Friday, February 26, 2010

Medial -ur-

On pg. 408 in discussing the desideratives in -uriō, I mention that the short u of this suffix violates Latin phonotactic constraints  since the only short vowel normally permitted before an r in a medial syllable is e.   As I mention in the footnote there are of course obvious analogical cases like augur, auguris, but there is perhaps another source for this sequence.  Original u like all short vowels became e (*swekurī 'father-in-law' gen. sg. > socerī), presumably by general weakening to i and lowering before, but if a labiovelar, and presumably also the  labiovelar glide, precedes the medial vowel the outcome is u, e.g. *per-kwatiō  > percutiō, iekwVna:num > iecunanum (glossed victimarium (Fest. p. 114 M) and derived from the old oblique stem of iecur according to Alan Nussbaum p.c.). What would happen if a u of this sort were followed by an r?  There are no clear cases but it is possible that if the u arose after or as the result of weakening, as is likely, it would be retained. A potential case might be decuria if from *dekwiria (more about that elsewhere).  This idea doesn't lead immediately to an immediate solution to the -uriō question, but it does open some interesting vistas.

Tuesday, February 23, 2010

The stem-vowel of the gerund and gerundive

On pg. 444 I address the variation in the vowel before -nd- in gerunds and gerundives formed to 3rd and 4th conjugation verbs.  In addition to the observations recorded there I should also have mentioned the discussion of Alfonso Traina in the 2nd edition of his book, Forma e suono. Da Plauto a Pascoli. Bologna: Patron, 129–143. Based chiefly on an examination of the invariant cases of transmission in the manuscripts of Terence, Traina establishes the following tendencies:

After -u, -ṷ-, -qu- the form is always -end- (vivendus).

After a syllable containing o or u the form is predominantly -end-

After i the form is predominantly -und- (faciundus).

Thus it seems likely that the choice of vowel was governed largely by avoidance of two segments with identical values for roundness.

Sunday, February 21, 2010

Lesser-known Historical Grammars of Latin XIII

The book pictured at left was published in 1962 by the  Italian linguist Carlo Tagliavini.  It is the 3rd edition of a work first published in 1938 and 1939. Tagliavini was a master expositor and like his outstanding Le origine delle lingue neolatine this book is a very readable and well organized survey of the material. The most interesting fact I picked up from this book was Tagliavini's statement in the preface that he was not able to complete the section on verbal morphology in 1939 (later added in the second edition of 1949) because he was called away in March of that year on a mission for the Ministero degli Esteri in Poland.  I haven't been able to find anything about this mission in other sources although I haven't examined Jerzy Borejsza's book about Poland, Germany and Italy on the eve of  WWII.  Interestingly, Tagliavini was a noted Albanologist and Italy's invasion of Albania took place in April of 1939.  Could this have been the real reason for his sudden departure? Pure speculation.

Friday, February 19, 2010


On pg. 357 I give the proto-form of iuxtā as *iugistā. The zero-grade u with a short vowel goes 
against de Vaan 2008:318 and Watkins 1975:530 who specifically notes "a ma connaissance aucune évidence ne nous empêche de postuler un ū longe." But the evidence of Old French, which has joste 'next to' and the derived verb joster 'put side by side, joust' points clearly to a short vowel. See the Dictionnaire étymologique de l'ancien français s.v. Forms with u in the various Romance languages are secondary. This fact has interesting consequences for the morphological analysis of this form. It makes connection with the neuter s-stem iūgera even more difficult, but it also doesn't exactly favor Watkins's deverbal superlative analysis since those forms too tend to have full-grades. 

In any case connection with the root of iungō is unavoidable and this means that the g of iug- has been devoiced before a voiceless consonant without the operation of Lachmann's Law.  The absence of lengthening—cf. fūsus from *fud-to- with LL—might be explained in a number of different ways.  It might be the case that LL didn't apply to secondary sequences arising by syncope (hence maximus, despite its lone apex, really doesn't have a long vowel). Or it might be the case that it did apply to such sequences but only if the voiced stop was temporarily reintroduced on the basis of related forms.  In the case of iuxtā the derivational connection with iungō was obscure enough that this didn't happen.

Sunday, February 14, 2010

Lesser-known Historical Grammars of Latin XII

The work pictured at left is the first of a two volume grammar of Latin combining general didactic info with historical comparative detail in smaller print.  Volume 2 (1918) covers syntax. A third volume to cover prosody and stylistics was promised but apparently never produced. The author Eduardo García de Diego was a student of the great Hispanist Ramón Menendez Pidal (1869-1968!) to whom this volume is dedicated. He was also an editor of the Latin glossaries from the monastery of Silos (1934). The book is only found in this hemisphere at Stanford and UC Berkeley.  The degree of detail is average; the layout attractive.

Saturday, February 13, 2010

Old Italian fie, fieno

On pg. 525 I say that erō is the only Latin future to survive into Romance in OFr. ier, OOc. er, and Sp. 2nd sg. pres. eres. But this is not correct.  The Latin future of fiō, fiam, fiēs etc. did survive into Old Italian, and beyond as an archaism as an alternative future for essere 'to be'. The expected forms should have had an alternation between an a proper to the first singular and an e proper to all the other persons. However, the a allomorph was generalized to the other persons pretty early.  Here are some examples:

3rd sg. fie: (Albertan. cap. 22): dopo me avrò erede lo quale io non so s'egli fie savio ovvero stolto, e fie signore del mio acquisto, e delli miei beni.

3rd sg. fia (Vita dei santi. padri. (S. Antonio, Abate, pag. 25): allora certo fia diritta l'anima nostra, quando la naturale sua integritade non sia maculata da peccato.

3rd pl. fien(o) (Petr. Canz. 127.4): Quai  fien ultime, lasso, e qua' fien prime?

3rd pl. fian(o) (Tass. Ger. 20.16): Fian per lo piu senza vigor, senz'arte.

Friday, February 12, 2010

Numasioi Vindicated

On pg. 23 n. 2 I mention Hartmann's vindication of the genuineness of the Praenestine fibula and its inscription. This famous inscription (MANIOS MED FHE:FHAKED NVMASIOI), if genuine, would be our single oldest piece of Latin.  Hartmann—to the extent that I am qualified to judge—seems to make a good case that the scientific testing supposedly proving the falsity of the document was flawed or inconclusive, but the form NVMASIOI still seemed a bit of a problem.  Numasios would appear to be the ancestor of praenomen Numerius, but Numerius if related to numerus 'number' should go back to *numesios vel. sim.  So either  (1) Numasios is not the ancestor of Numerius, (2) Numerius has nothing to do with numerus, or (3) the fibula is a fake.  Possibilities (1) and (2) are perfectly plausible, so no strong argument against the genuineness of the inscription can be drawn from the form NVMASIOI, but recent developments have put to rest any doubts about this name.
Through the kindness of Rex Wallace I've been informed about a recently published archaic Etruscan inscription on an aryballos from the "area ceretana" bearing the inscription (Poetto and Facchetti 2009:369): 

mi mlac mlakas larθus elaivana araθia numasianas
‘io (sono) il buon/bel (vaso) oleario di Araθ Numasiana per il buon Larθu’.

The form numasianas supports Numasios and makes it certain  that Numasios is unconnected with numerus

Poetto, Massimo and Giulio Facchetti, 2009. L’aryballos di Araθ Numasiana. Oebalus 4:365-384.

Thursday, February 11, 2010

Quantity of Vowels Before Final -M

On pg. 125, fn. 2 and fn. 50 I discuss the much-debated question of the length of vowels before final -m in Classical Latin.  The conclusion I reach there is that it is very difficult to determine anything about the quantity of these vowels for the Classical period.  However,  there are two additional arguments bearing on this question that I would like to mention. First, in favor of lenghthening of vowels before final -m Safarewicz 1935 notes that metrical inscriptions sometimes require a long scansion for a vowel before underlying final -m even when that m is omitted in the orthography, e.g. CIL 4.6892:

Quisquis amat nigra nigris carbonibus ardet;
 nigra cum video mora libenter aedo.
Whoever loves a dark-skinned girl burns on dark coals.
  When I see a dark-skinned girl, I gladly eat blackberries. 
(at least that is one possible interpretation; see Varone 2002:57 for other suggestions)

In the pentameter nigra(m) must scan as a spondee and Safarewicz argues that this shows that it was the vowel itself that was bimoraic.  But I'm not sure this is conclusive especially since there are also metrical inscriptions which treat the vowel with unindicated -m as short, e.g. CIL 6.1975:

quae tibi crescenti rapuit iuvenile figuram

where the e of iuvenile(m) is the first short of the 5th dactyl. See Fink 1969:448 for more examples.

In favor of a short scansion the always incisive Enrico Campanile 1973 made two observations. First  in Latin loanwords into Brittonic sequences of final -um do not behave like -ō or -ū in that they do not trigger vowel affection of the preceding vowel.  Second, it is a well established fact that Latin hexametric poets of the Classical period avoided elision of cretic sequences. The exceptions in Vergil involve final -ō in 3rd declension n-stem nominative singulars (Ecl. 3.84 Pollio amat nostrum quamvis est rustica Musam) and the 1st sg. ending -ō (Aen. 11.503 Audeo et Aeneadum promitto occurrere turmae).  But these exceptions are only apparent since it is very probable that these vowels were for Vergil synchroncically short as a result of so-called cretic shortening.  When it comes to vowels before final -m in words whose preceding two syllables form a trochee, Vergil has no compunction about eliding the final syllable, e.g. Aen. 2.667 Alterum in alterius mactatos sanguine cernam; Aen. 4.387 Audiam et haec manis veniet mihi fama sub imos. This suggests that these were not cretic sequences and therefore that the vowel was short.  I am inclined to agree with Campanile especially since this evidence agrees with Priscian's statement, and the fact that apices are never used on a vowel before final -m in inscriptions which use apices correctly.

Campanile, Enrico. 1973. Sulla quantità della vocale che precede -m in latino. Italia dialettale 36:1–6.
Fink, R. O. 1969. A long vowel befor final -m in Latin? American Journal of Philology 90:444–52.
Safarewicz, Jan. 1935. Les voyelles nasales en latin. In Atti del III congresso internazionale dei linguisti (Roma, 1933). Florence 176–9. 
Varone, Antonio, 2002. Erotica pompeiana. Love inscriptions on the walls of Pompeii. Rome: Bretschneider. 

Wednesday, February 10, 2010

Lesser-known Historical Grammars of Latin X and XI: Two from Napoli

The two grammars pictured on the left and the right were both produced in Naples by the same publishing outfit in consecutive years. The Sbordone volume is the 3rd edition published in 1964.  The Cupaiuolo tome is from 1963 and specifically notes on the title page that the lectures contained therein were delivered to "gli studenti dell'Inst. Univ.  "Orientale" di Napoli."  F. Sbordone is known mainly as a papyrologist and editor of Greek texts.  Fabio Cupaiuolo published a number of very useful bibliographical surveys on  Latin grammar and metrics, which I give references for in my book, plus a book on Latin adverbs (Cupaiuolo 1967), which I didn't cite.   The Sbordone book is really very barebones and sketchy. The one noteworthy feature is the the built-in selection of archaic inscriptions illustrating pre-classical norms.  The Cupaiuolo book is a lot more detailed and devotes about 60 pages (about 20 percent of the book) to a discussion of the formation of adverbs, clearly a subject of special interest to Professore Cupaiuolo. How Libreria scientifica editrice was talked into publishing both these works virtually at the same time is a bit of a mystery!

Cupaiuolo, F. 1963. Lezioni di grammatica storica latina. Naples: Libreria scientifica editrice.
———. 1967. La formazione degli avverbi in latino. Naples: Libreria scientifica editrice.
Sbordone, F. 1964. Grammatica storica della lingua latina. Con appendice di testi epigrafici, 3rd edition. Naples: Libreria scientifica editrice.

Monday, February 8, 2010

New Tocharian A Resource

Add Carling et al 2009 to the basic bibliography for Tocharian given on pg. 21 n. 79.

Carling, Gerd, in collaboration with Georges-Jean Pinault, and Werner Winter, 2009. Dictionary and thesaurus of Tocharian A. Volume 1: A-J. Wiesbaden: Harrassowitz.

More on More on Proclitics

Another bit of evidence for the proclitic nature of prepositions is that a "word end" falling between a preposition and its object is not treated as a genuine word end by Bentley-Luchs's Law or Meyers Law. For example, Bentley-Luch's Law states that if in iambo-trochaic verse there is a word end after the final B the preceding A must be bimoraic, but Curc. 477 (tr7) shows that a sequence of a preposition plus object was permitted in this position:

confidentes, garrulique et malevoli supra lacum.
B    c  D  A     B    c D A          b b  c D a B c D

Sunday, February 7, 2010

More on Proclitics

On pg. 111 in discussing clitic elements I simply define proclitics as elements that do not receive their own stress, but form a prosodic unit with the following tonic word, but I didn't give any Latin examples or evidence. 

It is general thought that a preposition was proclitic on the noun it governed.  There are a number of pieces of evidence for this.  

First, Quintilian reports (1.5.27) cum dico circum litora, tamquam unum enuntio dissimulata distinctione, itaque tamquam in una voce una est acuta. "For when I say circum  litora I pronounce the phrase as one word, concealing the fact that it is composed of two, consequently it contains but one acute accent, as though it were a single word.  [translation H. E. Barber, Loeb edition].  

Second, we often find prepositions written together with the noun they govern, or, in text which regularly use interpuncts to separate words, with omitted interpuncts. For example, APVRFINEM (ILLRP 7, from Marsian territory), INFRONTE (CIL 12.1319) and with missing interpuncts from the letters of Rustius Barbarus from the Ostraka of Wadi Fawâkhir, 1st or 2nd cent CE: 1.4f. ·per Popilium·, cf. 1.6 ·per Draconem·.

Incidentally the term proclitic, unlike enclitic, does not have ancient roots and was coined by Gottfried Hermann (pictured above) in De emendanda ratione Graecae Grammaticae.

See Adams, J. N. 1996. Interpuncts as evidence for the enclitic character of personal pronouns in Latin. ZPE 111:208–10.

Friday, February 5, 2010

Lesser-known Historical Grammars of Latin IX

A Görög és Latin Nyelv Hang- és Alaktana (I think that translates as Phonology and Morphology of the Greek and Latin Languages) was published in Budapest in 1932 by János Szidarovszky.  It is owned by exactly one library in the States, Columbia.  I haven't been able to discover anything about the author yet except that János Harmatta published an obituary of him in Antik Tanulmányok, 1954. I also can't say anything about the quality of this work except that the author certainly seems to have been au courant with the latest in IE scholarship to judge from the works cited. But I can say I've held it in my hands and leafed though it. Is this the sole historical grammar of Greek and/or Latin written in Hungarian?

Wednesday, February 3, 2010

Update on Casus Interrogandi

In a previous post I mentioned Nigidius Figulus' discussion of the intonation and/or stress of the vocative and genitive of Valerius.  I didn't say anything about Nigidius' odd term casus interrogandi, in this context, apparently meaning the genitive.  This term and the interpretation of this passage and others are the subject of a book-length treatment by Walter Belardi and P. Cipriano, Casus interrogandi. Nigidius Figulus e la teoria stoica della lingua. They argue that casus (plural)  interrogandi refer to the oblique cases in general, a term originating in Stoic grammatical theory.

The authors specifically lament the fact that English-speaking scholars, especially on the subject of ancient Grammar, don't read works written in Italian (p. 159).  I try very hard to read what Italian scholars produce, but I missed this one.  

Belardi W. and P. Cipriano, 1990. Casus interrogandi. Nigidius Figulus e la teoria stoica della lingua. Rome: Libreria Herder.

Worst Mistake Discovered Yet!!

On pg. 60 fn. 52 at the end of the note I wrote:

Quintilian (1.7.23) mentions that Servius Sulpicius, presumably the consul of 53 BCE, was criticized for removing final s whenever the next word began with a vowel.

This sentence is embarrassingly full of errors. First, the passage where Quintilian discusses Servius Sulpicius' treatment of final -s is not 1.7.23 (that's the famous one about dice and facie), it's 9.4.38. Second, what Quintilian reports there is as follows:

Quae fuit causa et Servio Sulpicio, ut dixit, subtrahendae s litterae quotiens ultima esset aliaque consonante susciperetur, quod reprehendit Luranius, Messala defendit. Nam neque Lucilium putat uti eadem ultima, cum dicit "Aeserninus fuit" et "dignus locoque", et Cicero in Oratore plures antiquorum tradit sic locutos.
This was the reason (i.e. avoidance of stridor) why Servius, as he himself has observed, (or "as I said" if ut dixi is read) dropped the final s, whenever the next word began with a consonant, a practice for which Luranius takes him to task, while Messala defends him. For he thinks that Lucilius did not pronounce the final s in phrases such as, Aeserninus fuit and dignus locoque, while Cicero in his Orator records that this was the practice with many of the ancients. [Translation H. E. Butler, Loeb edition]

So Sulpicius is said to have eliminated s when the next word began with a consonant, presumably an affectation modeled on old writers of the Republic, not before a vowel as I erroneously reported.

Third, Servius Sulpicius was consul in 51 BCE, not 53 BCE.  Finally—this last one is not really an error on my part—it is not certain that the Servius Suplicius referred to here is the consul of 51 BCE or his son.  Syme 1981 argued for the later.

So the sentence should be emended to:

Quintilian (9.4.38) mentions that Servius Sulpicius, either the consul of 51 BCE or more probably his son (see Syme 1981), was criticized for removing final s whenever the next word began with consonant to avoid a cacophonous effect. He was probably modeling this archaizing affectation on early Republican authors.

Syme, Ronald. 1981. A great orator mislaid. The classical quarterly. 31.2:421–7.

More on Final -s in Old Latin

On pg. 60 I touch upon the fact that final -s was sometimes omitted in Old Latin inscriptions and that it did not make position in the thesis in non-dramatic verse.  I didn't say anything explicit about dramatic verse except to make reference to Rex Wallace's article that argues that non-observance of final -s is more common in spoken verses than in the cantica.

The standard view, as canonized for example in Questa 2007:32–3, is that in dramatic verse s was "reduced", i.e. potentially did not make position,  in polysyllables, following a  short vowel before a consonant initial word.

Here are some examples where s fails to make position:

Plaut. Bacch. 313:  ibidem publicitus servant:: occidisti(s) me (ia6)
                                a  a  B     C    d d A  B     C       D  a   B  c    D

But note that the final s of publicitus does make position.

Ter. Ad. 839:  exporge frontem:: scilicet ita tempu(s) fert (ia6)
                        A    B  c    D   A        B c  D a a  B       c      D

To which we can contrast:

Ter. Eun. 1048: an mei patris festivitatem et facilitatem?  O Iuppiter!
                           A    b b   c   D   A  B c D      A   b b c D         A  B     c  d+

where s does make position in patris.

This account raises at least three questions.  First, how do we know that s was retained after a long vowel since V:s and V:(s) are metrically equivalent? Second is there any pattern to the observance and non-observance of s?  Third what happened to final s prevocalically?

(1)  I don't think we can be sure that final -s had a different treatment after long vowels than after short vowel.  There can be no metrical evidence as far as I can see.  Further non-notation of final -s after a long vowel (see Vine 1993:22) is found in inscriptions at Rome, and not just dialectically—an incidentally Latin inscriptions from the 3rd century are more common found outside of Rome.

(2) I don't know if there is a pattern or not.  It doesn't seem that anyone has pointed to an obvious one in the literature.  Since there is a difference in non-dramatic verse between the strong and the weak positions within the foot, it would be interesting to explore whether anything could be made of that for iambo-trochaic verse.  It will be important in this investigation to omit all pyrrhic/iambic sequences from this investigation since Iambic Shortening is an alternative explanation in those cases.

(3) In Plautus and Terence, there do not seem to be any certain cases of loss of s in prevocalic position with subsequent elision—mage rarely occurs prevocalically and elides, but we are not certain that mage is phonologically derived from magis.  This suggests that in the relevant idiolects final s resyllabified into the onset of the following word  But there does appear to be evidence for  an alternative pronunciation where s was lost before a vowel.  In Cicero's Orator 153 we find the following sentence in the OCT edition of Wilkins:

Sine vocalibus saepe brevitatis causa contrahebant, ut ita dicerent: multi' modis, in vas' argenteis, palmi' crinibus, tecti' fractis.

palmi’ Ribbeck; palmet A (Abrincensis = from Avranches pictured in the postcard above) : palma et L (= consensus FOPM)

Here Cicero seems to be referring to reduced pronunciation and/or spellings in some way analogous to elision. In the case of multi'modis Cicero may be referring to multimodīs, the adverb which is a hypostasis of multīs modīs. Here multi- stands form multīs but is a combining form, not a phonological reduction.  The second example, however,  can only be explained through the omission of final s after a long vowel followed by elision: vasīs argenteīs > vasī' argenteīs > vas'argenteīs.  The third example, where the manuscript tradition clearly points to palm'et crinibus for palmis et crinibus, requires the same explanation.  The final form with no elision simply seems to refer to non-notation of final s.

The loss of s in vidēn et sim. seems to be a different phenomenon.

Misleading Typo

On pg. 60 I quote the form STATIA for gen. sg. stātiās from a 3rd–2nd century BCE inscription from Rome as an example of the omission of final -s in Old Latin.  This form is said to come from  CIL 12.480, but that's wrong it comes from CIL 12.480, i.e. the second edition of vol. 1 of CIL not from CIL 12 which covers Gallia Narbonensis.     There may be other places where the 2 failed to get superscripted. 

Update: Yes, there is another place: pg. 60 n. 51 last line: For "CIL 12.592.2" read "CIL 12.592.2"

I will have more to say about the treatment of final -s in a future post.  

Monday, February 1, 2010

Quintilian on C

On pg. 58 in discussing the pronunciation of as a voiceless velar stop, I might have mention this passage from Quintilian (Inst. 1.7.10) where Quintilian mentions that he thinks the letter k should only be used in those words which can be abbreviated by k alone (e.g. K for Kalendae, Kaeso, etc.):

Hoc eo non omisi quod quidam eam quotiens a sequatur necessariam credunt, cum sit c littera, quae ad omnis vocalis vim suam perferat.

I mention the fact because some hold that k should be used whenever the next letter is an a, despite the existence of the letter c which maintains its force in conjunction with all the vowels. [Translation H. E. Butler, Loeb edition]

This clause [quae ad omnis vocalis vim suam perferat] might be taken to mean that c had no significant allophones and therefore that c was still a voiceless velar stop before a front vowel.  Fantasy image of Quintilian from the 15th century Nuremberg Chronicle.