1991 (B) - Changing Trends PDF
1991 (B) - Changing Trends PDF
1991 (B) - Changing Trends PDF
JENS HØYRUP
Revised Contribution to the Conference
«Contemporary Trends in the
Historiography of Science»,
Corfu, May 27 – June 1, 1991
FILOSOFI OG VIDENSKABSTEORI PÅ
ROSKILDE UNIVERSITETSCENTER
1991 Nr. 3
ISSN 0902-901X
To the memory of the founding fathers:
Otto Neugebauer, François Thureau-Dangin,
Solomon Gandz, and Kurt Vogel
TABLE OF CONTENTS
1
To some extent, of course, all history of science is ridden by the same dichotomy:
Is history of science to be done and judged as history, or does it belong within the
realm of the sciences. Logically, one would perhaps opt for the former answer;
according to the down-to-earth sociology of the pay-roll and the institutional
affiliations of most historians of science, however, most historians of science are
scientists.
Yet, even if the problem is shared by all history of science, it becomes more
outspoken when the philology and history of the period involved gives outsiders the
impression of an occult science, as it is the case of Assyriology.
I. THE HEROIC ERA, 1930 TO 1940
2
The rather few mathematical texts which we know from the late period, it is true,
were written by and for members of the astronomical environment, a context which
seems to have influenced the mathematical mode of thought.
3
A survey of publications from the period 1854 to 1929 with relevance for the
understanding of Mesopotamian mathematics will be found in Friberg 1982: 1-36.
-3-
Assyriology or in relation to Classical lore (»Plato’s number«, etc.).
Whatever one may think with hindsight of the principle, this turn was
obviously necessary to crack those complex texts which were now taken
up. This appears with great clarity if one compares Neugebauer’s
analysis (1929) of texts concerned with the partition of trapeziums with
the first attempts at translation of the same texts made by Carl Frank
(1928). Neugebauer’s prophetic conclusion (p. 79f) should be quoted:
Man darf wohl sagen, daß in den vorliegenden Texten ein gutes Stück
babylonischer Mathematik zutage liegt, unsere nur alzu dürftigen
Kenntnisse dieses Gebietes um wesentliche Züge zu bereichern. Ganz
abgesehen von der Verwendung von Dreiecks- und Trapezformel sehen
wir, daß komplizierte lineare Gleichungssysteme aufgestellt und gelöst
werden, daß man ganz systematisch Aufgaben quadratischen Charakters
stellt und zweifellos auch zu lösen verstand—und all dies mit einer
Rechentechnik, die der unseren völlig äquivalent ist. Bei einer solchen
Lage der Dinge bereits in altbabylonischer Zeit wird man in Hinkunft
auch die spätere Entwicklung mit anderen Augen anzusehen lernen
müssen.
In a postscript added in proof. Neugebauer acknowledges the
decisive role of H. S. Schuster in the interpretation of the text4. In the
following (second) fascicle of Quellen und Studien, Neugebauer (1930) and
Schuster (1930) each had an article dealing (in Neugebauer’s case among
other things) with Old Babylonian and Seleucid5 solutions of second-
degree problems, respectively. Already in the first fascicle, Neugebauer
and Struve (1929a) had investigated the Babylonian way of dealing with
circles, circular segments, and truncated cones.
To which extent these publications mark a watershed is revealed
by an slightly ironical remark made by Neugebauer in the preface to
4
According to what I was told in 1985 by Kurt Vogel, Schuster was in fact the first
to discover the Babylonian solution of second-degree problems.
5
The Old Babylonian period goes from 2000 B.C. to 1600 B.C. (the mathematical texts
seem to come from the later half of the period); the Seleucid period goes from 312
B.C. to 64 B.C.
-4-
MKT I (p. v):
Wenn ich sage, daß es von Anfang an meine Absicht gewesen wäre, eine
Edition aller erfaßbaren mathematischen Keilschrift-Texte zu veranstal-
ten, so soll das heißen, daß diese Arbeit zwar nie ihre Grundtendenz
verändert hat, um so mehr aber ihren Umfang. Das erste, bereits 1929
»Druckfertige« Manuskript erfaßte nur die ca. zwei Dutzend Tabellen-
texte aus Hilprechts Publikation BE 20,1, die drei Londoner Texte BM 85
194 und BM 85 210 aus CT IX, BM 15 285 aus RA 19 (Gadd), die beiden
Pariser Texte AO 6456, AO 6484 (TU 31 und 33), und schließlich die
sechs Texte aus Frank SKT. Das war nicht die hälfte der jetzigen Kapitel
I bis III und Kapitel V.
Apart from the corpus of tables, even this early list consists of texts
which had not been interpreted before—among which those which
Schuster, Struve and Neugebauer dealt with in Quellen und Studien in
1929-30. That is, already the supposedly »print-ready« manuscript from
1929 was a decisive leap forward—yet the impetus created by this initial
breakthrough made the leap look like a quite modest step in the
perspective of 1935.
It is characteristic that Schuster’s and Neugebauer’s articles (and a
number of others) were published in the newly founded Quellen und
Studien zur Geschichte der Mathematik, Astronomie und Physik, and not in
the Assyriological literature. Schuster could also point out (1930: 194) that
Thureau-Dangin, in the first publication from 1922 of the Seleucid tablet
under discussion, had only identified its contents as »opérations
arithmétiques«. However, Thureau-Dangin (who in fact had contributed
decisively since the mid-nineties to the knowledge of Mesopotamian
metrology and computational techniques) immediately took up the
challenge, and investigated texts of similar mathematical complexity in
the Revue d’Assyriologie. But still, and in spite of the immediate audience
of the Revue and a sometimes more precise reading of the texts, the
approach was the same: Babylonian mathematics was related by
Thureau-Dangin no less than by Neugebauer to categories of later
mathematics.
-5-
Why was the breakthrough produced at Neugebauer’s seminar at
Göttingen University and not by a competent philologist like Thureau-
Dangin, whose interest in the matter was conspicuous, and whose
mathematical competence turned out in the 1930s to be fullyy sufficient?
Paradoxically, the answer to this question has to do with that very
complexity of cuneiform writing which would make one expect the
philologist to be best fit for the task.
Firstly, most cuneiform signs are plurivalent. They may carry one
or several logographic meanings (not necessarily semantically related),
to which comes one or more groups of phonologically related syllabic
readings. Specific text types have their particular usages, which reduce
the ambiguity—but only when the characteristic usages have been
discovered.
To this we can add, secondly, the terminology itself. Like all tech-
nical terminologies, Babylonian mathematical terminology was ultimately
derived from daily language—but often technical meanings cannot be
guessed from general meanings, even when these are known. They have
to be derived from the mathematical procedure used—which itself is
hard to get at, if one is does not understand the terminology6.
Without having observed the processes directly, one may surmise
that only scholars with thorough mathematical training (and, certainly,
with a level of cuneiform competence approaching that of many
Assyriologists) would possess sufficient creative phantasy to crack the
codes from the numbers in the tablets (numbers mostly written in a
sexagesimal place value system without indication of absolute place and
hence ambiguous) and a rudimentary understanding of the connecting
words. Once this step was taken, Assyriologists with less exhaustive
mathematical competence would be able to join in and improve readings
by philological means.
This was what happened, and the 1930s were dominated by a
passionate though correct and very polite race between Neugebauer and
Thureau-Dangin, whose philological level was supreme, whose interest
6
Further explanation and exemplification in Høyrup 1990: 43-45.
-6-
in mathematics was longstanding, and whose understanding of mathe-
matics and knowledge of its early history proved entirely adequate7. A
number of articles from Neugebauer’s hand culminated in the publication
of the Mathematische Keilschrift-Texte I-III in 1935-1937, while Thureau-
Dangin’s decisive achievements were published in article form, not least
in the Revue d’Assyriologie, which he directed together with Vincent
Scheil. His Textes Mathématiques Babyloniens from 1938 was presented as
an attempt to »mettre des documents à la disposition des historiens de
la pensée mathématique« (TMB, xl) at a more accessible price than the
MKT (von Soden 1939: 144). For this reason, the philologically »in-
convenient« method of transcribing Sumerograms into Akkadian was
7
It may be accurate that the two »hated each other«, as I was told by
Olaf Schmidt, who was close the Neugebauer in the later 1930s—while
Bruins’ statement (1984: 107) that Thureau-Dangin considered the MKT
as »une mer d’erreurs« probably misrepresents Thureau-Dangin as much
as his following remarks on the latter’s intention by publishing the Textes
Mathématiques Babyloniens (compare Bruins 1984: 107 with TMB, xl, the
final paragraph). But even if their mutual feelings may have been
acrimonious, the fruitful outcome of the process shows the function of
scholarly mores at their best. Competition never prevented any of them
from giving advice or learning from criticism, nor from emphasizing the
merits of the other’s publications. It would be difficult to find more
indisputable corroboration of Robert Merton’s theses (1942) concerning
the function of the »institutional imperatives of science«.
One possible exception to this optimistic verdict should perhaps be
mentioned, even though I only know about it from rumours and have
not been able to verify it: Thureau-Dangin is claimed to have taken care
that Neugebauer should not get access to the extremely important
mathematical texts from Susa, which had been found already in 1933,
and which were only published in problematic form in 1961 as TMS (cf.
below). Against the rumour speaks »die Großzügigkeit, mit der mir der
Text [AO 8862] zugänglich gemacht und die Publikationserlaubnis
gegeben wurde« by Thureau-Dangin (Neugebauer 1932a:3). Such
generosity is no matter of course among Assyriologists.
-7-
adopted—for precise philological purposes, the reader was referred to
original publications (at times Thureau-Dangin’s own, at times those in
MKT).
Important contributions to the field were also published by Kurt
Vogel and Solomon Gandz, both of whom had their main interests
elsewhere, and both of whom brought their distinctive perspective. In
spite of this, however, and in spite of the different starting points of
Neugebauer and his collaborators on one hand and Thureau-Dangin on
the other, a coherent approach to the history of Mesopotamian mathe-
matics emerged and came to monitor the way both general historians of
mathematics and Assyriologists saw the matter for decennia.
First of all, the field was seen specifically as "Babylonian mathema-
tics«. What little was known about practical computation in the Sumerian
third millennium disappeared from view (presumably as »not really
mathematics«8). »Genuinely« mathematical texts were only known from
the Old Babylonian and the Seleucid period (with one or possibly a few
exceptions, which might be a bit younger than Old Babylonian). None
the less, »Babylonian mathematics« stood forward as one immutable
entity. Schuster (1930: 194) had found it important to point out that his
investigation of a Seleucid tablet gave insight »in die mathematischen
Kenntnisse [...], die noch zur Griechenzeit in Babylonien existierten« and
demonstrated »Kontinuität der orientalischen Tradition von sumerischer
Zeit bis weit in den Hellenismus«, while on the other pointing to a
conspicuous change in mathematical terminology taking place between
the Old Babylonian and the Seleucid period (originally distinct operations
8
In MKT I, a number of presumed Ur III (21st century B.C.) tables of reciprocals had
been listed. Still, the mathematical substance of these was evidently soon exhausted,
and as long as mathematical procedures and techniques were asked for, only the late,
multi-place tables were subjected to further investigation.
The above statement does not mean that nobody looked at older mathematical
techniques. For one, F.-M. Allotte de la Fuye, who had produced important
publications on such subjects for decades, continued to do so. But his text material
and his results were not understood as belonging to the history of (Babylonian)
mathematics.
-8-
losing their proper designation and thus—we may add—perhaps their
proper identity)9. At the end of the decade, the former conclusion had
become a trivial matter of fact—maybe because of its agreement with the
stereotype of an »immutable Orient«. The latter observation had largely
come to be neglected—it regarded only the »history of terminology«,
seemingly a purely philological and somewhat pedantic concern.
The separation of the history of mathematics from the history of
terminology is a particular instance of another characteristic of the
resulting ruling approach: The separation of philology and mathematics,
and the exclusive reading of the sources for their mathematical content.
This had not been Neugebauer’s intention. To the contrary, he had
claimed (1932: 222) that the transcription of Sumerograms into Akkadian
destroyed the »fundamentale sachliche Rolle der Ideogramme: daß sie nämlich
volkommen wie mathematische Symbole wirken«. But precisely his
emphasis on this aspect of the relation between terminology and
mathematical thought was taken as justification for a translation into
modern mathematical symbols, and thus as a reading as modern
mathematics—in particular when the statement was read by others who
knew neither Sumerian nor Akkadian but trusted the translations and the
mathematical commentaries.
Rather unreflecting10 reading, if not as then through the categories of
9
Neugebauer (1932a: 6) had been even more cautious; he presented the existence of
a Sumerian prehistory to Old Babylonian advanced algebra as a hypothesis which
was close at hand but unsupported by positive evidence. An important part of the
same article is also dedicated to terminological differences and changes, and the
statement that »sich das inhaltliche Niveau [from c. 1700 to c. 300 B.C.] nicht sehr
erheblich verändert hat« is characterized as »selbstverständlich nur eine Aussage ’in
erster Näherung’«.
10
A distinction between »unreflecting« and »critical« reading through the categories
of more familiar mathematics is important. Explanation always has to represent the
categories which are to be explained by others which can be supposed to be known,
and which are necessarily different. »Unreflecting« translation of categories is »one-to-
one«, while »critical« translation will be »network-to-network«. In itself there is
nothing wrong in describing a problem »I have added the measuring number of the
side and the area of a square, and the result was 110« as »an equation«; this is in fact
-9-
more recent mathematics was in fact what characterized the main
workers in the 1930s albeit with important shades, and what distinguish-
es no less the average picture of Babylonian mathematics which emerged
and was accepted.
It is often claimed that Neugebauer was the most modernizing of
all. This is more than a half mistake, a mistake which is due to careless
reading of MKT. In commentaries to the texts, it is true, unrestricted use
is made of symbolic algebra; but the aim is to show that the computa-
tional procedures employed by the Babylonian calculators are correct (or,
at rare occasions, mistaken). The mathematical commentary is not
claimed to map the ideas or methods of the Babylonians11. When making
general statements, Neugebauer would normally take care to put the
terms Algebra and algebraisch in quotes (so 1932a: 24, and MKT III, 79).
His idea was that Babylonian mathematics was numerical, springing from
the use of sexagesimal computation and fertilized by the advantage
offered by ideographic writing (MKT III, 79). He warned against
overrating Babylonian mathematics, which he felt contained »an keiner
Stelle etwas, was als unerwartete Glanzleistung angesehen werden
müßte«, when only »die ungeheure Schwierigkeit und Langsamkeit der
Entwicklung der allereinfachsten mathematischen Grundbegriffe, vor allem
einer wirklichen Rechentechnik« had been overcome (ibid, 80).
Regarded closely, Neugebauer’s »modernization« in the MKT thus
reduces to the application of numerical conceptualization. Even though his
restorations of damaged texts shows him to have been very sensitive to
the terminological distinction between different »additive« and different
»multiplicative« operations (cf. below), he understood the operations as
the closest we can get in terms of familiar notions. But an explanation which stops at
this point, instead of discussing the particular character of the »equation«, the way it
differs from and the way it is similar to a modern equation in x and y, is no
explanation but a replacement of an ancient by a modern conceptual structure.
11
A few cases can be found where Neugebauer is mislead himself and takes the
justification to be the only possible interpretation. So in his (1932a: 21f), a commen-
tary to problem #3 of the tablet AO 8862. In the discussion of the same problem in
MKT (I, 120), however, the mistake is eliminated.
- 10 -
addition and multiplication of numbers.
This does not apply, it is true, to Neugebauer’s more popular
lectures on Vorgriechische Mathematik12. Here it is stated, e.g., that
Babylonian mathematics »ihrem ganzen Niveau nach eine algebraische
Stufe erreicht hat, die erst zu Beginn der neueren Geschichte wieder
errungen worden ist« (Neugebauer 1934: 172). One of the texts which he
had discussed very cautiously in his article from 1929 is also presented
in a way which makes it impossible to distinguish justification through
from interpretation as algebra, and it is stated quite bluntly that »die
Formulierung ist zwar hier noch eine geometrische, aber die Ausrech-
nung selbst ist nichts als ein rein algebraisches Bestimmen von Unbe-
kannten auf Grund gewisser gegebener Relationen« (1934: 179).
On this point, Thureau-Dangin’s stance was identical with the
attitude which Neugebauer had expressed in the context of popular-
ization. So, in an article on "L’Équation du deuxième degré dans la
mathématique babylonienne" he tells his interpretation through symbolic
algebra to be a reconstruction of demonstrations not given in the text but
evidently lying behind (1936: 28). In his introduction to the TMB he also
speaks about algebra without hesitation or qualification, while suggesting
that the kind of algebra involved is similar to the one later taught by al-
Khwārizmı̄—a position which is made more explicit in an article on
"L’Origine de l’algèbre" (1940: 301).
Thureau-Dangin’s concept of »algebra« was completely numerical.
Second-degree problems were formulated so as to deal with plane
figures, he claimed (1940: 302), simply because
une figure plane, telle notamment qu’un triangle, un carré ou un
rectangle, donne facilement lieu à une équation du second degré, mais
les problèmes qu’en tirent les Babyloniens ne relèvent pas plus de
l’algèbre géométriques que, par exemple, les problèmes indéterminés que
traite Diophante dans son livre VI et dont il emprunte les éléments au
triangle rectangle. Il s’agit dans les deux cas de problèmes purement
12
Another exception is his article on »geometric algebra« (1936), to which I return
below.
- 11 -
numériques.
So far, Thureau-Dangin’s position was thus more or less shared
either by Neugebauer the high-level popularizer or by Neugebauer the
meticulous scholar. A point where they differ is in Thureau-Dangin’s
repeated reference to the »method of false position« (1938; 1940: 316f),
which Neugebauer seems never to have mentioned.
Like the comparison with al-Khwārizmian rhetorical algebra, this
is an illustrative instance of Thureau-Dangin’s tendency to read
Babylonian mathematics through the concepts and categories of other
pre-modern mathematical cultures—a tendency which he shares with
both Gandz and Vogel, while Neugebauer followed the maxim Hypo-
theses non fingo in this question as closely as possible.
Gandz contributed to the field in various ways, not least through
his competence as a Hebrew scholar. His contribution to the profile of the
field, however, was a monographic article on "The Origin and Develop-
ment of the Quadratic Equations in Babylonian, Greek, and Early Arabic
Algebra" (1937). As suggested by its title, the article claimed that both
Greek and al-Khwārizmian algebra descended from the corresponding
Babylonian discipline. »Greek algebra« not only meant the algebra of
Diophantos (more precisely, that modest part of Diophantos’ Arithmetic
which is concerned with determinate second-degree problems) but also
the »geometric algebra« of Elements II—an issue to which I shall return
below.
Making use of earlier work by Vogel, Gandz introduced a
comparative classification, which has remained influential, not least
because it brought some order to the vacillating identification of
Babylonian problems with modern symbolic equations. At the same
occasion, however, it disseminated the belief that this classification as
well as its formal expression corresponded directly to what was found
in the Babylonian (and Greek and Arabic) texts13.
13
The problematic nature of this belief can be illustrated on Euclidean material. Even
if we accept the theses that, e.g., Elements II, prop. 5 should be read, firstly, as algebra,
and, secondly, as an equation and not as an algebraic identity, how do we know that
- 12 -
Algebra was thus established as a discipline which the Babylonians
had created, with the Medieval algebra type as the example through
which the term was primarily understood. Mathematical fields and
concepts without pre-Renaissance antecedents were mentioned occasion-
ally as a characterization of one or the other Babylonian text but with
much greater caution. »Logarithms« turn up in MKT I, 362, in the
statement that a reverse compound interest problem »der sache nach«
asks for a solution which »mit n=log2(K/a) irgendwie äquivalent sein muß«;
but it is argued very clearly in the following pages that this does not
correspond to the Babylonian procedure; the same rejection of the
logarithmic interpretation is given in (Neugebauer 1934: 198). Nor is
»theory of numbers« postulated directly,—only, reticently, as »eine Art
elementarer Zahlentheorie«, which is then referred to »Pythagorean«
arithmetic (MKT III, 80).
A substantial share of the Babylonian mathematical problem texts
are concerned with practical problems involving metrological conver-
sions, norms for work etc. and with the determination of volumes.
Considerable effort was devoted not only by Thureau-Dangin (who had
been interested in such matters since the beginning of his career in the
late 19th century) but also by Neugebauer to the analysis of the precise
technical meaning and the techniques of these texts.
All in all, the reading of the Babylonian texts »as mathematics« was
thus no uncritical identification of Babylonian mathematics with
»immature modern mathematics«. Yet the tendency to concentrate on the
mathematics of the texts, necessary as it probably was as a »first
the statement »if a straight line be cut in equal and unequal segments, the rectangle
contained by the unequal segments of the whole together with the square on the
straight line between the points of section is equal to the square on the half« (transl.
Heath 1926: I, 382) is to be translated into x+y=a, xy=b, and not into ax-x2=b? Indeed
Heath, in his commentary, gives the latter equation as his main interpretation and the
former only in passing. In certain Babylonian problems, the situation is definitely no
better.
- 13 -
approximation« if the code should be cracked14, invited to extrapolation:
From unreflecting characterization by means of modern mathematical
concepts to interpretation in terms of these. This is what came to
characterize the following period. Before we leave the thirties we shall,
however, look closer at an important spin-off of the discovery of
Babylonian mathematics, viz the idea that Greek »geometrical algebra«
was nothing but Babylonian numerical algebra in geometrical dress
(necessitated by the discovery of irrationals).
The idea that Elements II should be understood as algebra was not
new. It had been formulated explicitly by Zeuthen (1886: 5ff), in his
interpretation of Apollonios’ Conics and has antecedents far back15. On
this background, the discovery of Old Babylonian second-degree
14
And also, it should be remembered, by the lack of obvious connections between the
sophisticated mathematical texts and what else was known about Babylonian culture:
»Man darf [...] nicht vergessen, daß wir über die ganze Stellung der babylonischen
Mathematik im Rahmen der Gesamtkultur praktisch noch gar nichts wissen«
(Neugebauer 1934: 204).
It was understood that the texts which we possess are training problems,
constructed backwards from the solution, and thus school exercises. But texts
elucidating the structure, curriculum and ideology of the Babylonian school have only
been published since the late 1940s. In 1934 Neugebauer was fully right in
maintaining that only a negative conclusion could be attained: Babylonian mathe-
matics was not a child of astronomy and astrology, and not born from religious
concerns.
Even the relation between »practical« mathematical problems and real computation-
al practice was difficult to specify at a time when tables of practical (»igi-gub«)
constants were unknown (the first were to be published in MCT).
15
According to al-Nayrı̄zı̄’s commentary to the Elements (ed., transl. Besthorn &
Heiberg 1893: II,i, 27), already Hero had begun proving the theorems of book II »by
means of analysis«, which is at the very least a step in the direction toward an
algebraic interpretation (depending, of course, of our definition of that term, but in
agreement with Viète’s understanding of his own accomplishment as a redemtion of
analysis).
In the 13th century, Jordanus de Nemore modelled his whole reconstruction of
Arabic algebra after Elements II and the corresponding propositions of the Data (cf.
Høyrup 1988: 332-36). In his case, the idea that Elements II was a metatheoretically
more satisfactory version of al-jabr is thus indubitable.
- 14 -
»algebra« invited the evident and still open-ended conclusion that »man
in Hinkunft auch die spätere Entwicklung mit anderen Augen anzusehen
lernen [muß]« (Neugebauer 1929: 80).
How these other eyes should look at things was later specified by
Neugebauer (1936: 250) as follows, after he had presented Zeuthen’s
concept with approval:
Die Antwort auf [...] die Frage nach der geschichtlichen Ursache der
Grundaufgabe der gesamten geometrischen Algebra [i.e., the application
of an area with deficiency or excess], kann man heute vollständig geben:
sie liegt einerseits in der aus der Entwicklung der irrationalen Größen
folgenden Forderung der Griechen, der Mathematik ihre Allgemeingül-
tigkeit zu sichern durch Übergang vom Bereich der rationalen Zahlen
zum Bereich der allgemeinen Größenverhältnisse, andererseits in der
daraus resultierenden Notwendigkeit, auch die Ergebnisse der vorgriechi-
schen »algebraischen« Algebra zu übersetzen.
Hat man das Problem in dieser Weise formuliert, so ist alles
Weitere vollständig trivial und liefert den glatten Anschluß der babyloni-
schen Algebra an die Formulierungen bei Euklid.
—not least, thus Neugebauer in the following passage, because Babylon-
ian »’algebraic’ (i.e., numerical) algebra« was »translated« into geometry
already in the Babylonian sources: E.g., the problem xy=a, x+y=b into a
problem concerned with a rectangle with given area and given sum of
length and width, i.e., into the simplest version of »application with
deficiency«.
As we have seen above, the thesis was taken over as trivially
unproblematic by Gandz (1937). This is how its further career began.
- 15 -
II. THE TRIUMPH OF TRANSLATIONS, 1940 TO 1975
The heroic epoch can be taken to have ended around the beginning
of the Second World War. Admittedly, another important collection of
texts, some of them unprecedented (the igi-gub-tablets and the tablet
Plimpton 322 with its »Pythagorean triplets«), was published by
Neugebauer and Sachs in 1945 as Mathematical Cuneiform Texts. Yet from
around 1940 »everybody« knew that Babylonian mathematics was as
described by Neugebauer and Thureau-Dangin. With few exceptions,
Assyriologists finding a tablet containing too many numbers in place
value notation would put it aside as »something for Neugebauer«, while
mathematicians and general historians of mathematics would know all
they wanted from the translations contained in MKT (TMB only rarely
except in Francophonic areas) or, all too often, from the few examples
rendered in German, English or symbolic translation in the secondary
literature16.
Since the secondary literature was more prone than the (generally
cautiously formulated) text editions to subscribe to modernizing readings
of the texts and would neglect all references to the terminology and its
development, it was soon conventional wisdom that »Babylonian
mathematics« could be treated as one thing from Old Babylonian through
16
In early years not least Neugebauer 1934 (reprinted 1969) and Gandz 1937; later
also Neugebauer 1969 (1st ed. 1952) and van der Waerden 1962 (1st Dutch ed. 1950,
with English transl. 1954). Vogel 1959 and Vajman 1961 have (undeservedly) been
much less influential, in Vajman’s case because of the language in which the book
was written, in Vogel’s perhaps because its appearance in a series of high-school
textbooks veiled its qualities.
- 16 -
Seleucid times17; that Babylonian mathematics could be adequately
described in terms of symbolic algebra and other recent mathematical
techniques18; and finally that Greek »geometric algebra« was really a
geometricized algebra derived from the Babylonian prototype.
Van der Waerden’s Science Awakening (1962), probably the most
influential work of all, was explicitly intended (among other things) »to
explain clearly how Thales and Pythagoras took their start from Babylonian
mathematics but gave it a very different, a specifically Greek character« (p. 5).
Van der Waerden’s presentation of »Babylonian algebra« (pp. 63-75)
is still undogmatic as far as modernization is concerned. Admittedly,
along with a number of moderately straightened translations of texts it
brings translations into symbolic algebra. At the same time it suggests,
however, that the thought process behind a particular solution »is
expressed better by [a certain intuitive argument ascribed to a hypotheti-
cal ’elementary school teacher’] than by the elaborate algebraic trans-
formations, which Neugebauer gives« (p. 67); it also conjectures (pp. 71f)
that fundamental algebraic identities »like (a-b)(a+b) = a2-b2« can have
been found by means of geometric diagrams, while still maintaining that
we must guard against being led astray by the geometric terminology.
The thought processes of the Babylonians were chiefly algebraic. It is
true that they illustrated unknown numbers by means of lines and areas,
but they always remained numbers. This is shown at once in the first
example [of the preceding], in which the area xy and the segment x-y are
17
Neugebauer (MKT III, 5 n.20) had explained his choice of what he considered as
»sachlich adäquaten« instead of literal translations by the observation that »wer
terminologiegeschichtliche Studien an Hand einer Übersetzung machen will, dem ist
doch nicht zu helfen«. If this was read at all, then only as a statement that »termino-
logiegeschichtliche Studien« were irrelevant to the study of the history of mathematics,
and that translations could thus safely be relied upon.
18
Even though a few writers have maintained, basing their understanding upon one
or two simple examples borrowed from the secondary literature, that Babylonian
mathematics contained nothing but empirically established numerical schemes.
Familiarity with only a modestly broader sample of translations taken from MKT or
TMB would have prevented the mistake.
- 17 -
calmly added, geometrically nonsensical.
The tendency to replace the Babylonian texts by modern mathema-
tics becomes more outspoken and much less reflecting if we go to general
histories of mathematics19. Here, furthermore, practically oriented
mathematics disappeared from view apart from rudiments: interest in the
(symbolically expressed) formulae for areas and volumes, and succinct
statements that mathematics was used for this or that practical purpose.
The sexagesimal place value system is a recurrent pièce de résistance, but
the restricted role of this system and the existence of other, unambiguous
notations used for practical purposes is bypassed in silence.
An early example is Hofmann’s Geschichte der Mathematik (1953).
According to this book, slopes are measured by their »Rücksprung
(cotg)«, while no word is wasted on that absence of a general notion of
angle which had been pointed out time and again by the original
workers. Equations are presented in symbols without a word as to their
original, verbal formulation, and evidence from all ages is presented
without distinction. Neugebauer’s idea of the function of ideograms as
operators is taken over, but now referred to those practical problems
where ideograms can surely be maintained to serve as mere technical
abbreviations. Perhaps because of the formalization of which the
secondary literature makes use, perhaps because mathematics is thought
of as identical with formalization in the century of Hilbert and Bourbaki,
it is finally stated that the rich material gives us an interesting insight »in
die formale Höhe der babylonischen Mathematik« (emphasis added), while
workers closer to the original texts had rather been impressed by the
contentual level which was reached in spite of the absence of formaliza-
19
The same strengthening of the tendency can be noticed in a large article on "Die
Algebra der Babylonier" (Goetsch 1968), which builds exclusively on translations and,
even more, on the mathematical commentaries of original editions (see, e.g., p. 118),
and whose only reserve against symbols arises when the author does not understand
that Neugebauer’s justifications should not automatically be understood as interpreta-
tions (p. 103). The form of the article is illustrative of the general expectation as to
how the history of Babylonian mathematics was to be dealt with.
- 18 -
tion20.
Carl Boyer’s History of Mathematics (1968) is less concise, more
factually precise and much richer in details and examples. Yet al-
Khwārizmı̄’s classification of mixed quadratic equations, translated into
symbols, is stated to be the classification used »in ancient and Medieval
times«, and we are informed that »all three types are found in Old
Babylonian texts« (p. 34f), and that the type »x2+q=px« »appears
frequently in problem texts, where it is treated as equivalent to the
simultaneous system x+y=p, xy=q«, without any attempt being made to
explain that this is a symbolic interpretation of something different (nor
of course that the texts in question contain no hint of the idea of one
form of the problem being equivalent to another formulation). Elsewhere,
the tablet Plimpton 322 (the table based on Pythagorean triplets) is told
to have »deep mathematical significance in the theory of numbers« (p.
37).
There is no reason to go on with detailed exemplifications, even
though more analysis of other works would enrich the picture with
shades. Eves (1969: 31), e.g., explains that Babylonian algebra (which is
taken for granted) is a »rhetorical, or prose, algebra«. Kline (1972: 8f), in
an otherwise reasonable exposition, manages to explain that the problem
of finding »a number which, added to its reciprocal, yields a given
number« is »a fundamental problem of the older Babylonian algebra«,
and that the problem of finding two numbers with given sum and
product was »reduced« to this form (the original text of the tablet YBC
6967 shows that it is rather the opposite reduction which takes place, the
number and its reciprocal being understood as the sides of a rectangle,
the area of which is explicitly spoken of as such—see MCT, 129, but not
the translation). Etc.
That the authors of general histories tend to believe in the
secondary literature written by specialists (and to overemphasize that
modernizing aspect of the specialists’ exposition with which they are
20
It is immaterial for the present purpose that the presentation is also ridden by
actual mistakes.
- 19 -
familiar at the cost of qualifying remarks) should not cause bewilder-
ment. It is more amazing that the same trend can be found in the
specialist literature itself, and that Assyriologists took over the moderniz-
ing interpretation.
The first, and perhaps the most amazing instance is Neugebauer’s
and Sachs’s Mathematical Cuneiform Texts from 1945. Evidently, the
superficiality encountered in the expositions in general histories is as far
removed from this careful volume as at all possible. The relations
between practical mathematical problems and technical practice are
carefully investigated; and far from pretending that everything Babylon-
ian looks like anything else irrespective of chronology, the volume
contains a chapter by Albrecht Goetze where linguistic differentiation is
used to distinguish localities and time of origin within the Old Babylon-
ian epoch.
When it comes to mathematics, however, the tendency to interpret
unreflectingly through modern concepts is indubitable. Plimpton 322 is
taken, not precisely as an expression of »eine Art elementarer Zahlen-
theorie«, as the reticent words of MKT III are quoted, but as »a text of
purely number theoretical character«, and as »investigation of the
fundamental laws of numbers themselves« (p. 41).
As it will be remembered, the relevance of the concept of loga-
rithms had been rejected both in Neugebauer (1934) and in MKT. But as
a commentary to an inversion of the table of »powers« (rather, »repeated
products«, since this is what is stated in the »direct« table) it is said in
MCT (p. 35) that
We now have an Old Babylonian tablet which answers the question: to
what power must a certain number a be raised in order to yield a given
number? This problem is identical with finding the logarithm to the base
a of a given number«.
In the end of the discussion it is then stated that
In a comparison with our concept of logarithm, the only missing element
is the selection of a common base and the tabulation for constant
intervals, which would be needed if the tables were to be used for
- 20 -
practical computations in general. It is accordingly clear that the Old-
Babylonian mathematicians were very close to an important discovery
but failed to take the final, essential step.
Forgotten is, firstly, that the »tabulation for constant intervals« is not just
one element of a modern table of logarithms but the only element—no
table of decadic logarithms bothers to tell the logarithm of 10, 100, etc.,
which on the other hand is the only thing listed in the Babylonian table,
merely with base 2 instead of base 10. Forgotten is, secondly, a question
which would probably have been asked by Neugebauer 10 years earlier:
Was »practical computation in general« what the author of the table was
after? Forgotten, finally, the question whether logarithms are really
transcendentally important or only important in the context of Early
Modern to contemporary mathematical theory and computational
techniques.
When Neugebauer, the paragon of translators, and Sachs, the
»scientific humanist«, are thus bound by the spell of their own conceptu-
al translation, it is only to be expected that modernizing interpretations
were as a rule also accepted by Assyriologist in general when dealing (on
rare occasions) with mathematical texts21.
21
I shall restrict myself to a single reference: the unreserved use of symbolic algebra
in Gundlach & von Soden 1963. The reason to pick out precisely this thoughtful
publication is that von Soden was almost the only scholar at the time to point out the
dangers inherent in unreflective modernization—thus in a slightly later publication
on »language, thought and concept formation in the Ancient Orient«: »Die Mathe-
matikhistoriker setzen die babylonischen Ausrechnungen m.E. vorschnell in uns
gewohnt Gleichungen, noch dazu oft mit allgemeinen Zahlen, um und werden
dadurch der Andersartigkeit des mathematischen Denkens im alten Orient nur
unzureichend gerecht« (von Soden 1974: 28). In spite of the authors’ own doubts
concerning the procedure, there was no other way to present Babylonian mathematics
at the time.
Apart from the Susa texts (on which below) and a smaller bunch of tablets from
Tell Harmal (Baqir 1950, 1950a, 1951; Goetze 1951), only very few new texts were
published between 1945 and 1970. That they were treated according to the tradition
which had been established by Neugebauer, Sachs and Thureau-Dangin goes more
or less by itself, and calls for no supplementary commentary in the present context.
- 21 -
So far, everything seems to agree with an almost Kuhnian scheme:
After an initial phase where methodological and philosophical problems
are amply discussed follows another where scholars do »business as
usual«, convinced by the success of the first generation that it was
right—more firmly convinced, indeed, than this generation had dared to
believe itself—boiling the methodological message down to a simplified
textbook version while refining and extending actual results. Eventually,
even the founding fathers become convinced22. One aspect of the
process, however, falls outside this general logic of the development of
knowledge though under the more general heading menschliches,
allzumenschliches. Notwithstanding the principle nihil nisi bene it has to be
mentioned, since much of what happened to the field would else be
unexplainable.
As mentioned above, Neugebauer was not given access to the
mathematical tablets from Old Babylonian Susa, for reasons which I have
not traced. Instead, the task was entrusted to the historian of mathema-
tics Evert M. Bruins in collaboration with the Assyriologist Marguerite
Rutten, who took care of copying and—so it appears—was main
responsible for transliterations. Bruins was responsible for the mathemati-
cal commentary and apparently for most of the Akkadian transcription
from Sumerograms and for the translation into French23. The outcome
of this collaboration appeared in 1961 as Textes Mathématiques de Suse
(TMS), after ten years where Bruins had informed about one or the other
tablet in various articles.
The tablets are of extreme importance for the understanding of the
higher levels of Babylonian mathematics, in particular the »algebra«.
22
As I discussed the process with my colleague Michel Olsen he commented that this
was exactly what also happened within the field of structuralist text analysis.
23
»Apparently«, since the preface only states that the translation (which seems to
encompass everything between copying and mathematical commentary) was made
in cooperation (p. xi). It is obvious, however, that much in the translation into French
and even in the transcription into Akkadian has been derived backwards from the
mathematical commentary; the transliteration, on the other hand, is relatively free of
this backward influence.
- 22 -
They are difficult, and at times very different from anything known
beforehand. Bruins has thus had an indubitably difficult task, and he
should be praised for finding sometimes ingenious interpretations. On
other occasions, however, his transcriptions into Akkadian contradict the
most elementary rules of the Akkadian vocabulary and grammar; he
overlooks that two consecutive problems on a tablet are different and
spins a long story out the existence of two different solutions to what he
believes to be one problem; in a standard construction he takes an
Akkadian possessive particle -šu, »its«, for a Sumerian šu, »hand«; etc.24.
According to normal rules and experience, others should have
continued work on the texts, confirming sound conclusions and
eradicating obvious mistakes. This never happened; instead, the
interpretations remained almost fully unchallenged until a few years ago,
and the fanciful mathematical commentary was accepted by eminent
scholars without specialized knowledge of Akkadian, and even by many
Assyriologists, who may have been as scared by the mathematics as
other scholars by the cuneiform script. Both groups, of course, were
entitled to believe that everything was sound as long as those who
should have done so did not object.
The reason that almost nobody objected is obvious from what
happened to the sole scholar who tried to do so. Wolfram von Soden
made a review (1964), which was precise but quite gentle in tone. In 1963
Karl-Bernhard Gundlach and he had also dared to disagree with another
one of Bruins’s interpretations. As a result, von Soden was submitted to
almost 30 years of defamation, expressed in a language and with a self-
assurance which nobody is expected to use in scholarly discourse unless
his cause is impeccably sound.
Neugebauer and others who had dared to disagree fared no
better25, and appear to have decided to ignore the pest. This might have
24
For documentation, I shall only refer to (Høyrup 1990: 299-302, 320-327). The
-šu/šu-mistake, not mentioned there, is TMS, p. 52.
25
Bruins could never agree with himself whether it was Neugebauer or Derek Price
who should have been caught in the Plimpton collection trying to break off a piece
- 23 -
been a sensible strategy (and was indubitably sound for their mental
health), if Bruins had not had free access to publication channels—princi-
pally in the journal Janus, of which he was the main editor, and whose
deficit he paid. As things were, however, his verbal violence and his
assurance were liable to deceive everybody who was not extremely
familiar with the matters in question and with all relevant earlier
publications26.
Bruins was thus widely held to be a highly competent scholar with
a most difficult temper, and he was able to maintain his status as an
expert almost to the end. While the relation between Neugebauer and
Thureau-Dangin can be taken as an exemplary instance of the functional-
ity of the norms of the scholarly community, the Bruins phenomenon
shows their possible dysfunctionality. Owing to the general conviction
that nobody advances devastating criticism without support in strong
arguments or indisputable facts, Bruins could retain his monopoly on the
interpretation of the Susa tablets almost up to the present date, thus
delaying advances in the field for decades.
from Plimpton 322 in order to make the counter-evidence to his theory disappear. He
told the story regularly but with changing protagonist. A third variant—less
obviously absurd—can be found in Bruins 1984: 118.
26
Once Bruins discovered that he had made a mistake he would cite in future
publications himself for the correct opinion and make somebody else responsible for
the erroneous point of view—preferably the one who had pointed out his mistake.
This can be exemplified by the sequence (Powell 1976: 432), (Bruins 1978), (Høyrup
1982: 32 n.6), and (Bruins 1984: 134 n.5). In the first of these, Powell had pointed out
that two mid-third millennium texts solve the same mathematical problem, one
correctly and another wrongly, and based his interpretation on analysis of the error;
in the second, Bruins rejected Powell’s interpretation of the first tablet without
noticing that his own interpretation was contradicted by the second; in the third, I
permitted myself to mention this neglect in a footnote; in the fourth, Bruins accuses
Powell of having overlooked the existence of the two parallel texts (and identifies
them wrongly).
David Fowler commented upon this example with the words »I could put together
a similar sequence over the Rhind papyrus 2/n table«.
- 24 -
generally influential though they may be, to broader issues. The first
challenges to the orthodoxy of the postwar period turned up between the
late 1960s and the mid-1970s, not from within but from outside the field:
they were formulated by scholars who knew considerably to much less
about Babylonian mathematics that the fathers of orthodoxy, but who
were more alert to metatheoretical questions than the disciples of these
(and as alert as the fathers had been in the 1930s). The issue was the
combined question of »Babylonian algebra« and »geometrical algebra«.
In 1969 a reprint of Neugebauer’s Vorgriechische Mathematik from
1934 appeared. In this work, we remember, Neugebauer had been much
more explicit on the algebraic interpretation than in his text editions.
Michael Mahoney, particularly well read in the history of that algebra
which was »a creation of the seventeenth century—AD!« (1971: 375), took
advantage of the occasion to ask in an essay review in which sense
»Babylonian algebra« could be taken to be algebraic. Distinguishing the
mere algebraic approach from algebra as developed from Viète to Descartes, he
argued that only the former term characterized the Babylonian type of
mathematics, which (in the reading of the texts that had been established
in MKT, TMB and MCT) contained only recipes for numerical proce-
dures. He made a plea (p. 377)
to wield Ockham’s razor when dealing with Babylonian mathematics
and not to assign to the Babylonians any concept, or form of mathemati-
cal thought, for which there is no explicit documentation, nor even need.
Apart from the choice of the term, Mahoney was broad-minded
concerning the idea of a Greek »geometrical algebra« inspired from
Babylonia, maintaining (p. 371) that
the theory can marshall a great deal of indirect evidence in its support
(neither it nor its opponents have anything like direct evidence).
Moreover, like most good theories, it explains phenomena it was not
originally intended to explain. For at the same time that it reveal
continuity, it throws discontinuity into sharper focus«.
Others were more sanguinary. A first attack had been launched by
Arpád Szabó (1969: 455ff). Granted »daß es eine ’babylonische Algebra’
- 25 -
wirklich gegeben hat—wovon O. Neugebauers Forschungen uns
überzeugen möchten« (p. 457), he rejected as extremely implausible that
the Greeks should have known about it—and if they had, he doubted
that they would have borrowed it. Instead he argued for an autochtonous
development of insights like those of Elements II,5, suggesting a starting
point in the kind of geometry told about in Plato’s Menon (82B-85E).
An even stronger rejection of the »monstrous, hybrid creature, a
contradiction in terms, a logical impossibility« was formulated by Sabetai
Unguru (1975: 77), in the context of a general attack on modernizing
interpretations of Greek mathematics. The argument does not really
involve Babylonian mathematics—which is claimed to belong to an
»arithmetical stage [...] in which the reasoning is largely that of elementary
arithmetic or based on empirically paradigmatic rules derived from
successful trials taken as a prototype« (p. 78), on the faith of Abel Rey’s
book on Greek mathematics from 1935, certainly less well-informed on
Babylonian topics and much more speculative than both Neugebauer and
van der Waerden.
Neither Szabó nor Unguru were really concerned with Babylonian
mathematics, so there may be no particular reason to blame them for
treating the subject superficially or for speaking from mere hearsay. None
the less, their interventions made it clear to historians of mathematics in
general that the orthodox interpretation was an orthodoxy and no
necessary plain truth—not least Unguru’s sharply formulated interven-
tion, which was answered by a no less sharp retort by Hans Freudenthal
(1977), by a venomous commentary by André Weil (1978), and by a
gently reasoned reply from van der Waerden (1976), which together
could not but arouse attention. They thus inaugurated the beginning of
a third phase in the study of Babylonian (now rather Mesopotamian)
mathematics, which was to reintegrate it into the general pattern of the
study of cultures and into the broader context of Mesopotamian
culture—while making perhaps the Babylonian calculators less interesting
for mathematicians as »the first of our kind«
- 26 -
III. FRESH START FROM SOURCES THROUGH NEW
APPROACHES, 1971 ONWARDS
- 27 -
in more or less corrupted Sumerian by Babylonians during the second
and first millennium.
These points were brought home even more clearly by the other
strand of his work from the early 1970s, which culminated in an article
on »The Antecedents of Old Babylonian Place Notation and the Early
History of Babylonian Mathematics« (1976). Here, he succeeded in
pushing back the firm terminus ante quem for the creation of the
sexagesimal place value system to the mid-21st century B.C. Only slightly
older mathematical exercise texts, on the other hand, were shown to
presuppose ideas which were to go into the place value system without
as yet possessing the tool. In this way Powell could make plausible a
connection between the invention of the place value notation and the
needs of the particular, tightly administrated Sumerian economy of the
21st century (»Ur III«).
In the same article Powell analyzed a number of mathematical
school exercise texts (i.e., genuine mathematical problem texts, as this term
has been used since the 1930s) from the mid- to late third millennium,
thus opening up a new vista even for the received conception of what
the history of Mesopotamian mathematics should be about.
Schmandt-Besserat’s contribution to the process of revitalization
was of a completely different character. Herself a Near Eastern archaeo-
logist, she discovered that a system for recording based on small clay
calculi (»tokens«) in varied shape and magnitude and previously only
noticed in late fourth millennium Susa had been widely used in the Near
East since around 8000 B.C. A number of proto-cuneiform signs seemed
to be pictures of tokens—a set of very early token forms (large and small
cones and spheres, the most common types from the very beginning) re-
emerging as sexagesimal number symbols: 1, 10, 60, 3600.
The discovery, which was speedily and efficiently published27 and
for this reason and because of its striking character soon widely known,
brought nothing immediately to the study of the history of Mesopo-
27
First in Syro-Mesopotamian Studies (1977) and soon in Discovery (1977a) and Scientific
American (1978).
- 28 -
tamian mathematics—nothing was changed in the interpretation of the
proto-literate number signs. This task was left to Jöran Friberg.
Friberg, a mathematician, started to look at cuneiform, proto-
cuneiform and proto-Elamite mathematics, computation and metrology
in the later 1970s. His first important discoveries, (inefficiently) published
in (1978) and (1979) and indeed primarily spread through personal
interaction during the first years, changed the whole understanding of
the earliest numerical and metrological notations28. In the best tradition
of the ingénu Friberg took a second look at the corpus of published proto-
cuneiform and proto-Elamite tablets29. Most of these contain numerical
or metrological notations, often accounts with single contributions and
total. Through analysis of the summations he was able to demonstrate
that the conventionally established interpretation was partly a myth, and
to single out in the proto-cuneiform material a number of metrological
sequences »integrating quantity and quality«30, together with a number
notation containing the steps 1, 10, 60, 12031. The same signs were used
in the various sequences, but with different mutual ratios; there was thus
28
Once again, Aisik Vajman should have been mentioned, if only his earlier works
on the same matters had not been even more badly published, and not backed by
personal contacts. Nobody outside the Soviet Union (and few scholars there) seems
to have taken serious note of them before Friberg.
29
»The proto-literate period« in Mesopotamia, to which the proto-cuneiform script
belongs, is dated approximately 3300 B.C. to 2900 B.C. (according to a compromise
between not too firmly established calibrated radiocarbon dates and stratigraphic
evidence). Proto-Elamite writing was used in the Iranian region during the second
half of this period. It appears to have been inspired by the invention of writing in
Mesopotamia, but makes use of a different inventory of signs; the metrologies,
however, are largely but not fully identical.
30
That is, for instance: the area 3 iku is denoted by threefold repetition of the sign
iku; in our metrology, on the contrary, three hectars are written »3 ha«, with
separation of quantity (»3«) from quality (»ha«).
31
It has later turned out that the proto-cuneiform accounting tablets make use of two
different counting systems used for counting objects belonging to different categories:
One, sexagesimal, with the steps 1, 10, 60, 600, 3600, and 36000; another, »bisexagesi-
mal«, containing the steps 1, 10, 60, 120, 1200, and 7200 (Damerow & Englund 1987:
126f, 133f, 165).
- 29 -
no longer any reason to believe that Schmandt-Besserat’s tokens
demonstrated the existence of a pure number system back to 8000
B.C.—the tokens could just as well have stood for specific measures of
grain, as held indeed by Schmandt-Besserat in later publications.
Friberg (1979: 33-43) was also able to decipher a complex computa-
tion text (probably a school exercise) from the later proto-literate period,
inaugurating thus the study of the mathematical techniques of this period
(500 years earlier than any genuinely mathematical text analyzed before).
32
Soubeyran published and discussed a collection of mostly mathematical texts from
Mari in (1984), while 10 mathematical problems from Tell Haddad have been
published and discussed by al-Rawi & Roaf (1984). The discovery of Ebla has brought
three texts with mathematical contents (analysis and previous publication history in
Friberg 1986).
Several new texts have been located by Friberg, cf. below.
33
So, critical reflection on Schmandt-Besserat’s thesis led Lieberman (1980) to
investigate the Sumerian use of two different ways to write numbers (»curviform«
and »cuneiform«) throughout the third millennium and connect it to a conjectural use
of tokens as a computation device (Lieberman did not know about Vajman’s and
Friberg’s work, and therefore accepted the identification of tokens with sexagesimal
numbers). Whiting (1984) analyzed Powell’s evidence and some supplementary texts
in an attempt to push backward the ante quem of the place value system, but
neglected to observe his distinction between the prerequisite idea of sexagesimal
regularization and extension and the establishment of a place value system strictu
sensu.
- 30 -
Nissen, and others34. I have also had the pleasure myself to belong to
this informal group.
Two members, Peter Damerow and I, had brought questions and
ideas inspired by cognitive psychology, by the sociology of knowledge,
and by anthropology into the field already before the formation of the
group. Peter Damerow, primarily a psychologist, took up the study of
early numerical notations and arithmetical techniques (tokens, Egyptian
and early cuneiform writing) as an approach to historical genetic
epistemology and to historically oriented philosophy of knowledge. The
first outcome was published in (1981)35. Soon afterwards, Peter Dame-
row joined the »Uruk Project« directed by Hans Nissen, and undertook
the computerization36 of a complete edition of proto-literate tablets from
Uruk; together with Robert Englund, an Assyriologist also engaged in
investigation of the administrative system of the Uruk III period, he
analyzed the complete numerical and metrological evidence in the proto-
literate tablets and was thus able to confirm and complete the results of
Vajman and Friberg (Damerow & Englund 1987). Afterwards, Damerow
and Englund (1989) applied the same method to Proto-Elamite material
from the locality Tepe Yahya, thus again confirming and completing
conjectures and preliminary results of Vajman and Friberg on this topic.
Together, Damerow, Englund and Nissen (1988a, 1988b) have drawn up
the resulting picture of the emergence of writing and numerical
notations, while Robert Englund has been able to demonstrate that the
specific administrative calendar used during Ur III for the computation
of rations, fodder and work obligations was used already in the proto-
34
I persist in disregarding astronomy—for which I apologize to Hermann Hunger,
who participated in several workshops. I also omit what a number of regularly
participating »general discussants« have contributed from their general competence
as historians of science or as Assyriologists: Kilian Butz, Jean-Pierre Grégoire,
Wolfgang Lefèvre, Johannes Renger, Jim Ritter, Arpád Szabó, Sabetai Unguru, as well
as everybody who only participated once.
35
A more refined analysis has been published (with considerable delay) as Damerow
1988.
36
Described in Damerow, Englund & Nissen 1989a.
- 31 -
literate period.
My own first contribution (1980) was part of a larger comparative
investigation of the interplay between institutional and social context and
mathematical mode of thought. I shall leave the word to Jöran Friberg
(1982: 13737):
[In] »an inter-cultural investigation of the role that the existence of an
institutionalized teaching of mathematics may have played for the
evolution and inner organization of mathematical thinking«, H. follows
[...] the gradual development of mathematical ideas and principles, and
the changing role of the profession of scribes and teachers of mathema-
tics, from the proto-literate period in Mesopotamia (when there are clear
signs of efforts to establish coherence and uniformity in the numeratio-
nal and metrological notations), via the school of scribes in the Ur III
period (when there was no room in the curriculum for »l’art pour l’art«)
to the proud and self-conscious Old Babylonian mathematicians (in a
time of far-reaching individualization of the economic and social life),
and finally to the time of the militaristic Kassites and their successors
(when mathematical traditions were kept alive only through the efforts
of a few »families of scribes«).
The specific aim of the chapter dealing with Mesopotamia was thus to
delineate the historically changing character of Mesopotamian mathema-
tical thought and to relate it to its use by a particular professional group
and through this to the broader context of the history of Mesopotamian
social structure and ideology.
Soon afterwards, a random question asked by Peter Damerow
drove me into another direction: a reinterpretation of the terminology
and (as a consequence) the substance and techniques of Old Babylonian
»algebra«. Through a method which can be characterized as »structural
semantics«38 I was able to show that the operations spoken of in the Old
37
Strictly speaking, Friberg does not report my original publication but a slightly later
Danish essay.
38
So I later found out—but my real inspiration for the method was vaguely
structuralist text analysis.
- 32 -
Babylonian texts could not possibly be genuinely arithmetical operations
with numbers39. Instead, the texts seemed to describe analytical opera-
tions on geometric figures, whose character can be described as »naive«
like those of Plato’s Menon, but whose substance is of course much more
sophisticated.
The outcome, which was first fully described in a fairly illegible
publication from (1984)40, put the question of the conjectural Babylonian
inspiration behind Greek »geometric algebra« in new light, since this
inspiration would be precisely of the type which Szabó had suggested
(cf. above).
While Damerow’s intervention as well as my own were thus
governed by questions and methods different from those of earlier times,
Friberg has demonstrated how far established questions and methods in
stubborn and bold combination can carry. One facet of his work has been
a continuation of the reading of the texts »as mathematics« unhampered
by too many metatheoretical scruples. But while the orthodoxy of 1940
to 1970 à la (Goetsch 1968) would do so without reference to the original
text, basing itself at best on a translation, Friberg respects the cuneiform
original more scrupulously than the best philologist. His view of what
pertains to mathematics has been as broad as that of the orthodoxy is
narrow, and encompasses every text and every publication concerned
with matters numerical, metrological, and computational.
One outcome of this was a Survey of Publications on Sumero-Akkadian
Mathematics, Metrology and Related Matters (1854-1982) from (1982), an
extensive annotated bibliography of which I have made ample use while
preparing the present paper, and which has set the stage for a new
delimitation of the field in better agreement with the place and function
39
The texts distinguish sharply between two different operations both traditionally
translated as »addition«, similarly two different »subtractive operations«, no less than
four »multiplications«, and two different »halves«.
40
A more readable exposition was published recently as Høyrup 1990.
- 33 -
of Mesopotamian mathematical activity in its own historical context41.
Another result has been the discovery and analysis (and, to some extent,
the publication) of a number of new mathematical tablets, some of them
from periods which hitherto had been completely devoid of mathematical
texts42. Finally, Friberg has provided new insights through his analyses
of a large number of texts—once again, mostly unpublished as yet.
Marvin Powell, the final regular member of the workshop circle
contributing actively to the field, has continued his work on the
development of metrologies in their technical and philological context.
What has then been achieved since 1971? What I have listed
appears to be an array of disconnected approaches, and it may perhaps
seem improbable that it should be possible to distinguish any trend, in
spite of the opportunity offered by a more or less regularly held
workshop.
In order to find out whether a trend can be distinguished we may
look at the outcome of the seemingly disparate approaches of the 1970s
and 1980s concerning specific problem fields.
Firstly the emergence of mathematics in the Near East. The
approaches of Schmandt-Besserat, Friberg, Damerow, Englund, Nissen
and to a lesser extent myself, and the dialogue between these approaches,
has made it possible to see the function of the token-system and its role
41
Unfortunately but for reasons of space, his recently published article "Mathematik"
(1990) in Reallexikon der Assyriologie was not allowed to cover the subject-matter as
broadly.
42
Friberg 1981; Friberg, Hunger & al-Rawi 1990a.
- 34 -
in the emergence of script, mathematical notations and mathematical
conceptualizations in the light of general cognitive psychology and
anthropological state formation theory, and to integrate the insights
thereby obtained with what else is known about the specific development
of social structure and culture in early Mesopotamia and its Near Eastern
surroundings.
A similar integration of mathematics into general history and
culture has been achieved for later periods. A first condition for this to
happen was that the myth of timeless Babylonian mathematics be
exploded—which, again, could only be done if texts illustrating
development and change could be discovered and analyzed. This was
done by Powell as far as the third millennium is concerned, and by
Friberg, in part together with Hunger, as regards the Late Babylonian
epoch. Powell also inaugurated the investigation of development in his
analysis of third millennium texts and of the changing character of
metrological systems, while Friberg’s and Hunger’s work on Late
Babylonian material has demonstrated how new text types and new
techniques connected to the new metrology had come into existence. My
own analysis of »algebraic« texts has brought back into focus the large
difference between the Old Babylonian and the Seleucid terminologies
(which was pointed out by Schuster (1931: 159f) and Neugebauer (1932a:
6f) but since then forgotten as unimportant); it has furthermore allowed
the conclusion that the differences in terminology reflect different
conceptualizations.
However, integration into general history and culture presupposes
more than this. Here it has been of extreme importance that Assyrio-
logists do not any longer automatically consider everything which looks
mathematical as »a matter for Neugebauer« but as a legitimate part of
their own field containing important information on society and culture.
An as yet unpublished example is Karen Nemet-Nejat’s work on
"Cuneiform Mathematical Texts as a Reflection of Everyday Life in
- 35 -
Mesopotamia"43. Englund’s analysis of the administrative calendar and
(to a lesser extent) his research in Ur III administrative procedures,
aswell as Powell’s investigations of the relations between metrology and
agricultural practices have added other facets to the »onionology« of
mathematics, to borrow a term coined by Ignace Gelb (1967: 8) for the
study of low-status but vital subjects like the distribution of onions, as
opposed to the more celebrated interest in gods and myths.
The outcome of my own work on »algebra« has been described by
Horst Klengel44 as a parallel to what has happened to the study of
Mesopotamian law: Instead of being analyzed in terms of its relation to
Roman Law—once so to speak the embodiment of the very Idea of
Law—it has come to be seen as an expression of Mesopotamian culture
and mode of thought, and concerned with the problems of Mesopota-
mian society.
Only one problem field has not given rise to a crystallization
involving several approaches: until now, I have been alone in recasting
theories about the transmission of Babylonian mathematical knowledge
and techniques to later cultures (with appurtenant transformation) and
about the relation between practitioners’ mathematics, scribal mathema-
tics and »scientific« mathematics45. In so far as this makes part of a
trend it is not inside the small community of students of Mesopotamian
mathematics but due to my interaction with scholars from neighbouring
fields.
Apart from that, it should be clear that the development over the
last 10 years does expresses a trend, viz away from the attitude that the
mathematical knowledge of the Babylonians should be studied as a step
in the ladder leading to, and hence from the perspective of modern mathe-
matics, and toward the position of a multi-dimensional anthropology
43
Drafts of this work were presented at the 1988 Berlin Workshop, but the author had
taken up the subject independently of the Berlin collaboration.
44
In private conversation, and thus not quoted verbally.
45
In this connection I disregard van der Waerden’s work on Geometry and Algebra in
Ancient Civilizations (1983), since the fundamental perspective, though certainly a
recast, is not that of Babylonian mathematics.
- 36 -
where mathematics is primarily studied in relation to its historical
context46, and where the distinction between »external« and »internal«
causation is regarded as only relative.
Thus seen, the »new trend« in the historiography of Babylonian
mathematics is not specific to this field. It can be found in many quarters
of the history of science, and in the humanities in general. It is therefore
not so strange that the historiography of Mesopotamian mathematics has
performed what Michael Mahoney hoped for (but doubted could be done
from extant sources): bring about the »transition from mathematics to
mathematical thought« (1971: 378). Indeed, Mahoney’s hopes coincided
with the motivation of many scholars, including some of those who took
up new approaches to Babylonian mathematics.
That it could be done, in spite of Mahoney’s pessimism, depended
on the real interdisciplinarity of the group which was engaged in the
task—not only as a collectivity but also individually (which is the
presupposition that collective interdisciplinarity can work): negation of
the mutual segregation not only of the two but of the N cultures was
essential.
Neither Neugebauer nor Thureau-Dangin respected the segregation.
In their time, however, the source situation was still so that Neugebauer
(1934: 204) was forced to conclude that »wir über die ganze Stellung der
babylonischen Mathematik im Rahmen der Gesamtkultur praktisch noch
gar nichts wissen«, as quoted above. Afterwards, as the source situation
improved, orthodoxy took care that historians of mathematics did not
46
This does not prevent that categories of modern mathematics can be used when
needed as analytical tools (cf. note 9)—even if we stop asking as our fundamental
question »how the equations of the Babylonians looked« we may still take notice that
a problem »I have added [the measuring numbers of] the side and the area of a
square, and the result was 110« shares essential features with modern equations, and
can be termed no more adequately by a single word. We still need an Archimedean
point from where to describe the world, and purist who refuse to speak about
»algebra« and insist, e.g., on »numerical mathematics« are mistaken—»numbers« are
no more transhistorically immutable than »algebra«, as shown by Vajman, Friberg
and Peter Damerow.
- 37 -
discover or bother, while Assyriologists got no opportunity to tell. It has
only been since 1970-80 that an improved source situation, disrespect for
»cultural« boundaries within the scholarly world, and the combination
of individual and collective interdisciplinarity have created the break-
through to those new questions which could now be answered. The
process has also revealed the distinction internal versus external to be a
sociological accident just as much as an absolute cognitive category.
The new approaches have certainly brought about new insights on
many levels. They have probably also created a void, or at least a loss:
The orthodox picture of Babylonian mathematics allowed mathematicians
the illusion that they could overcome the two-culture split without
leaving home: »we mathematicians got our own humanities, Greek and
Babylonian mathematics—they look precisely as mathematics, and are
written in x and y«. Once the new picture has been discovered by general
historians of mathematics and has gone into the textbooks, it will perhaps
be less easy to use Babylonian mathematics as historical staple food for
such justifications of present-day mathematics. May the new insights
contribute instead to overthrowing all 2- and N-culture distinctions in
earnest.
POSTSCRIPT
- 38 -
orientations of a field within which I work myself has been severely
distorted by the necessary false consciousness of a participant; what I do
know is that the preceding essay is a hybrid: not an unpolluted insider’s
report, because of the Fall from innocence caused by the author’s attempt
to apply a voyeurist, metatheoretical and sociological point of view; nor
on the other hand a real historical or sociological account, since any
participant’s belief that he can step outside the process is illusory.
In any case: I learnt quite a bit about a field with which I felt to be
familiar through looking at what used to be (good or bad) theory under
the aspect of primary sources. It is a pleasure no less than an obligation to
express my gratitude to the Greek Society for the History of Science and
Technology for giving me the occasion to do so.
It is also obligation as well as pleasure to thank those participants
in the conference who reacted to my talk and to the preliminary written
version of the present paper—in particular to David Fowler, who also
had the kindness to correct the English of a number of passages.
Baqir, Taha, 1950. "An Important Mathematical Problem Text from Tell Harmal".
Sumer 6, 39-54.
Baqir, Taha, 1950a. "Another Important Mathematical Text from Tell Harmal".
Sumer 6, 130-148.
Baqir, Taha, 1951. "Some More Mathematical Texts from Tell Harmal". Sumer 7,
28-45.
Boyer, Carl B., 1968. A History of Mathematics. New York: Wiley.
Bruins, Evert M., 1978. [Critical abstract of Powell 1976]. Zentralblatt für Mathematik
und ihre Grenzgebiete 357, 7-8.
- 39 -
Bruins, Evert M., 1984. "Requisites for the Interpretation of Ancient Mathematics".
Janus 71, 107-134.
Böhme, Gernot, Wolfgang van den Daele & Wolfgang Krohn, 1973. "Die
Finalisierung der Wissenschaft". Zeitschrift für Soziologie 2, 128-144.
Damerow, Peter, 1981. "Die Entstehung des arithmetischen Denkens. Zur Rolle der
Rechenmittel in der altägyptischen und der altbabylonischen Arithmetik", pp.
11-113 in P. Damerow & W. Lefèvre (eds), Rechenstein, Experiment, Sprache.
Historische Fallstudien zur Entstehung der exakten Wissenschaften. Stuttgart: Klett-
Cotta.
Damerow, Peter, & Robert K. Englund, 1987. "Die Zahlzeichensysteme der
Archaischen Texte aus Uruk", Kapitel 3 (pp. 117-166) in M. W. Green & Hans
J. Nissen, Zeichenliste der Archaischen Texte aus Uruk, Band II (ATU 2). Berlin:
Gebr. Mann.
Damerow, Peter, 1988. "Individual Development and Cultural Evolution of
Arithmetical Thinking", pp. 125-152 in S. Strauss (ed.), Ontogeny, Phylogeny, and
Historical Development. (Human Development, Volume 2). Norwood, N.J.: Ablex
Publishing.
Damerow, Peter, Robert K. Englund & Hans J. Nissen, 1988a. "Die Entstehung der
Schrift". Spektrum der Wissenschaften, February 1988, 74-85.
Damerow, Peter, Robert K. Englund & Hans J. Nissen, 1988b. "Die ersten
Zahldarstellungen und die Entwicklung des Zahlbegriffs". Spektrum der
Wissenschaften, March 1988, 46-55.
Damerow, Peter, & Robert K. Englund, 1989. The Proto-Elamite Texts from Tepe
Yahya. (The American School of Prehistoric Research, Bulletin 39). Cambridge,
MA.: Peabody Museum of Archaeology and Ethnology / Harvard University
Press.
Damerow, Peter, Robert Englund & Hans Nissen, 1989a. "Zur rechnergestützten
Bearbeitung der archaischen Texte aus Mesopotamien (ca. 3200-3000 v. Chr.)".
Mitteilungen der Deutschen Orient-Gesellschaft 121, 139-152.
Eves, Howard, 1969. An Introduction to the History of Mathematics. Third edition.
New York etc.: Holt, Rinehart and Winston. 1st ed. 1964.
Frank, Carl, 1928. Straßburger Keilschrifttexte in sumerischer und babylonischer
Sprache. (Schriften der Straßburger Wissenschaftlichen Gesellschaft in
Heidelberg, Neue Folge, Heft 9). Berlin & Leipzig: Walter de Gruyter.
Freudenthal, Hans, 1977. "What is Algebra and What Has It Been in History?"
Archive for History of Exact Sciences 16, 189-200.
- 40 -
Friberg, Jöran, 1978. "The Third Millennium Roots of Babylonian Mathematics. I.
A Method for the Decipherment, through Mathematical and Metrological
Analysis, of Proto-Sumerian and proto-Elamite Semi-Pictographic Inscriptions".
Department of Mathematics, Chalmers University of Technology and the University
of Göteborg No. 1978-9.
Friberg, Jöran, 1979. "The Early Roots of Babylonian Mathematics. II: Metrological
Relations in a Group of Semi-Pictographic Tablets of the Jemdet Nasr Type,
Probably from Uruk-Warka". Department of Mathematics, Chalmers University of
Technology and the University of Göteborg No. 1979-15.
Friberg, Jöran, 1981. "Methods and Traditions of Babylonian Mathematics. II: An
Old Babylonian Catalogue Text with Equations for Squares and Circles". Journal
of Cuneiform Studies 33, 57-64.
Friberg, Jöran, 1982. "A Survey of Publications on Sumero-Akkadian Mathematics,
Metrology and Related Matters (1854-1982)". Department of Mathematics,
Chalmers University of Technology and the University of Göteborg No. 1982-17.
Friberg, Jöran, 1986. "The Early Roots of Babylonian Mathematics. III: Three
Remarkable Texts from Ancient Ebla". Vicino Oriente 6, 3-25.
Friberg, Jöran, 1990. "Mathematik". Reallexikon der Assyriologie und Vorderasiatischen
Archäologie VII, 531-585. Berlin & New York: Walter de Gruyter.
Friberg, Jöran, Hermann Hunger & Farouk al-Rawi, 1990a. "»Seeds and Reeds«:
A Metro-Mathematical Topic Text from Late Babylonian Uruk". Baghdader
Mitteilungen 21, 483-557, Tafel 46-48.
Gandz, Solomon, 1937. "The Origin and Development of the Quadratic Equations
in Babylonian, Greek, and Early Arabic Algebra". Osiris 3, 405-557.
Gelb, Ignace, 1967. "Approaches to the Study of Ancient Society". Journal of the
American Oriental Society 87, 1-8.
Goetsch, H., 1968. "Die Algebra der Babylonier". Archive for History of Exact
Sciences 5 (1968-69), 79-153.
Goetze, Albrecht, 1951. "A Mathematical Compendium from Tell Harmal". Sumer
7, 126-155.
Gundlach, Karl-Bernhard, & Wolfram von Soden, 1963. "Einige altbabylonische
Texte zur Lösung »quadratischer Gleichungen«". Abhandlungen aus dem
mathematischen Seminar der Universität Hamburg 26, 248-263.
Heath, Thomas L. (ed., trans.), 1926. The Thirteen Books of Euclid’s Elements,
Translated with Introduction and Commentary. 2nd revised edition. 3 vols.
Cambridge: Cambridge University Press / New York: Macmillan. 1st ed. 1908.
Reprint New York: Dover, 1956.
- 41 -
Hofmann, Joseph Ehrenfried, 1953. Geschichte der Mathematik. 3 Bände. (Sammlung
Göschen 226, 875, 882). Berlin: Walter de Gruyter, 1953, 1957, 1957.
Høyrup, Jens, 1980. "Influences of Institutionalized Mathematics Teaching on the
Development and Organization of Mathematical Thought in the Pre-Modern
Period. Investigations into an Aspect of the Anthropology of Mathematics".
Materialien und Studien. Institut für Didaktik der Mathematik der Universität
Bielefeld 20, 7-137.
Høyrup, Jens, 1982. "Investigations of an Early Sumerian Division Problem, c. 2500
B.C.". Historia Mathematica 9, 19-36.
Høyrup, Jens, 1984. Babylonian Algebra from the View-Point of Geometrical Heuristics.
An Investigation of Terminology, Methods, and Patterns of Thought. Roskilde:
Roskilde University Centre, Institute of Educational Research, Media Studies
and Theory of Science.
Høyrup, Jens, 1988. "Jordanus de Nemore, 13th Century Mathematical Innovator:
an Essay on Intellectual Context, Achievement, and Failure". Archive for History
of Exact Sciences 38, 307-363.
Høyrup, Jens, 1990. "Algebra and Naive Geometry. An Investigation of Some Basic
Aspects of Old Babylonian Mathematical Thought". Altorientalische Forschungen
17, 27-69, 262-354.
Kline, Morris, 1972. Mathematical Thought from Ancient to Modern Times. New York:
Oxford University Press.
Lieberman, Stephen J., 1980. "Of Clay Pebbles, Hollow Clay Balls, and Writing: A
Sumerian View". American Journal of Archaeology 84, 339-358.
Mahoney, Michael S., 1971. "Babylonian Algebra: Form vs. Content". Studies in
History and Philosophy of Science 1 (1970-71), 369-380.
MCT: O. Neugebauer & A. Sachs, Mathematical Cuneiform Texts. (American
Oriental Series, vol. 29). New Haven, Connecticut: American Oriental Society,
1945.
Merton, Robert K., 1942. "A Note on Science and Democracy". Journal of Legal and
Political Sociology 1, 115-126.
MKT: O. Neugebauer, Mathematische Keilschrift-texte. I-III. (Quellen und Studien
zur Geschichte der Mathematik, Astronomie und Physik. Abteilung A: Quellen.
3. Band, erster-dritter Teil). Berlin: Julius Springer, 1935, 1935, 1937. Reprint
Berlin etc.: Springer, 1973.
Neugebauer, O., 1929. "Zur Geschichte der babylonischen Mathematik". Quellen
und Studien zur Geschichte der Mathematik, Astronomie und Physik. Abteilung B:
Studien 1 (1929-31), 67-80.
- 42 -
Neugebauer, O., & W. Struve, 1929a. "Über die Geometrie des Kreises in
Babylonien". Quellen und Studien zur Geschichte der Mathematik, Astronomie und
Physik. Abteilung B: Studien 1 (1929-1931), 81-92.
Neugebauer, O., 1930. "Beiträge zur Geschichte der Babylonischen Arithmetik".
Quellen und Studien zur Geschichte der Mathematik, Astronomie und Physik.
Abteilung B: Studien 1 (1929-31), 120-130.
Neugebauer, Otto, 1932. "Zur Transkription mathematischer und astronomischer
Keilschrifttexte". Archiv für Orientforschung 8 (1932-33), 221-223.
Neugebauer, Otto, 1932a. "Studien zur Geschichte der antiken Algebra I". Quellen
und Studien zur Geschichte der Mathematik, Astronomie und Physik. Abteilung B:
Studien 2 (1932-33), 1-27.
Neugebauer, Otto, 1934. Vorlesungen über Geschichte der antiken mathematischen
Wissenschaften. I: Vorgriechische Mathematik. (Die Grundlehren der mathemati-
schen Wissenschaften in Einzeldarstellungen, Bd. XLIII). Berlin: Julius Springer.
Neugebauer, O., 1936. "Zur geometrischen Algebra (Studien zur Geschichte der
antiken Algebra III)". Quellen und Studien zur Geschichte der Mathematik,
Astronomie und Physik. Abteilung B: Studien 3 (1934-36), 245-259.
Neugebauer, O., 1969. The Exact Sciences in Antiquity. Reprint of 2nd Edition. New
York: Dover. (1st ed. 1951).
Powell, Marvin A., 1971. Sumerian Numeration and Metrology. Dissertation,
University of Minnesota.
Powell, Marvin A., 1972. "Sumerian Area Measures and the Alleged Decimal
Substratum". Zeitschrift für Assyriologie und Vorderasiatische Archäologie 62 (1972-
73), 165-221.
Powell, Marvin A., 1976. "The Antecedents of Old Babylonian Place Notation and
the Early History of Babylonian Mathematics". Historia Mathematica 3, 417-439.
Rawi, Farouk N. H. al-, & Michael Roaf, 1984. "Ten Old Babylonian Mathematical
Problem Texts from Tell Haddad, Himrin". Sumer 43, 195-218.
Schmandt-Besserat, Denise, 1977. "An Archaic Recording System and the Origin
of Writing". Syro-Mesopotamian Studies 1:2.
Schmandt-Besserat, Denise, 1977a. "The Invention of Writing". Discovery: Research
and Scholarship at the University of Texas at Austin 1:4, 4-7.
Schmandt-Besserat, Denise, 1978. "The Earliest Precursor of Writing". Scientific
American 238:6, 38-47 (European edition).
- 43 -
Schuster, H. S., 1930. "Quadratische Gleichungen der Seleukidenzeit aus Uruk".
Quellen und Studien zur Geschichte der Mathematik, Astronomie und Physik.
Abteilung B: Studien 1 (1929-31), 194-200.
Snow, C. P., 1964. The Two Cultures: And a Second Look. An Expanded Version of the
Two Cultures and the Scientific Revolution. London: Cambridge University Press.
Soubeyran, Denis, 1984. "Textes mathématiques de Mari". Revue d’Assyriologie 78,
19-48.
Szabó, Arpád, 1969. Anfänge der griechischen Mathematik. München & Wien: R.
Oldenbourg / Budapest: Akadémiai Kiadó.
Thureau-Dangin, F., 1936. "L’Équation du deuxième degré dans la mathématique
babylonienne d’après une tablette inédite du British Museum". Revue
d’Assyriologie 33, 27-48.
Thureau-Dangin, F., 1938. "La Méthode de fausse position et l’origine de
l’algèbre". Revue d’Assyriologie 35, 71-77.
Thureau-Dangin, F., 1940. "L’Origine de l’algèbre". Académie des Belles-Lettres.
Comptes Rendus 1940, 292-319.
TMB: F. Thureau-Dangin, Textes mathématiques babyloniens. (Ex Oriente Lux, Deel
1). Leiden: Brill, 1938.
TMS: E. M. Bruins & M. Rutten, Textes mathématiques de Suse. (Mémoires de la
Mission Archéologique en Iran, XXXIV). Paris: Paul Geuthner, 1961.
Unguru, Sabetai, 1975. "On the Need to Rewrite the History of Greek Mathema-
tics". Archive for History of Exact Sciences 15, 67-114.
Vajman, A. A., 1961. Šumero-vavilonskaja matematika. III-I Tysjačeletija do n. e.
Moskva: Izdatel’stvo Vostočnoj Literatury.
van der Waerden, B. L., 1962. Science Awakening. 2nd Edition. Groningen:
Noordhoff.
van der Waerden, B. L., 1976. "Defence of a »Shocking« Point of View". Archive for
History of Exact Sciences 15, 199-210.
van der Waerden, B. L., 1983. Geometry and Algebra in Ancient Civilizations. Berlin
etc: Springer Verlag.
Vogel, Kurt, 1959. Vorgriechische Mathematik. II. Die Mathematik der Babylonier.
(Mathematische Studienhefte, 2). Hannover: Hermann Schroedel / Paderborn:
Ferdinand Schöningh.
von Soden, Wolfram, 1939. [Review of TMB]. Zeitschrift der Deutschen Morgenländi-
schen Gesellschaft 93, 143-152.
von Soden, Wolfram, 1964. [Review of TMS]. Bibliotheca Orientalis 21, 44-50.
- 44 -
von Soden, Wolfram, 1974. Sprache, Denken und Begriffsbildung im Alten Orient.
(Akademie der Wissenschaften und der Literatur. Abhandlungen der Geistes-
und Sozialwissenschaftlichen Klasse, 1973 Nr. 6). Mainz: Akademie der
Wissenschaften und Literatur / Wiesbaden: Franz Steiner.
Weil, André, 1978. "Who Betrayed Euclid?" Archive for History of Exact Sciences 19,
91-93.
Whiting, Robert M., 1984. "More Evidence for Sexagesimal Calculations in the
Third Millennium B.C." Zeitschrift für Assyriologie und Vorderasiatische Archäo-
logie 74, 59-66.
- 45 -
Author’s Address:
Jens Høyrup
Institute of Communication Research,
Educational Studies Studies
and Theory of Science
University of Roskilde
P. O. Box 260
DK-4000 Roskilde
Denmark