Reconstructing Akkadian Texts and Singing the Twelve Days of Christmas – By Jonathan Stökl

Jonathan Stökl on Martin Worthington’s Principles of Akkadian Textual Criticism

Martin Worthington, Principles of Akkadian Textual Criticism, de Grutyer, 2012, 350 pp. $140.00.
Martin Worthington, Principles of Akkadian Textual Criticism, De Grutyer, 2012, 350 pp. $140.00.

This book is a rare jewel: an enjoyable, learned and very well informed read on textual criticism. Most works on textual criticism tend to be much more difficult (indeed, at times, boring). But not so Worthington’s.

It is somewhat surprising in a discipline such as Assyriology that relies as heavily on ancient manuscripts that this is the first treatment (hence the word ‘principles’ in the title) focused on textual criticism. Worthington points out that many useful and erudite comments can be found in footnotes of text editions, etc., but there is no book-length study of the issue as yet. Principles of Akkadian Textual Criticism has changed that state of affairs admirably.

Worthington starts by giving an introduction to the purposes of textual criticism: the establishment of the earliest achievable text-form as well as all intervening forms of the text. It is important to note that, in line with most recent text-critical approaches, the search for a putative original text is not really the aim of textual criticism, as the original text usually lies outside the realm of the possible. The greater part of Worthington’s theoretical frame-work is based on work done in Classics and Medieval (English) literature. The absence of virtually all (Hebrew) Bible text-criticism surprised me, especially because of the relative level of sophistication reached in that discipline.

Worthington provides examples of the text-critical principles from disciplines such as ethnomusicology, Italian politics and English literature. My personal favourite is from a study of the evolution of a line in the Christmas song The Twelve Days of Christmas:

‘Within a list of different types of bird, this carol features “five gold rings”. Rings are out of place in a list of birds, and indeed originally the song ran “five gold wrens”’ (p. 43).

This particular example illustrates the value of studying the intervening stages that a text goes through. If we were simply interested in the earliest form of the text we would focus on the ‘golden wrens’, but if we also want to know how the text was sung and understood, say, by a twentieth century audience, we would need to look at the text which includes the ‘golden rings’. The case is identical to that of an ancient text where a variant was transmitted through the ages. It is good to see Worthington’s double focus on the early text as well as on the various forms a text has taken throughout its history.

In its basic principles text criticism as applied to texts written in (non-alphabetic) cuneiform is no more complicated that in any other writing system. In practice, however, it is significantly more complex. Many mistakes in the transmission of a text arise out of the confusion of signs by ancient copyists (and modern readers), simply because there are many more signs and the differences between them can be minute. This is complicated yet further by the polyvalence of Akkadian cuneiform, which poses an extra problem for the textual critic: it is possible that an obscure word should actually be read as a different word, or, if a scribe made a mistake, we might not realise the mistake straight away because we read the sign in question as a different one, and therefore the entire word differently. Let me give an example drawn from Worthington’s study, who, in turn, quotes Stefan Maul and Rita Strauß. In the edition of a Neo-Assyrian ritual tablet they read a certain line as marṣa(tu.ra)[1] šulna-a tulabbassu(mu4me?-su)[2]. Another tablet which contains the same text writes the second word slightly differently: túgšà-ha-a. The sign šul can be read as šáh in Neo-Assyrian. According to Maul and Strauß the reading of the following sign as na is certain. Maul and Strauß draw the conclusion that šulna-a’ should be read as dun-na-a[3] and they posit an otherwise unattested word dunnû. Alternatively, Worthington suggests, they could have regarded the na as a mistake for a ha; in that case they could have read the sentence as marṣa šah(h)â tulabbassu (‘you clothe the patient with cloth’).

Worthington’s careful awareness of the issue helps him to create better models for understanding the transmission of texts and the possible errors that can creep in.

Some readers of the book might take issue with the vast amount of documentation. Worthington defends this approach as necessary because all variance is variant’ only on account of being different from the norm. Thus, we only know that the North-Eastern Scottish English expression ‘I says’ is a variant to the normal expression in English (‘I said’) because of frequency and distribution of use (this is the reviewer’s example, not Worthington’s). The same is true in antiquity, and Worthington points to the example of the lack of doubling of consonants in Neo-Assyrian (p. 84): Old Babylonian Akkadian writes the form iparras both as i-pa-ra-as and (more frequently) as i-pa-ar-ra-as. In Neo-Assyrian the spelling i-pa-ra-as seems to be more common. This could mean that Neo-Assyrian Akkadian simply uses a different spelling of the same word iparras or that in Neo-Assyrian the correct form would have been ipāras. Without counting and listing examples it is difficult to come to a conclusion (and in the case of Neo-Assyrian ipar[r]as we simply do not have enough evidence to determine which is the case). As Worthington puts it: ‘In grammar, as in war, there is strength in numbers’. But he also makes it clear that more theoretical thinking is required: thus, not only is the frequency of a spelling important, but also the relative frequency and the distribution through time and space (whether it is only used in one manuscript, only in one dialect, only in one region, etc.). This is undoubtedly the case, and it is good to have an understanding of the frequency of certain mistakes and by-forms, etc. (only through detailed statistics is it possible to discern that peculiar forms may just be regional or temporal by-forms. The North-Eastern Scottish narrative[4] form ‘I says’ is a good example of that). Even though some readers may chafe at the amount of documentation that Worthington provides, most of the examples are interesting and I would happily have read even more.

Any subject working with manuscripts, particularly ones that have been transmitted over a long period, needs to develop a form of textual criticism appropriate to these manuscripts.

I share Worthington’s hope that his book will help toward achieving that aim in Assyriology and Cuneiform studies in general. If you are interested in textual criticism, Akkadian literature, or the transmission process of any form of literature, and you want the experience to be a good one, read this book.


[1] The Sumerian word tu.ra is used to spell the Akkadian word marṣu/-a (‘patient’).

[2] The Sumerian word mu4me is a way of spelling any form of the Akkadian verb labāšu (‘to dress’). The ending (‘phonetic complement’) su indicates that the form ends with –su. The context suggests that mu4me?-su be read as tulabbassu (you shall dress him).

[3] Dun is another reading of the sign šul.

[4] A ‘narrative’ is a particular form of the verb used in stories (narratives). Most languages use some form of past tense for this purpose.

(Visited 56 times, 1 visits today)