More about michi
Harrington’s Medieval Latin suggests that “the letter H was already weak in CL, and the Latin-speaker was likley to be uncertain about where to use this sound (cf. Catullus 84). Internally, the H was used to mark a syllable but was not pronounced (mihi, nihil, prehendo). In spite of the pretensions of African orators who affected to pronounce an initial H at the time of St. Augustine (Conf. 1.18.19), there is no trace of Roman H in Romance (the Fr. aspirate H is Germanic in origin, the H of Spanish haber a learned spelling.)
[with regard to michi = mihi, etc.]: after H lost all sound, scribes often wrote -CH- to indicate a disyllable. nichi (in Spain nicil) = nihil. In Spain, also mici, arcivum, macina.” (Harrington, Medieval Latin, p. 5)
A thread on ChoralNet raises another good consideration regarding the role of medieval orthography in the whole mess:
The rogue “c”s in “michi” and “nihil” may also be there for the same reason that medieval scribes put an “h” in Ihesus (I can remember some bs and ps appearing as well, but can’t think of an example off the top of my head – oh yes, the transmutation of the digraph “nt” to “mpt” or “mbt” – cf. redentor -> redemptor.). The medieval pen stroke called a “minim” was just, effectively, a vertical line, so three vertical lines could have stood for ni, ui, in, iu, or m… all of which could have led to substantial confusion, even disregarding the difference between vocalic and consonantal u – what we now call u and v. And consider also that a slightly longer vertical line would have signified an l…
Hence the appearance of extra letters – to tell the reader exactly what word they were looking at. Otherwise ||||||| (that’s an exaggeration) for MIHI could have been any grammatically correct collection of Latin letters. A strategically placed C (||||C|||) gave the reader at least some clue.
Renaissance reform of Latin by such as Poliziano and Ronsard, ironing out some of these rogue letters, went hand in hand with clearer writing and font sizes. It is perhaps interesting to note that I remember no such rogue Cs from the Latin of Charlemagne’s time, which was written in uncial – a script much more legible than the usual medieval Black Letter.
None of this is to deny the points about pronunciation made by other contributors to this thread; however, accounts also need to remember the existence of Latin as a _written_ language, indeed the language written most in the Medieval period.
An illustration of the above is the notorious mimi numinum nivium minimi munium nimium vini muniminum imminui vivi minimum volunt, a scribal exercise which, in addition to being a terrific tongue twister and using a only a few letters, is virtually unreadable in Black Letter script–a good argument for dotting i’s, adding extra letters not made up solely of vertical strokes, etc. (It translates to something like “The very short mimes of the snow gods do not wish at all that the very great burden of distributing the wine of the walls will be lightened in their lifetime.”) Beautiful as Black Letter is, we are all grateful for the Renaissance italic.
I remember that tongue twister! I mean, I don’t really remember it, but I remember remembering it. Wow, those last few words might be hard to read without the Renaissance, too.
Emily
16 December 2007 at 5:39 pm
1)I disagree somewhat with Harrington on his interpretation of Catullus 84 as evidence that “the Latin-speaker was likley to be uncertain about where to use this sound”. For the poem’s joke to work, there could be no reasonably accepted ambiguity among its intended audience (i.e., upper class Romans) regarding correct Latin pronunciation.
84 is more likely poking fun at lower-class/uneducated pronunciation (cf. commentary by E.T.Merrill: [5-6] “The point of these two parenthetical verses (cf. the verse introduced by credo in Catul. 2.8) seems to be that this super-aspiration was considered to be a characteristic of low-born and uneducated people (Gell. 13.6.3); and as the relations cited are all on the mother’s side, it looks as though the ancestry of Arrius in the female line had already been the subject of jest among his acquaintances (cf. Cicero’s remark concerning him in Cic. Brut. 243 infimo loco natus). The point of liber as an adjective and not a proper name is then clear, if infimo loco be understood of the condition of slavery: his maternal uncle (perhaps only one of his uncles on that side) was a libertus, and the social standing of the entire family is thus indicated.”).
The existence of this linguistic quirk among freedmen seems plausible given the large numbers of Greek slaves in Rome, i.e., there must have been a critical mass among the families of freedmen for the endurance of a Greek accent in Latin speech. Moreover, the standard experimentally-based model for linguistic changes predicts that one’s native phonology is the characteristic most readily transferred on to a learned language. Apropos this, graffiti in Pompeii demonstrates transferred characteristics that are far more advanced (e.g., Greek inflectional morphology on Latin stems), so Arrius’ superfluous aspirations seem rather tame compared to what they could have been.
In any event, we really shouldn’t equate having a foreign accent with being uncertain about the appropriate use of a particular phoneme. If it’s true that Arrius’ accent was based in his freedman (read: “Greek”) origins, then that somewhat undercuts Harrington’s use of him as an exemplary “Latin-speaker”. The ‘h’-sound may have been weak (it probably was), and many speakers may have been uncertain about its use (they probably were), but that all depends on which class of speakers were referring to. And given the Late Republican Roman elite’s demonstrable potential for linguistic anal-retentiveness, I find it hard to believe that educated Latin-speakers would have had any doubt about what was correct until much later.
Moving on to a slightly more tenuous argument, even if educated Latin-speakers knew what was correct, that doesn’t necessarily mean that their scribes did. In other words, a foreign scribe taking dictation might easily slip over an unpronounced or lightly pronounced intervocalic ‘h’. So if our concern is with orthography, then this throws a wrench into the works. It seems to be a small wrench, though. We do have some evidence, however meager, that intervocalic ‘h’ was retained in writing at least into the 2nd century AD, since the word ‘mihi’ can be found among the papyrus letters of Claudius Terentianus. As always, more study is necessary before definitive conclusions can be reached
2) If the French aspirate is Germanic in origin, then that suggests a basis for the view that medieval michi/nichil reflects a regional difference in pronunciation. A friend of mine has observed that initial ‘h’ is often absent from Italian music MSS he’s looked at, but present in French and English MSS. Alright, who’s going to do the grunt work of tabulating the data?
3) I’m intrigued by those Spanish spellings, esp. since for ‘mici’, ‘arcivum’, and ‘macina’ we get in modern Spanish ‘mi’, ‘archivo’ (as in “cheese”), and ‘maquina’ (as in “key”). But then you also get ‘ariquitectura’, not ‘architectura’….and “archipiélago”, not “arquipiélago”. I’m inclined to think that the modern -ch- in these words was a later introduction…but why?
evovae
18 December 2007 at 9:25 pm
[...] away under the stacks at Tübingen or Heidelberg. But that won’t stop us!) and the most recent exchange brought up Catullus 84 as an example of how you could be hammered in Ancient Rome for incorrect [...]
It’s not what you say… « Tota Nostra Est…
18 December 2007 at 9:49 pm