right ear but not a left is good, while a left ear without a right is bad.42 A deformity to the right ear is bad, to the left ear is good.43 Two right ears are good whereas two left ears are bad.44 The same is applied to nostrils (having only a left nostril is bad, only a right is good45), hands and fingers,46 feet and toes.47 If an izbu has horns (tablets V and IX), something pertaining to the right horn is good, the left bad, in the same way as with other features of the body. Similarly, in lunar crescent omens from Enuma Anu Enlil, in which the moon has “horns,” if the moon’s right horn “pierces the sky” there is a good outcome,48 or if the moon’s right horn is long and the left is short, this is a good omen.49 The same prin- ciple is evident in the medical diagnostic omens. Symp- toms observed on the right or left will be interpreted as less severe or more severe, respectively: “if a man’s right ear is discolored, his disease will be severe but he will re- cover,” versus “if a man’s left ear is discolored, he is in a dangerous condition.”50

42 Leichty, The Omen Series Summa Izbu, Izbu III: 20-21.

43 Leichty, ibid., Izbu III: 5, 11, 14 and 16 and 9, 12, 15, and 17. 44 Leichty, ibid., Izbu III: 18 and 19. 45 Leichty, ibid., Izbu III: 30-31. 46 Leichty, ibid., Izbu III: 48-58. 47 Leichty, ibid., Izbu III: 58-62. 48 Hunger, Astrological Reports to Assyrian King, p. 35, no. 57: 5. 49 Hunger, ibid., p. 212, no. 373: 5; p. 284, no. 511: 5′. 50 R. Labat, Traite akkadien de diagnostics et pronostics

medicaux (Leiden: E. J. Brill, 1951), 68: 1-2.

567

Journal of the American Oriental Society 119.4 (1999)

These forms of empirical judgment, loaded with cul- tural content, belie the facile construction of the “empiri- cal” as “the objectively true.” As Sahlins puts it, “One

simply cannot posit another person’s judgments of ‘reality’ a priori, by means of common sense or common humanity, without taking the trouble of an ethnographic investiga- tion.”51 Assyriologists are at an obvious disadvantage here. But decisions regarding boundary lines between “reality” and “fantasy” in this case are not simply a matter of how relativistic we are about the validity of certain beliefs in the context of particular world-views.52 It is the way em-

piricism itself is construed that will determine whether

interpreters of Babylonian omens view the unobserved

phenomena incorporated within these texts as empiri- cally “valid” or simply absurd. Limiting the empirical to

sensory perception is as inadequate and distorting in relation to ancient sciences as it is to modern.

OBSERVABLES IN MESOPOTAMIAN DIVINATION

It has become commonplace to admit that theory and inference are embedded in observations and that, in fact, one observes a field of phenomena, not blindly, but for the purpose of discovering evidencefor something. The

conceptual context of an inquiry as a whole defines the interests and the problems for inquiry, including defining what is taken to be empirical for a particular inquiry. As discussed above, the most basic of all premises of Meso-

potamian divination, namely, that “signs” in nature are

produced by deities for the purpose of communicating with humans, suggests that ominous phenomena belonged to a conceptual framework representing the world as cre- ated and manipulated by deities. The expectations of what kinds of “signs” might occur resulted from this very inclusive concept of possible divinely produced phenom- ena. Therefore, dreams, lunar eclipses, puddles, disease

symptoms and malformed fetuses, could all and equiva- lently be “signs.” No category of phenomena had greater or lesser epistemological status, as all of the signs yielded “predictions” of what was in store for the observers. The

periodic nature of astronomical phenomena rendered this

category of signs amenable to prediction, which, from our point of view, distinguishes celestial divination from other forms of scholarly divination. It seems likely that the predictability of the signs themselves would be

advantageous to the practice of divination, giving the di-

51 Marshall Sahlins, How “Natives” Think, 162-63. 52 For an interesting discussion of this problem and its impli-

cations, see Dan Sperber, “Apparently Irrational Beliefs,” in Hollis and Lukes, Rationality and Relativism, 149-80.

viner advance knowledge of the occurrence of signs and

accordingly more time to prepare for their anticipated consequences.

In the absence of sources making explicit what the

Babylonian scribes’ criteria for empirical entities might be the schemata themselves indicate what was deemed “observable.” The schemata typically exhibit symmetries of direction (right and left, high and low, up and down, or north, south, east, and west), temporal relationships (beginning, middle, end), or other descriptive features and their opposites (bright and dull, light and dark, thick and thin), all, in principle, observable features of phe- nomena. One such scheme is found in the omens refer-

ring to the color of the object of interest, for example, the color of ants crossing the threshold of a house, the color of a lunar eclipse, the color of a sick person’s throat, or the color of a dog that urinates on a man. These items from the series Summa dlu, Enuma Anu Enlil, SA.GIG, and Izbu, respectively, are all organized similarly, i.e., white, black, red, green-yellow, and variegated. Simple numerical expansions are also characteristic, e.g., the

multiple births running from two to ten,53 or parhelia (mock suns) numbering from one to four,54 some of which seem to be beyond the range of the empirically verfiable. Such patterns are also characteristically found in impetrated omens, such as lecanomancy (oil divina-

tion): “if (the oil) becomes dark to the right/left,” or “if it dissolves to the right/left”; or libanomancy: “if the smoke, when you scatter it, rises to its right, but does not rise to its left,” or “if the smoke, when you scatter it, rises to its left but does not rise to its right.” The formal patterns into which phenomena could be organized in omen lists indicate that the goal was to determine whether a phe- nomenon appeared in a particular configuration (e.g., up, down, to the right, left, etc.) in accordance with such pat- terns. One may argue that it would only have been in terms of established ideas about the behavior of phenom- ena that any phenomenon would have been interpretable. Or, put another way, the kinds of phenomena under ob- servation were shaped by the scribes’ traditional ideas of what was deemed ominous. In and of themselves, strictly sensory impressions would have been meaningless. The

phenomena collected as omens in the series include both those potentially accessible to the senses and those that were not, but both were apparently considered (poten- tially) meaningful in conjunction with apodoses.

Within the diviners’ conceptual framework, what was “observable,” in the sense of being worthy of observation,

53 Leichty, The Omen Series Summa Izbu, Izbu VI: 46-58.

54 See the samgatu omens in ACh Suppl. 2 32.

568

ROCHBERG: Empiricism in Babylonian Omen Texts

was not limited to what could be seen. Of course, this statement requires that we make no distinction between omens as factual or fanciful and simply take the evidence of the omen lists “at face value.” We can imagine that, in the process of compiling the omen series and out of the scribes’ consideration of empirical symmetries, logical possibilities were raised of “phenomena” that had not been observed, but yet were conceivable and could ra- tionally be included within the omen series. Since the omens do not seem to recognize a categorical distinction between observed and unobserved phenomena, both be-

longed to the category of “observables,” in that the divin- ers watched for them all in order to predict the future.

CONCLUSION: THE CLASSIFICATION OF BABYLONIAN

DIVINATION AS SCIENCE

If the character of observables is indeed “the index of the whole framework of science” (Wartofsky), then the system of Babylonian divination as science encompassed a broad spectrum of the phenomenal world. Its interests spanned a diversity of natural and human phenomena. A classification of Babylonian omens on the basis of crite- ria of scientific truth currently accepted can only lead to a view that this system is a non-science, from its premises to its approach to the observable. Aspects of this ancient system, however, find many parallels with other historical sciences, and to the extent that the system had as its chief interest the phenomenal world, and indeed many natural phenomena, it shares something fundamental with other sciences, including modern.

Whether or not one classifies Babylonian divination as science seems to me best pursued, as Hesse put it, “in the spirit of the principle of no privilege.”55 As far as the omen protases referring to (for us) ontologically suspect

55 Mary Hesse, Revolutions and Reconstructions in the Philos-

ophy of Science (Bloomington: Indiana Univ. Press, 1980), 147.

or even impossible entities are concerned, these were ap- parently an important part of the scientific enterprise of

Mesopotamian divination, which suggests to me that we need to consider them if we are to engage not in a judg- ment, but an investigation of the nature of this system. The presumption of a commonsense empiricism defined in terms of categorical distinctions between the real and the fantastic and predicted on a notion of an objective re- ality corresponding to sense data does not take adequate account either of the evidence collected in omen protases or of the meaning of empiricism in any other scientific context. Even if we are able to explain away some phe- nomena, such as the lunar eclipses that are “observed” when they cannot occur, treating these not as astronomi- cal, but rather as meteorological eclipses, we are still left with the “observations” of ghosts, gods, and three-headed sheep. On the basis of the omen series themselves, we can only note that the status of such phenomena as observ- ables was as legitimate as those we recognize as empiri- cally true. To bracket these protases, designating them as “absurd,” and then going on to discuss the empirical core of Babylonian omen science, limits our ability to under- stand what phenomena the Babylonian scribes regarded as subjects for observation. It limits our ability to pene- trate, if it is at all possible, something about their per- ception of the world and what it meant within their own “conceptual framework.” If we do not regard contempo- rary physical science as “just a systematic exposure of the senses to the world,” but “a way of thinking about the world,”56 I wonder why we have not approached the sci- ence of Babylonian omens in the same way. It seems to me perfectly reasonable to apply the term empirical in a characterization of Babylonian divination, and to view all the subjects of the protases as “observables,” because these things belonged to the domain of their inquiry.

56 N. R. Hanson, Patterns of Discovery, 30.

569