7 Apr 2008

Defining valid Etruscan word-initial clusters

Okay, long story short, it's true what they say: Bad things come in threes. One of the problems lately has been that my internet connection has been for some reason impaired and I will have to get that fixed. The cable gods must be angry. So I'm here now at a library making my next blog entry. Other crap has happened at the same time and maybe some day I'll share that with you all. At any rate, let's get back to the last subject I wrote about: Etruscan phonotactics.

Etruscan phonotactics are intriguing to me because it's the kind of details that never ever show up in current books on the Etruscan language. Most authors seem to be effectively hijacked by an overhyped mystery of Etruscan civilization to be capable of moving forward and coming up with new questions to resolve and explore with the reader. I feel as though the available reading materials are impoverished and I'm starving for something new. So I'm treading new ground here and I realize in hindsight that my definition of these onset clusters isn't complete. None of you piped up to correct me either! This saddens me but nonetheless at least I'm interested in these nagging details and my devil's advocate is fully oiled.

I feel compelled to expose glaring counterexamples to my previous claims such as fler 'offering' and the rather uncommon gentilicium Fnesci which I failed to account for. In these instances, we have a bilabial fricative followed by a resonant. While rarer than word-initial stop+resonant sequences, such clusters appear to be legal formations in Etruscan. Certainly the cluster fn- is so unusual that it emphasizes my point that at some point in a Pre-Etruscan stage, accent must have fell at times on the second syllable and then had succumb to syncope to produce these and other fascinating clusters (i.e. the word tmia 'temple' also comes to mind). So it seems that we need to expand these rules a bit before they're acceptable.

Let's then restate the rule as the following and see whether this sits well on my conscience: Valid word-initial clusters in Etruscan are either of the form FR-, sS- or (s)SR- where F = fricative (/s/, /ɸ/), R = resonant (/m/, /n/, /l/, /r/, /w/) and S = stop (/p/, /t/, /k/, //, //, //). In this way, lexical items like sren, fler and Fnesci are covered by FR-; staile and scuna are covered by sS-; and finally tmia, tnucasi, tleχe, tva, clen, θresu, and streteθ are covered by (s)SR-. Does this sound good? Speak to me people! Share questions.

10 comments:

  1. I know nothing about Etruscan. But something that comes to mind is: how do you that something like "fn" or "tm" aren't contracted spellings for something else?

    ReplyDelete
  2. Seems right, but I'm no expert on Etruscan.

    I would have to agree with the notion of the syncopation to allow such bizarre clusters, there aren't many other things one can come up with to account for the emergence of such clusters.

    ReplyDelete
  3. Glen,

    Don't know how to contact you directly but I am a student of linguistics and I happen to be doing some work on grammatical case. Wondering if you had any information on the case system used in Proto-Semitic. Worlds of information is to be found on the grammar of biblical Hebrew but I want to look beyond that and encompass the related languages of Northern Africa (e.g. Ge'ez).

    Anything you got will be helpful.

    By the way I really enjoy your blog.

    ReplyDelete
  4. I'm going to answer everyone in one big lump. Here we go! :)

    Goofy: "But something that comes to mind is: how do you that something like "fn" or "tm" aren't contracted spellings for something else?"

    Something else? Like what? The only conceivable thing it could represent besides a cluster is a consonant and syllabic resonant. However, I don't think there's reason to think this. There are living languages with clusters like this today. Note Klallam sqməy̕ 'dog' and Greek πνευμα pneuma 'breath'. In fact, even French retains strange clusters of the pC- variety from Greek loanwords such that psychologue 'psychologist' is actually pronounced with initial /ps-/ and pneu 'tire' is vocalized as it is spelled, /pn-/.


    Phoenix: "I would have to agree with the notion of the syncopation to allow such bizarre clusters, there aren't many other things one can come up with to account for the emergence of such clusters."

    Yep, it seems pretty clear-cut here. One issue I still need to resolve though is where Etruscan f comes from. My spidey senses are telling me that it didn't exist in previous stages of the language. Many instances of f seem to be attributable to earlier p neighbouring tautosyllabic u but it doesn't explain all instances. So I'm not sure whether I'm trying to fit a round peg into a square hole or not. We'll see.


    Zach: "Don't know how to contact you directly but I am a student of linguistics and I happen to be doing some work on grammatical case. Wondering if you had any information on the case system used in Proto-Semitic."

    Regardless of whether you're a university student or a passionate hobbyist, there should be plenty of literature at one of the libraries at your local university. Proto-Semitic is one of the more popular protolanguages studied around the world so you shouldn't have trouble finding anything on its case system.

    For starters, I recommend Louis Gray's Introduction to Semitic Comparative Linguistics. And even if you have trouble finding that book, parts of it are available online through Google Books. Click here for the topic of declension.

    If you really need to get a hold of me, my contact address is available through any of my draft pdfs of my ongoing Etruscan dictionary project.

    ReplyDelete
  5. What I meant was, what if the rather uncommon word Fnesci is a contraction, like say "Mrs" or "Esqu".

    ReplyDelete
  6. Can you provide the complete inventory? I took a look at your dictionary but that's obviously just a subset of the entire corpus, and seems to eg. miss the sSR formula entirely. Might be interesting to try & see if there are any non-random-looking holes in the system. Eg. are there any /pm pw/ (two labials)?

    ---

    As for what I can see now, the system looks remarkably similar to what IE languages tend to have, especially Greek... areal influence, probably? But I notice at least one difference: the presence of /tl/. I can't think of any IE language that has coronal + /l/. Not sure what, if anything, that signifies.

    ---

    Metathesis (as in Slavic) would be another way for initial clusters to arise. I haven't studied Etruscan either, however, so I can't comment if that might fit; but if the system lacks any RR combos, probably not at least as a recent development. Hmm... are there any other mecanisms? Prefixes?

    ReplyDelete
  7. goofy: "What I meant was, what if the rather uncommon word Fnesci is a contraction, like say 'Mrs' or 'Esqu'."

    What makes what-ifs so entertaining and so insidious at the same time is that they are limitless. Of course this is possible, especially considering that Etruscans did make use of abbreviations (normally for the praenomen of the deceased in funerary inscriptions).

    However, beyond an idle possibility, I don't see what more we can do with this thought. It's not considered a title at all, but a name. I feel that it's a name too.

    ---

    Tropylium: "Can you provide the complete inventory?"

    How do you mean? I haven't keyed everything in yet and am still working on existing words. If you have nothing to do, why don't you be a sport and dig for counterexamples? I can't be expected to do everything, you know :)

    However the sSR- pattern is really quite rare. I can only think of streteθ in Liber Linteus VI as an example. I don't believe I have that listed in the pdf yet because I'm not sure what meaning to give it, being that it's a curious little hapax legomenon.

    Tropylium: "Might be interesting to try & see if there are any non-random-looking holes in the system. Eg. are there any /pm pw/ (two labials)?"

    Yes, holes make my devil's advocate giddy with glee but I've honestly never encountered these two clusters in Etruscan thus far.

    Tropylium: "As for what I can see now, the system looks remarkably similar to what IE languages tend to have, especially Greek... areal influence, probably?"

    I sometimes wonder, yes.

    Tropylium: "but if the system lacks any RR combos, probably not at least as a recent development. Hmm... are there any other mecanisms? Prefixes?"

    Prefixes are unusual for languages with an SOV word order. Etruscan appears to have none, and only suffixes. When syncope as I propose here occurs in a language, it's not unusual for additional tactics to be employed to avoid phonotactically invalid clusters like *RR-. In fact, a cluster with two liquids seems to be universally avoided in world languages. It would be great for an alien language, though :)

    ReplyDelete
  8. Just a few points regarding Tropylium's comments, if I may, as well as yours, perhaps...


    1) As for the /tl-/ cluster purportedly "unattested" in any IE language, it is quite common in the Czechoslovak dialect continuum (e.g. [tlak] "pressure", [tlama] "muzzle, jaw, mouth", [tlapa] "paw", [tlEn^i:] "rot", [tli:t] "to rot", [tlOut_sI] "to beat", [tlumEni:] "subdued", etc.). Of course, most of them have arisen via the metathesis of liquids (PSlavic *tVlC- > *tlVC-).


    2) As a matter of interest, Czech has some complex initial clusters, which you might consider "bizzare", but they are normal in my mother tongue, e.g.:

    [pstrUx] "trout"; [pStrOs] "ostrich"; [fspr^i:t_SIt] "to jam"; [fskr^i:sIt] "to resurrect"; [fstr^Ebat] "to absorb"; [fskvE:tat] "to flourish"; [fsprUz^It] "to pep"; [fsplanOut] "to kindle"; [fspjEra] "sprag"; [fskli:t_SIt] "to germinate"; [vzHli:ZEt] "to look up"; [pSka:t] "to fart (formal)"; [bzd^i:t] "to fart (formal)"; [tr^t^Ina] "(sugar) cane"; [t^pIt] "glitter"; [kr^t^InI] "baptism"; [dSt^i:t] "to spout"; [kSt^it_SE] "thatch of hair"; [mdli:] "faint (adj)"; [lpjEt/ "to cling"; [t_SpjEt/ "to stink"; [rdOusIt/ "throttle"; [vzlIkat/ "to sob"; [svrab] "scab"; [rmOut^It] "to afflict"; [mrak] "cloud"; [lnOut] "to adhere"; [mlE:kO] "milk"; [mnOut] "to whet"; [mn^Elki:] "shallow"; [jmE:nO] "name"; [ln^Eni:] "linen (adj.)"; [mr^i:Z] "trellage"; [Stka:t] "to sob"

    Also, consider the cluster [-str^t_s-] in [OstrOpEstr^t_sE] "thistle (gen.sg.)" (which is a real tongue-twister for me as well).

    All those clusters have arisen via syncope (of what once were jers) by the way.

    ReplyDelete
  9. Petusek: "As a matter of interest, Czech has some complex initial clusters, which you might consider 'bizzare', but they are normal in my mother tongue"

    Czech?! "Bah!" I say! In the grand scheme of things, Czech is far too soft and cuddley for my liking! Nothing and I mean nothing beats a Salish language for fear-inducing word-initial clusters!

    However, back to Proto-Indo-European itself, one might also consider the instrumental suffix *-tlo-.

    ReplyDelete
  10. Alright, alright, Klallam is a little more difficult to speak, but in comparison with English, Czech IS somewhat more complex. ;-)

    ReplyDelete