Mecislau wrote:Generally, the language is completely intolerant of word-final consonant clusters in native words, but the 2sg fem form of the past tense is formed with the ending -t, which usually ends up forming a cluster.
Out of curiosity, what happens when you apply a suffix pronoun to the verb? I know we avoid this problem in Levantine Arabic by inserting an epenthetic vowel in between the verb and suffix(es):
ḍarab-(a)t
hit-you.MSG "you strike"
ḍarab-t(i)-nī
hit-you.MSG-me "you struck me"
Mecislau wrote:In fact, for verbs I can list them all: PS *mV- (nominalizing), *nV- (detransitive), *tV- (reflexive), *šV- (causative), not counting personal agreement affixes.
As a fiddly little aside, all of these morphemes show up in other branches of AA as well. Probably at the time of PAA, these weren't quite bound morphemes yet as 1) they often show up
suffixed to the verb in Cushitic and 2) the prefix/infix *t-/*-t- shows up in Egyptian as both an impersonal pronoun (similar to french
on) as well as the subject of "passive" verbs:
*radj.n.tw n.j ḥtp.w.t (probably pronounced something like: */laṭī.ni.tuwa n.ī ḥatap.aw.at/ )
give.PERFECT.one to.me offering.PL.FEM: "one has given offerings to me" or "the offerings were given to me"
*radj.n.tw n.j ḥtpwt jn wʕb (probably */laṭī.ni.tuwa n.ī ḥatap.aw.at ʔan waʕʕāb/ )
give.PERFECT.one to.me offering.PL.FEM by priest: "the offerings were given to me by the priest"
Furthermore, at least for Semitic, we can divide nominalizing *m- into *ma- for place and time, *mi- for instrument and *mu- for relatives (forms participles). I don't know how much of this might be obfuscated by Hebrew's fiddly vowel system.
That is, if the stem vowel is /i:/, the prefix vowel should be /a/; if the stem vowel is /a:/, the prefix vowel should be /i/. You can still see this in Hebrew:
Interesting. Arabic is quite different in this regard. The stem voweling is determined by the verb's aktionsart. "Permanant states" take 'u' in both the perfect & imperfect; كبُر /kab
ura/ -> يكبُر /yakb
uru/ "he got big -> is getting big", سمُنت /sam
unat/ -> تسمُن /tasm
unu/ "she got fat/ is getting fat" while "temporary/mental states" take 'i' in the perfect and 'a' in the imperfect; فهِم /fah
ima/ -> /yafh
amu/ "he understood/ understands" and transative verbs take 'a' in the perfect and either 'i' or 'u' in the imperfect كتَب /kat
aba/ -> يكتُب /yakt
ubu/ "he wrote/writes", ضرَب /ḍar
aba/ -> يضرِب /yaḍr
ibu/ "he hit/ hits". Meanwhile, the
prefix vowel is determined by the verb's
wazn. So, for example, all
faʕala (base, Form I) verbs take 'a' in their prefix (as seen above), while
faʕʕala (intensives, Form II) always take 'u' in their prefix; y
ukabbiru "he is enlarging"
So far as I know, this feature dates back to PS, as it shows up in both East Semitic and South Semitic as well. I've seen it argued that the perfect forms were actually a verbal noun + pronoun; compare the PS passive participle forms *paʕūlum/*paʕīlum and the PS infinitival form *paʕālum. However, I've seen people attribute Barth's Law to PS, so at least in that case, Hebrew may be more conservative (my memory on this is rather fuzzy at the moment, though).
TomHChappell wrote:Do any tetraconsonantal roots come from compounds of two biconsonantal roots? Do any quinquiconsonantal roots come from compounds of a biconsonantal root with a triconsonantal root, in either order?
This just reminded me of another form of compound in Arabic, called النحت an-naḥtu "carving/sculpting". It's used when elements of two or more words are combined to refer to them all. Off the top of my head:
x بسمل /basmala/ means "he said 'in the name of God'" from بسم الله /bi-smi-llāhi/ "in the name of God".
x سمعل /samʕala/ means "he said 'peace upon'
someone" from السلام عليكم /'as-salāmu `alaykum/ "peace upon you".
x الحوقلة /ʔal-ḥawqalatu/ refers to saying لا حول ولا قوة إلا بالله /lā ḥawla wa-lā quwwata illā billāhi/ "there is no power and no might but by means of God".
x المشألة /ʔal-mašʔalatu/ describes saying ما شاء الله /mā šāʔa-llāhu/ "as God wills".
As you can tell, many of these are religious in nature and were coined in the classical period. I'm pretty certain that there are more modern/seculare ones, but for the life of me, I can't think of any right now. There's actually a bit of science to this; these new words can't violate Arabic phonotactics - ie. they can't contain both س /s/ and ز /z/ as these never appear in a native root together (there are numerous other frobidden combinations, as I recall). I also believe they have to have at least one 'liquid' letters - ر /r/, ل /l/, م /m/ & ن /n/.
TomHChappell wrote:Lots of stuff about roots & phonotactic restrictions.
Again, medieval Arabic scholars spent a *lot* of time looking into this stuff. Unfortunately, most of what I know of this I read years ago, and all of it was written in Arabic. However, I'm pretty sure there are translations/modern studies written in english. I'll let you know if I come across any. If worse comes to worst, I suppose I could see about translating something. Again, this is interesting to me, but I unfortunately can only devote so much time to this.
With respect to your questions about syllable structure and consonant clusters, I can say that Arabic only allows clusters medially. Initial clusters are broken up with epenthetic vowels + glottal stop (hamza), like the استفعل
ʔistafʕala verbs mentioned above. Most modern Arabic dialects break up the final clusters that resulted from the loss of case/mood with anaptactic vowels: CA kalb-un -> *kalb-0 -> PA kal
ib. Basically, Arabic only allows CV, CVC, CV: or CV:C *if* (and only if) the second consonant is a geminate (CV:C:V).
Mecislau wrote:However, the majority of any such trends are most likely just historical accident.
Again, I'll disagree here (although I might be misreading what you mean by "accident"). I really wish I were better versed in this, but I can say with some confidence that there were definite restrictions in PS and PAA on what kinds of consonants could be in a root at that same time and what positions they could take.
Mecislau wrote:Final {h} not resulting from an earlier *w or *j is also very rare if not nonexistent.
There are actually a fair number of roots that end in /h/ in Arabic. Two that immediately spring to mind are W-J-H, as in وجه /wajhun/ "face" and perhaps more (in)famously Š-B-H, as in شبه الجزيرة العربية /šibhu-l-jazīrati-l-ʕarabiyyati/ "The Arabian Peninsula" (lit: the arabian semblance of an island).
Mecislau wrote:While I can't cite any examples off the top of my head, neighboring consonants (ie, C1 and C2, or C2 and C3) will often acquire the same pharyngealization or voicing state as the other (if possible).
This happens quite regularly in Arabic. For example, the -t- infix of
iftaʕala verbs takes on either voicing or emphasis (but not both) from on adjacent consonant: زهر /zahara/ "it shone" -> ازدهر /iz
dahara/ "it bloomed", ضرب /ḍaraba/ "he struck" -> اضطرب /iḍ
ṭaraba/ "he got agitated". There are other such changes/assimilations, but I think you get the idea.
Not very common. The only real infix is the -t- of the appropriately-named t-Stem (Hebrew hitpa`el, Arabic 'ifta`ala, etc), which is not universal. This is simply a case of a former prefix undergoing metathesis with the following consonant, either in all cases (as in Arabic) or in certain cases (as in Hebrew, where it only happened before /s š z ṣ/).
There actually are more, Akkadian verbs had a several 'awzan/binyanim that could combine two infixes (-ta- & -n-) + any of the prefixes described earlier (š-, n-, etc.) + a seperate infix, -ta- used to form the perfect. You could actually end up with surface forms along the lines of
*iptatarras (for *yap
tatanras) meaning something like "he had been constantly deciding" from the root P-R-S. The gemination present in Arabic and Hebrew "intensives" is almost certainly the result of assimmilating an earlier nasal infix (discussed earlier up-thread). There are Arabic forms that involve infixes besides, ie - ifʕa
wʕala, ifʕa
wwala, ifʕa
nlala, & ifʕa
nlā. Granted, those are pretty rare. Still, I thought it worth pointing out.
Mecislau wrote:My apologies for any inaccuracies or mistakes I've made in the descriptions of languages I'm less familiar with. My familiarity with Semitic and Afro-Asiatic outside of Hebrew is relatively limited, and I hope I was clear in saying so.
Hey, no worries! I'm sorry I'm a pedantic douche! I really think you've been most gracious in tolerating my penchant for nit-picking. And hey, it's not like I don't make mistakes or misremember things.
Mecislau wrote:That said, it is good to see you again, Wiseblood! You're certainly a lot more knowledgeable about Semitic and Afro-Asiatic languages than I am, and I always enjoy reading whatever information you provide.
You flatter me sir. I've just been very fortunate to have had the opportunity to do a fair bit of academic study on comparative AA. However, it *is* quite nice to be back.
TomHChappell wrote:Quite interesting. Are C1C2C1C2 the most frequent kind of tetraconsonantal roots in Semitic languages in general? Or at least the most-common other than "modern" (post-1492) loanwods?
Hmmm. I'm not exactly sure, but I'd say roots of the type ABAB are probably more common both in the classical as well as the modern period. I have to ask why you chose 1492 as the beginning of "modern" loanwords, though. I'm well aware of both Columbus and the Spanish expulsion, but I fail to see what these would have to do with loanwords.
TomHChappell wrote:Useful to know, and not what I expected.
I forgot to mention that based on spellings like 'm
uātu' "to die", some people believe that Akkadian, and thus PS, retained some labialized consonants from PAA; that is the word was pronounced something like /mʷa:tu/. Many Ethiopean Semitic languages have labialised consonants as well, but I'm pretty sure that those are of secondary origin. Just some food for thought.
TomHChappell wrote:I thought it was "house of bread" rather than "house of meat".
BTW what are "bakery", "butchershop", "sandwich-shop or delicatessen" in Semitic languages (especially Hebrew and Arabic)?
As I recall, the equivalent stem (L-Ḥ-M) often means "cattle" in South Semitic, so we might assume the original meaning was simply something like "food". Now that I think about it, the verb لحم /laḥama/ means "he fused" (it's actually used to describe soldering - لحام /liḥāmun/ - in modern contexts), so the Hebrew meaning might be the original one. The name itself is probably Aramaic in origin, in which case it would definitely have meant "house of bread".
In Arabic, a butchershop is called a ملحم /malḥamun/ "meat-place". The word مخبز /maxbazun/ from خبز /xabaza/ "bake" is used to refer to a bakery. A deli would either be a مطعم /maṭʕam/ "feeding-place" from طعِم /ṭaʕima/ "feed" or لحوميّة /luḥūmiyyatun/ "pertaining to meats", depending on the context. If you need to specify, you can refer to a sandwich shop as, wait for it - دكّان الشطائر /dukkānu-š-šaṭāʔiri/ - "shop of sandwiches". Shocking, I know.
Okay, I think I'll have to cut it short there. I'll try to get back to this soon.