Fwd: [IER] RE:Proto-Elamite, was : Literacy in Himalyas - a post by Zoller

[Guest guest]

Indo-Eurasian_research , jacob dahl

<jacob.dahl wrote:

 

Dear List,

 

First a brief introduction for those who do not know much about

proto-Elamite.

I will include here some notes on the early history of the discovery

and first attempts of decipherment, since this has some importance

for our problems. (sorry if this is going to be a bit long).

 

I will try to comment on Steve's thoughts on literate vs. illiterate

urban societies as I go along.

I think the work by the Berlin-Uruk group, now the CDLI

(http://www.cdli.ucla.edu) is a good example of what can be achieved

by studying ancient archives even when the " text cannot be read " . We

have more than 6.000 texts from Southern Mesopotamia dating to

between 3300 and 2900 BC, the majority of which are administrative

documents (some 15% school texts, mostly lists). The writing-system

of these texts mostly conform with what Damerow termed proto-writing

(Damerow 1999, see

http://www.mpiwg-berlin.mpg.de/Preprints/P114.PDF), i.e., a writing

system which was not invented to represent a (specific) language. Or

at least show a very weak link between writing and an oral language.

Yet we know an awful lot about late Uruk society (not least because

a

successor writing-system is known and deciphered). However, also in

Uruk the writing served but one purpose, administration, the school

served the administration too, by educating new administrators. It

is

probable that the school redefined itself at some point and began

inventing writing with no use for the administration. But that is

another question.

 

HISTORY OF THE PROTO-ELAMITE TEXTS:

The overwhelming majority of proto-Elamite texts (ca. 1600) were

found during the French excavations of Susa (in South-Eastern Iran)

in the beginning of the 20th century. Jacques de Morgan, the leader

of the excavations, until 1907, was in fact not an archaeologist,

but

a mining-engineer, his primary goal was to find " movable " artifacts

to fill the halls of the Louvre. He dug in increments of 5 meters,

apparently without any drawings neither of the trenches, nor of the

find-spots, etc. I don't want to engage in unnecessary criticism of

early " archaeologists " , I am only bringing this up to illustrate why

we have had so big problems with chronology, reconstructing

archives,

and distinguishing writing-phases, etc.

Some proto-Elamite texts have been found during later controlled

excavations at Susa, not always in a context that is easily

understood, but certainly helping our chronological understanding.

It is in fact possible to re-create some ancient Susa archives, but

many problems remain. If there is an interest in discussing archives

I would be happy to expand on this topic.

Later excavations at sites all over Iran have revealed samples of

writing that unquestionably can be defined as proto-Elamite

(although

sometimes with strong local variations).

This topic needs to be addressed also, when does a local variant

stop

being proto-Elamite, and begins to be something of its own? Texts

from Malyan, for example, are much closer related to the Susa texts

than texts from Yayha. This claim is simply based on occurrences of

signs and choice of sign-forms.

Traditionally the proto-Elamite texts are dated to around 3000 BC

(almost contemporary with the Uruk III period in Mesopotamia). The

writing system was used for a very brief period.

The name proto-Elamite was give to these texts since they were long

thought of as a hypothetical ancestor of an equally hypothetical

indigenous Elamite writing system. It is not known which language

the

texts were invented for, which they were used to write (if any), and

there is no proof of any relation to the writing system known as

linear-Elamite, known from some 20 artifacts written more than 750

years after the proto-Elamite tablets. Linear-Elamite too, remains

undeciphered. In the case of linear-Elamite where we have only a

couple of handfuls of texts (rumors about new longer inscriptions

don't help much before they materialize into publications), it is

doubtful that we can ever advance in a decipherment.

Another important point is that proto-Elamite apparently was in use

for a very brief period of time (fore-runners exist), perhaps only

one or two generations. That may vbe why we don't see much

standardization of the sign repertoire (see also below). However, I

think it will be possible to distinguish two phases of writing

(early

and standard).

 

EARLY WORK ON THE PROTO-ELAMITE TEXTS:

The majority of the proto-Elamite texts were published shortly after

their discovery, this is of course laudable in itself, but the

quality of most of the publications is below what could have been

asked for (as a rule top-edge inscriptions, often holding one

numerical sign, were left out: these markings may give vital

administrative clues of course! The reverse was often not copied,

etc. As J. Friberg has recently stated it is not to be expected that

a good copy can be made by anyone who doesn't understand the text).

Unfortunately no claim concerning proto-Elamite can be made without

reference to the originals.

Each major publication was accompanied by a sign-list, and in fine

agreement with theories, also put forward in Farmer/Sproat/Witzel

2004 (http://www.safarmer.com/indus/simpleproof.html), the number of

signs expanded with each text publication! The final list counted

more than 5000 signs (MDP 31, de Mecquenem 1949). This

unrealistically high number has now been reduced to about 1300 (de

Mecquenem's list included many variants, and broken signs, and

listed

many signs over and over again, because of a lack of understanding

of

slightly different hands, etc.).

In 1974 P. Meriggi published a new sign-list which, although it too

suffers from countless problems, has been used by members of the

CDLI

to generate a new sign-list. Meriggi also attempted to decipher

proto-Elamite using traditional linguistic methods. He was not

successful.

In the late seventies the Swedish mathematician Jöran Friberg got

interested in early writing, in particular proto-cuneiform, and

proto-Elamite. His new method of decipherment decoding the

metrological systems, has proven to be very profitable, and we are

now able to distinguish between counted objects and non-counted

objects, etc.

Proto-Elamite holds a special place in the history of writing since

it can be shown to be related, in some way or another, to the

slightly older proto-cuneiform writing system from Southern

Mesopotamia. I don't think many any longer believes that the scribes

at Susa simply borrowed their writing system from Mesopotamia.

Rather

the two systems seem to have had common roots.

This is a point that interest me much at the moment and I would be

happy to expand on this, e.g., the transmission of writing from one

culture to another, or as it is more likely in this case, the common

background of two writing-system.

 

A brief note on the often reported " new examples of proto-Elamite

writing " . I have seen no evidence for a any of the claims of writing

at Jiroft. It would not surprise me if Jiroft has a proto-Elamite

stratum too, but that remains to be found! Every so often a new

proto-Elamite tablet is discovered at a site in Iran, one of the

latest at Tepe Ozbaki, in the North, how we shall understand this is

also not certain to me. The reported proto-Elamite texts (often just

one " sign " itched into some random surface), coming from places far

away (Afghanistan, Oman, etc.) can all be disregarded, as far as I

am

concerned (and I have seen quite a few by now).

 

ABOUT THE PROTO-ELAMITE WRITING-SYSTEM:

All proto-Elamite texts, all, are administrative, one or two are

perhaps school texts (mathematical exercise texts); we have no

lists!

Compare this with the late-Uruk corpus (roughly contemporary texts

from Mesopotamia) where about 15% were school texts, i.e., lots of

lists. Another topic that would be worthwhile expanding on: the need

of structure in the process of creating a real writing system.

(Lists

in Mesopotamia does not appear before Uruk III, that is late Uruk, a

stage when proto-cuneiform underwent a massive standardization,

perhaps the proto-Elamite just gave up writing before it came to

this).

 

Now, proto-Elamite has long been considered a prime candidate for

decipherment, actually many have believed it is more suited for a

traditional decipherment than, i.e., proto-cuneiform. This is due to

the fact that proto-Elamite has a high number of sign-strings with 5

or 6 signs, or even more. Additionally the signs are very abstract.

In fact the majority of the signs do not look like anything we can

identify. Many of those signs that look like something don't stand

for what they look like, e.i., the image of a certain animal does

not

seem to be a counted object, but rather part of the string in front

of the counted object, etc.

We split up proto-Elamite texts according to entries holding a

counted object and a numerical notation (remember again that all

texts are administrative). It is relatively easy to isolate the

object signs, but the remaining signs are not easy to deal with (see

more below).

 

Initial graphotactical analysis did not, however, produce any

meaningful results (see Englund 2004: available as preprint at the

MPIWG server at: http://www.mpiwg-berlin.mpg.de/preprints/P183.pdf).

That is we were unable to find any meaningful repetitions.

Surprising

in a copus of more than 10.000 lines of text! But remember that

these

initial tests were made on a data-set compiled using the old

hand-copies, and the sign-list of Meriggi. Another note: the

sign-frequencies that I reported in 2001

(http://cdli.ucla.edu/pubs/cdlb/2002/cdlb2002_001.html) hold, more

or

less, even after collation of almost the entire corpus. Perhaps it

is

time I do some counting again though. However, a new test should

take

into account that proto-Elamite may have had two writing-phases, and

that we may see different results in each of these.

 

Proto-Elamite is unique for an early writing system since it uses no

word-dividers and the entries are not arranged in boxes. In

actuality, a proto-Elamite text is arranged sequentially, and not in

any visible order of hierarchies. This in-line representation of the

entries is quite different from all other early writing-systems and

it may carry certain elements of language coding (cf. Damerow 1999:

7, see http://www.mpiwg-berlin.mpg.de/Preprints/P114.PDF). The

entries of a proto-Elamite text can cover all surfaces, and can run

from one line to the next and from one surface onto the next. Each

entry consists of a string of graphemes and a numerical notation.

Most of the strings of signs in the proto-Elamite corpus are of

modest length (2 - 6 signs), but some longer strings exist. The

header and subscript is not followed by a numerical notation. In a

transliteration of a proto-Elamite text each entry is given its own

line-number, and its two parts are separated by a comma.

 

NEW WORK ON THE PROTO-ELAMITE TEXTS:

After collating the entire Louvre holdings of proto-Elamite and

cleansing the sign-list, renewed graphotactical tests have produced

some very interesting patterns. We can for example take a look at

the

following texts (links to on-line images inserted, old publications:

caution!) MDP 6, 204

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P=P008004); 225

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008024); 253

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008048); 337

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008123); 353

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008135); MDP 17, 93

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008291); 350

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008548); and 463

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008661). They all agree

with the same format. The first sign is a header, presumably

indicating the institution to which the following items or persons

belong. Next we find a short list of entries each consisting of a

string of " ideograms " followed by a counted object (M346 = sheep?)

and a numerical notation. The considerable length of the strings,

and

the fact that these strings can be divided into two different

segments that can be found in different constellations throughout

the

corpus is suggestive of an interpretation of these as " spelling " the

title and the name of the " owner " of the counted object.

Of course my " evidence " for this is more than I mentioned here, let

me know if I shall expand on this.

 

Now, it becomes really interesting when we start looking signs used

in different positions in the strings, since the signs used in the

initial positions seems to be used only there, the signs used for

writing the owners only used there, etc. This is of course very

preliminary, but it does points towards a more complex writing

system

than previously believed. Let me note, however, that most of the

texts, most, shows no sign of this system, but remains what Damerow

has called a proto-writing system

(http://www.mpiwg-berlin.mpg.de/Preprints/P114.PDF). It is possible,

however, that this differentiation pertains to different stages in

the writing system, i.e., that later texts have a more elaborate

structure. Consider for example the two following texts, MDP 17, 182

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P=P008380) and MDP 17, 151

(http://cdli.ucla.edu/cgi-bin/Pget.pl?P=P008349), they are in fact

two documents with almost exactly the same content (both presumably

recording a number of animals (goats), and by-products from a herd

of

goats). However, MDP 17, 151 has a much more evolved structure than

MDP 17, 151, and some of the signs have evolved too (physically too,

the tablets belong to different groups). It is particularly the

initial string of MDP 17, 151, presumably identifying the owner of

the animals which is much longer than the similar string of signs in

MDP 17, 182. In the latter text, the ownership identification is

taken care off with one sign (following the sign for nanny-goat). In

a couple of days I will have a preprint on-line of an article that

briefly discuss this.

I guess this is something I should return to in another post?

 

FUTURE WORK:

What is decipherment? That is a question I would like to put to this

list.

Consider the following, for example, I have recently worked on a

group of animal-herding texts, consisting solely of counted objects

and numbers, and perhaps owners marks (see for example MDP 17, 96 =

<http://cdli.ucla.edu/cgi-bin/Pget.pl?P= P008294>) (see also the two

texts quoted above). And I think I have sorted out Susa animal

terminology; the sign for sheep, goats, juveniles, and males and

females, etc. Of course I have no clue as to how these words were

pronounced by the scribes, but I still think it counts as a

decipherment.

Any thoughts?

 

Is it possible to prove that the use of proto-Elamite was indeed

very

shortlived? And if so, what are the implications? I for one think it

is quite possible that they " invented " their writing system and used

it for a short while where after it was abandoned, and lived happily

the next many centuries without using a writing system, but I may be

wrong.

 

 

The following three points are those I find most important:

1. The use of writing, administration, archives, schools

2. The early development of writing, invention of new signs, complex

graphemes (I will have an article out about this in a few days in

the

CDLJ as 2005:03 see http://www.cdli.ucla.edu/pub.html), and

standardization of sign repertoire (existence of lists, etc.), etc.

3. Methods of decipherment that fit the systems of (proto-)writing.

 

BE aware that much of the argument here cannot be proven without

access to the original texts (I am working on bringing these

exciting

texts out to you all in new editions), and that much come from

finished or almost finished work that I am just about to publish. I

really hope for some input, " deciphering " proto-Elamite is a rather

lonely profession!

 

Regards,

 

Jacob Dahl

CNRS, Paris