accusative and ergative languages

[Ralf-Stefan Georg]

Again without stage-directions:

>How clever to respond in this way! As if I was asking you to subscribe to
>Klimov's ideas simply because he is an eminent linguist!

Well, it *did* sound like that, or why pointing out Klimov's reputation so
verbosely ? I mean, if I agree with someone on a particular point, it
really doesn't matter whether that person is Roman Jakobson or my local
fishmonger. I agree, because I do, and any consequences of this I will have
to face myself. Sapere aude.

So, it would be enlightening to learn *what* the *specific* lines of
Klimov's argumentation are, you agree with. So far, we have only heard the
bottomline and the fact that Klimov was a deservedly prominent linguist.
Where's the beef ?

>I was simply pointing out, since you obviously missed my point, that highly
>qualified linguists do disagree; and so, your opinion (and even,
>occasionally, the consensus) may or may not be found ultimately correct even
>as the "consensus" once firmly rejected the laryngeal theory in any form.

That's a truism. What is interesting, is which particular things about the
synchrony and diachrony of ergative traits in observable languages
allow/force/disallow/prevent us from daring a determinist statement as that
which you brought forward.

The analogy you mention is irrelevant. While it is true enough that today's
communes opiniones once were universally rejected, the only thing which
follows is that we should use the notion of "truth" sparingly. What does
not follow is, that every communis opinio of today will of necessity be
debunked one day. But you carefully evade the task to expose Klimov's (or
your, where you differ) line of argumentation.

>QED. Just what do you believe your proved?

That Sumerian is just another split-ergative language, since you seem to
doubt my (oh, not *my*, I could refer you to *eminent* linguists ;-)
assertion that, while non-split ACC-languages do exist, non-split
ERG-languages are not known. You asked for the split in Sumerian, I gave it.

>And, I would like to ask you a question in view of your snide aside about a
>"non-linguist observer". Is it your opinion that no one is entitled to be
>considered a linguist, even an amateur linguist, if he/she does not possess
>a PhD in Linguistics?

No, this is patently not my opinion, especially since I was an excellent
linguist even before I was handed over my PhD diploma, which rules out this
possibility  ;-) ;-) (< --- see these !).

It is, however, true that not every scholar who took part in the
advancement of our knowledge of Sumerian, could be classified as a linguist
in the modern sense. This is perfectly OK with most of them, I'm pretty
sure, no offense intended. Linguistics would not be without the great
philologers. The same holds for other disciplines as well, where modern
linguistics (especially typological linguistics) sometimes has the right
and duty to "correct" (better: adjust) the findings of the great
philologists/grammarians (or better: not what they *found*, rather how they
interpreted it). And it is certainly true that notions of ergativity, not
to speak of split-ergativity, did not play a major role in the earlier days
of Sumerology, i.e. well into this century, let alone the cross-linguistic
typology of these phenomena (yes, I am aware of truly linguistic treatments
of Sumerian, which do exist). So, I would not be disparaging, say,  A. von
Gabain calling her Alttuerkische Grammatik the work of a non-linguist. It
is, and she knew it, and nevertheless it is a gold-mine for any linguist
working on Old Turkic, I know of no linguist who could say different things
about it, but the linguist's task is different from that of the pioneer
philologist and grammar-writer (and, of course, linguistics is a discipline
which sometimes makes progresses). Hope this makes that clear.
Anyway, I'm happy to accept the label of "Non-Sumerologist" for myself. Are
you a Sumerologist ? (Gonzalo, are you still with us ?)

>As for your characterization of the Sumerian imperfective system, which is
>properly called the maru: inflection *not* imperfective, just what
>characteristics do you *believe* it has that qualify as ACC?

Some Japanologists of my acquaintance (and some text-book authors) promise
to kill everyone who dares to speak of a "verb" in Japanese in their
presence, since these things are, of course, "properly called" /dousi/.
What is this about ?

The idea of typological linguistics is to *compare* languages, resp. their
structural makeup. In order to be able to do this, exotistic terminology is
best avoided. True enough, perfect matches between verbal (or other)
categories among languages in terms of their functions are rarely
encountered. However, to ensure comparability (with the usual disclaimers)
a modernization of terminology is always to be wished for. I can and will
object to your objection if, and only if, you (or someone else) will point
out why this "maru:"-inflection may, under no circumstances, be regarded as
a verbal category encoding imperfective aspect, as opposed to perfective
aspect encoded by the HamTu-inflection (you see, I am myself a great
"terminology-dropper").

You highlighted *believe* in your above response. Well, just as a small
philosophical aside, of course everything which we think to *know* is
really something we *believe*, but we may *believe* some things with
slightly greater confidence than others, iow., that's what science is
about. If you want to discuss this further, let's transfer this to the
radical-constructivism-list.

What makes me *believe* this, is data like the following: while it seems to
be clear that case-marking (and some people think that ergativity is only
about case-marking; however, I think highly enough of you not to assume
that you are of this lot; moreover, after Wolfgang's postings here, noone
else should be) in Sumerian operates ergatively regardless of TAM category,
verbal cross-referenceing doesn't, it is just this which makes up for an
ACC residue in the language. To wit:

lugal-le Hi-li ib2-dim2-me. "The king fashioned the wig"
The king is case-marekd as ERG (-(l)e), and the wig is ABS, so, in terms of
case-marking a perfect ERG construction.
The verb form, here given in transliteration, is morphologically to be
analyzed: i- (conjugation prefix for maru: or imperfective (horribile
dictu), -b- "personal affix for third person *inanimate*", so patently
cross-referencing the patient here (cross-referencing the king would
require the animate PA -n-), -dim- "The Root", -e imperfective suffixe (or
maru:-suffix, now you be quiet ;-).
Now, let's look at an intransitive imperfective sentence:

lugal i3-du. (case-marking is of course not ergative, i- the maru:-marker).
In order to have an ergative organization of verbal cross-referencing of
constituents, we should expect the *patient* in the transitive sentence
above, being treated *the same way* as the intransitive agent (some prefer
"subject") in the last example, i.e. by being cross-referenced as -b-. It
isn't. Actually it isn't overtly  cross-referenced at all, though some
Sumerologists prefer to insert a zero-affix cross-referencing the agent
here at the end of the suffix-chain. Bog s nimi.

The bottomline, the imperfective/maru:-system shows ACC-verbal
cross-referencing by virtue of treating transitive patient and intransitive
agent/suffix *differently*, the very gist of the definition of ergativity,
which, I hope, I won't have to rehash here.

Note, for completeness' sake, that this state of affairs doesn't repeat
itself in the perfective/HamTu-conjugation.
And, being myself, a stubborn: QED.

>I am also puzzled by your idea that Sumerian pronouns "operate on a fully
>ACC basis" since , e.g. the 1st and 2nd persons ergative g[~]a[2].e and za.e
>contrast with 1st and 2nd persons absolutive g[~]a[2] and za in the same way
>nouns show an ergative in -e and an absolutive in -0. Perhaps you could
>explain your ideas in greater detail.

It is true that g[~]a[2].e and za.e are formally ergative cases, by virtue
of -e. However, I'm unaware of a systematic contrast between ergative and
absolutive forms (i.e. without -e) used in a clear-cut ergative way in the
language. No doubt this reflects my superficial knowledge of it. Various
sources assure me that what seem to be "absolutive" forms  g[~]a[2] and za
are late Sumerian, and explicable as phonetically expectable reflexes of
the longer (and earlier) forms. Even then, they are used in ERG and ABS
functions indiscriminately, like the longer ones before.

It would help your case if you could demonstrate with text examples that
g[~]a[2].e and za.e are confined to ERG function, or better, that  g[~]a[2]
and za are, in Classical Sumerian (2600-2300 BCE) used in ABS function,
i.e. as intransitive subjects and patients of transitive verbs. I. for one,
don't know whether this is the case, but you seem to know, so it should be
legitimate to ask you for examples. Until they come forth (in which case I
will give this up happily), I will take this phenomenon as the second
instance of an ergativity split in Sumerian, though admittedly the first
one mentioned is the stronger one.

To conclude: I stand by my "belief" that there is no such thing as a fully
ERG language, i.e. one without any splits, as opposed to fully ACC
languages. Sumerian is no counter-example. Any takers ?

St. G.

Stefan Georg
Heerstrasse 7
D-53111 Bonn
FRG
+49-228-69-13-32