3 Apr 2008

Cool stuff about Etruscan phonotactics

While everyone else is hohumming about how the Etruscan language is a big, unsolvable mystery, I'm more interested in finding out new information and learning more. You don't need a degree to extrapolate and develop a good theory. You just need a brain and hopefully the one you're born with because they don't hand them out during graduation ceremonies. So I shamelessly ponder on the Etruscan language myself, and after having made my own language database on my computer, I'm starting to see new patterns in how Etruscan words are shaped.

One pattern that I saw long ago was the "intrusive y". You can see this, for example, by comparing the word śa 'six' with śealχ 'sixty' (TLE 98: śealχls-c 'and of sixty'). Notice the vowel change here? The reason why a changes to e is not by random. The vowel e is often derived from the Old Etruscan diphthong ai. So when also comparing Late Etruscan śealχ with the Lemnian root sialχv- present on the Lemnos Stele (sialχveiś 'of sixty'), we may postulate earlier *śa-i-alχu, showing clearly an intrusive glide between the terminating vowel of the root and the initial vowel of the decadic ending. It appears to me that this intrusive y pops up any time two non-high vowels would otherwise clash in a derivative form and is probably related to the fact that words avoid initial /y-/ altogether in all Proto-Aegean languages, including Minoan[1].

Then there's the nifty issue of initial consonant clusters in Etruscan. The language doesn't allow just any cluster to occur at the start of a word. Rather, the only legal initial clusters seem to be SR-, sSR- or sC-, where R is any of m, n, r , v or l, S is any stop, and C is any consonant in general. The words tmia 'temple' (PyrT 1.i), streteθ (LL 6.iii), and sren 'image' (TLE 399) are examples of the three possible types of word-onset clusters.

Just recently however, I noticed that the words in my database are showing me something else. It looks like Etruscan words have a resistence towards initial clusters with aspirated stops. So far, I've found no clusters of the SR- type with φ or χ even though there's nothing phonetically implausible about such clusters. Occasionally though, I have found the aspirated dental stop in just a few clusters (e.g. θresu in TLE 222). Here again though, it's rare which then makes me think that there's an underlying resistence to aspirated stops in Etruscan clusters. Just thought I'd share that with yo'll because I think that all these details might be useful to someone out there.


NOTES
[1] That is to say, the symbol used for A in the Minoan Linear A writing system and the symbol used for YA were used interchangeably in word-initial position by scribes. This can be seen in the word or name YA-SA-SA-RA-ME inscribed on libation tables which is written also as A-SA-SA-RA-ME. See for example Interaction and Acculturation in the Mediterranean, edited by Best and de Vries (1980), p.160 (see link).

UPDATES
(April 3 2008) Just after posting this, I thought of another letter that occurs in the R slot, namely v /w/ (as in tva 'it shows'). So I changed "R is any of m, n, r or l" to "R is any of m, n, r , v or l".

6 comments:

  1. Hi, Glen. Since you're well acquainted with Etruscan phonotactics (certainly much better than me), I'd like to ask you a few questions.

    1) As for the aspirated stops in clusters, I think there are typological parallels in other languages. For example, in English, or some varieties of it, at least, /p/, /t/, /k/ are not aspirated when they are preceded by /s/ (cf. top, pot, Kate vs. stop, spot, skate, etc.). Some phoneticians say this is due to the "aspirated" nature of /s/.
    On the other hand, /r/ and /l/ in "trail" and "play" are different from those in "rail" and "lay", aren't they? So, I wonder if it could be due to assimilatory or accommodatory changes.

    2) As for the shape of noun stems in Etruscan, what is the most frequent or typical pattern? I'm especially interested in the medial clusters. Since loanwords are often borrowed in a form that conforms to the native phonotactic canon, I wonder what shape a medial cluster of the /-bt-/ shape would take and whether it is allowed in Etruscan at all. Also, I'd like to ask what origin you suppose for the Etruscan /-mp-/ and /-mpH-/ clusters.

    Of course, I'd try to find that if I had some free time and if the lexicon you have been collecting were in form of a database (DBF, EXCEL, whatever).

    Anyway, thanks a lot for any response!

    Best,

    Petr

    ReplyDelete
  2. Peťusek: "1) As for the aspirated stops in clusters, I think there are typological parallels in other languages."

    I think it might have to do with how aspiration is released in such clusters. It probably makes sense that both θm- and tm- would be phonemically neutralized to /tm-/ since the resultant phonetic difference between θm- [tm̥ʰ-] and tm- [tm̥-] is minute (nb. aspiration being released with the trailing resonant, not with its preceding stop) and aspiration in solitary resonants themselves is non-distinctive in Etruscan anyway. So in a nutshell, the resonants would have motivated the neutralization.

    Peťusek: "Since loanwords are often borrowed in a form that conforms to the native phonotactic canon, I wonder what shape a medial cluster of the /-bt-/ shape would take and whether it is allowed in Etruscan at all."

    In Etruscan, /bt/ itself can never be allowed since there are no voiced stops in this language. It would no doubt be naturalized at the very least to /-p.t-/ or /-pʰ.tʰ-/. I use the period here to present syllable division within the cluster. (Note: Tautosyllabic stop-plus-stop clusters are impossible in Etruscan.)

    Peťusek: Also, I'd like to ask what origin you suppose for the Etruscan /-mp-/ and /-mpH-/ clusters."

    For the clusters themselves or for the origin of the word? I believe that word-medial nasal-plus-stop clusters were always native to Etruscan and its prestages. However the numeral semɸ "7" is without a doubt an Afro-Asiatic loanword.

    ReplyDelete
  3. Peťusek: Also, I'd like to ask what origin you suppose for the Etruscan /-mp-/ and /-mpH-/ clusters."

    Glen: For the clusters themselves or for the origin of the word? I believe that word-medial nasal-plus-stop clusters were always native to Etruscan and its prestages. However the numeral semɸ "7" is without a doubt an Afro-Asiatic loanword.

    For the clusters themselves. Especially, whether they could arise from non-labial clusters in any way (via effects like, say, affective nasalization, etc.).

    As for the numeral "7", by Afro-Asiatic you mean Proto-Afro-Asiatic or its daughter (such as PSem *sábʕ-u(m) or PEg *safḫaw)? What shape do you presuppose for the source?

    Would the pharyngeal have caused the nasalization (I've read nasalization may develop from pharyngeals and pharyngealization diachronically), or was it the PSem final *-m via regressive assimilation (and subsequent loss of the segment itself), or was the vowel in the second syllable syncopated and the newly arisen cluster metathesized (of course, with the necessary devoicing of *b)...?

    ReplyDelete
  4. Peťusek: "As for the numeral '7', by Afro-Asiatic you mean Proto-Afro-Asiatic or its daughter (such as PSem *sábʕ-u(m) or PEg *safḫaw)?"

    No, I meant "Afro-Asiatic" as in the Afro-Asiatic family of languages. (Proto-Afro-Asiatic itself dates to at least 10,000 years B.P. and so has nothing to do with Etruscan!) Specifically, the likeliest candidates for the donor language of the Etruscan numeral appear to me to be Ugaritic and Middle Egyptian.

    Peťusek: "What shape do you presuppose for the source?"

    You mean, what do I think was the Pre-Etruscan form at the time of borrowing? Proto-Etrusco-Cypriot *sempʰa circa 1200 BCE.

    ReplyDelete
  5. Glen: No, I meant "Afro-Asiatic" as in the Afro-Asiatic family of languages. (Proto-Afro-Asiatic itself dates to at least 10,000 years B.P. and so has nothing to do with Etruscan!)

    I thought so. I was just making sure you hadn't redone your hypothesis completely. I've seen more shocking things on the net! ;-)

    Specifically, the likeliest candidates for the donor language of the Etruscan numeral appear to me to be Ugaritic and Middle Egyptian.

    However, this is a problem. Neither Middle Egyptian (/sfḫw/, /sfḫwt/), nor Ugaritic (/šbʕ/ < */šabʕ/, */šabʕat)/ have */m/ before */b/.

    Peťusek: "What shape do you presuppose for the source?"

    Glen: You mean, what do I think was the Pre-Etruscan form at the time of borrowing? Proto-Etrusco-Cypriot *sempʰa circa 1200 BCE.

    No, I mean what form did the Semitic/Egyptian numeral have? Neither Middle Egyptian nor Ugaritic explain the */m/ in Proto-Etrusco-Cypriot */sempʰa/. From the little I know about Etruscan phonotactics (little indeed), I would expect it to borrow the numeral as **/sepʰa/, or something similar. Why would Etruscan add */m/? Was it regular? Was it expressive? Was it for fun?

    ReplyDelete
  6. Peťusek: "Neither Middle Egyptian nor Ugaritic explain the */m/ in Proto-Etrusco-Cypriot */sempʰa/."

    Considering so many words with word-medial -mɸ- (eg. hamɸeti) or -nθ- (eg. hanθe) it almost seems to me like there was some sort of early phonotactic pressure to introduce *m or *n before what would otherwise be an intervocalic aspirated stop.

    ReplyDelete