User:RandomCritic/Sandbox3

Conversion of Greek to Latin

A-L includes a large amount of Greek vocabulary; in principle, any Greek noun or adjective can be converted into an A-L word. There is a conventional set of equivalents between the letters of the Greek and Roman alphabets, which differs in some respects from the current mode of Romanizing Greek. This is laid out in the tables below:

Rh is used for Greek ρ at the beginnings of words, e.g. ρομβος (rombos) > rhombus. Rarely (and mostly in words relatively recently adapted from Greek), k is used to represent Greek κ. In such cases it is always pronounced [ k ] and never [ s ] (as it might be if spelled c) : e.g. σκελετον skeleton, not "sceleton".

Greek accent marks and breath marks, other than the "rough breathing" (first in the list of consonants above), are entirely disregarded; the Greek pitch accent is superseded by a Latin stress accent, which will be described below.

Frequently, but not universally, certain Greek nominative endings are changed to Latin ones which cannot be predicted from the tables above. Occasionally forms with both endings are found in A-L, for instance Latinized hyperbola next to Greek hyperbole. The most usual equations are found below:

Examples:

• Greek αγγελος (aggelos) > Latin angelus (γγ > ng, -ος > us)
• Greek ελλειψις (elleipsis) > Latin ellipsis (ει > i, ψ > ps)
• Greek μουσαιον (mousaion) > Latin musæum (ου > u, αι > æ, -ον > um)
• Greek μαιανδρος (maiandros) > Latin mæander (αι > æ, -ρος > er)
• Greek χρυσανθεμον (chrusanthemon) > Latin chrysanthemum (χ > ch, υ > y, θ > th, -ον > um)
• Greek διαρροια (diarroia) > Latin diarrhœa (ρρ > rrh, οι > œ)

Letters and sounds

• The letters b, f, k, l, m, p, v and z have each only one sound, which corresponds to the equivalent IPA symbols [ b f k l m p v z ].
• The letter j has the single sound [ dʒ ].
• The letter r has a single sound, [ ɹ ] in rhotic dialects of English. In non-rhotic dialects, it varies according to placement in a syllable. At the beginning of a syllable, it is pronounced [ ɹ ]. At the end of a syllable, i.e. between a vowel and a consonant, or after a vowel at the end of a word, it is dropped -- though not without, frequently, affecting the pronunciation of the previous vowel sound. If r occurs at the end of a word after a vowel, and the next word begins with a vowel, it is usually pronounced as the beginning of the first syllable of the next word. Rh and rrh are pronounced exactly like r and rr.
• When followed by a vowel, the combinations qu (always) and gu and su (usually) stand for [ kw ], [ gw ], and [ sw ] respectively.
• The combination ph is pronounced [ f ].
• The combination th is pronounced [ θ ].
• The combination ch is pronounced [ k ] in all environments.
• The letters c, d, g, h, n, s, t, x have different values depending upon surrounding sounds and syllable structure.

Phonemes

The underlying consonantal phonemes of A-L are close in most respects to those of Latin, the primary difference being that / w / and / j / are replaced in A-L by / v / v and / dʒ / j. The sounds / θ / th and / x / ch were borrowed from Greek; the latter became an invariable [ k ] subsequent to the split of original / k / c into [ k ] and [ s ].

Consonantal allophones

Openness

Openness is a quality of syllables, which may be either open, semi-open, semi-closed, or fully closed.

Stress

Primary stress

Stress is another characteristic of syllables. In A-L, it is marked by greater tension, higher pitch, lengthening of vowel, and (in certain cases) changes in vowel quality. Its exact concomitants in Classical Latin are uncertain. In Classical Latin the main, or primary stress is predictable, with a few exceptions, based on the following criteria:

• In words of one syllable, stress falls on that syllable, as marked in the following syllables with an acute accent: quá, nón, pár.
• In words of two syllables, stress falls on the first syllable of the word (the penult, or second from the end): e.g. bó.nus, cír.cus.
• In words of three or more syllables, stress falls either on the penult or the antepenult (third from the end), according to these criteria:
  • If the penult contains a short vowel in an open syllable, the stress falls on the antepenult: e.g. stá.mi.na, hy.pó.the.sis.
  • If the penult contains a long vowel; a diphthong; a closed syllable (with any length of vowel); or is followed by z, the stress falls on the penult.
    • Long vowel: cicāda > cicáda, exegēsis > exegésis.
    • Diphthong: musáeum, amóeba, Acháia, paranóia, thesáurus
    • Closed syllable: aórta, interrégnum, prospéctus, rotúnda
    • z: horízon

Primary stress can therefore be determined in cases where the penult is either closed or contains a diphthong. When it contains a vowel that may have been either short or long in Classical Latin, stress is ambiguous. Since A-L does not distinguish short from long vowels, stress becomes a lexical property of certain words and affixes. The fact that decorum is stressed on the penult, and exodus on the antepenult, is a fact about each of these words that must be memorized separately (unless one is already familiar with the Classical quantities).

Secondary stress

Secondary stress is dependent upon the placement of the primary stress. It only appears in words of four or more syllables. There may be more than one secondary stress in a word; however, stressed syllables may not be adjacent to each other, so there is always at least one unstressed syllable between the secondary and primary stress. Syllables containing semivowel e or i are never stressed.

• If a 4-syllable word has primary stress on the antepenult, there is no secondary stress: pa.rá.bo.la, me.tá.the.sis.
• If a 4-syllable word has primary stress on the penult, secondary stress is on the first syllable, marked hereafter with a grave accent: à.la.bás.ter, è.pi.dér.mis, sì.mu.lá.crum, prò.pa.gán.da, ùl.ti.má.tum.
• If a 5-syllable word has primary stress on the antepenult, secondary stress is on the first syllable: hìp.po.pó.ta.mus, Sà.git.tá.ri.us, Phì.la.dél.phi.a.

Secondary stress in words with three or more syllables before the primary stress is less predictable. Such words include those of five syllables with penult primary stress, and all words of six syllables in length or longer. The following generalizations about such long words may be made:

• The syllable immediately before the primary stress is never stressed.
• Words produced by derivation from a shorter word convert the primary (and, if any, secondary) stress of the stem into a secondary stress, as long as it does not fall immediately before the new primary stress: é.le.phant- + í.a.sis becomes è.le.phan.tí.a.sis
• Compounds of which the compound element consists of more than one syllable likewise convert the primary stress of their elements into secondary stress: phár.ma.co- + póei.a becomes phàr.ma.co.póei.a.
• If a primary stress is eliminated in compounding or derivation because it would stand next to another stress, secondary stress remains unchanged: pùsillánimus + itas becomes pùsillanímitas.
• Single-syllable prefixes and single-syllable compound-elements are generally unstressed: ac.cù.mu.lá.tor, im.pè.di.mén.ta, Her.mà.phro.dí.tus
• In other cases where the composition of the word may be unclear, every other syllable before the primary stress is stressed: a.mà.nu.én.sis, ò.no.mà.to.póei.a.

Unstress

Unstressed syllables are all others. They are always adjacent to a stressed syllable; that is, there can never be more than two unstressed syllables in a row, and that only when the first one follows a stressed syllable.

Mergers

The most notable distinction between A-L and other varieties of Latin is in the treatment of the vowels. In A-L, all original distinctions between long and short vowels have been obliterated; there is no distinction between the treatment of a and ā, etc., for instance. However, the subsequent development of the vowels depended to a large degree upon Latin word stress (which was preserved nearly unchanged in the mediæval period), and as this was in part dependent upon vowel length, in certain cases Latin vowel length contrasts have been preserved as contrasts in both stress and quality. However, the immediate governing factor is not length but stress; short vowels which were stressed for various reasons are treated exactly like stressed long vowels.

In addition to the merger of long and short vowels, other vowel mergers took place:

• the diphthongs æ and œ merged with e
• the vowels i and y merged
• the diphthong ei (also æi, œi), when still written distinctively, in pronunciation was merged with i or (more frequently) e

The merger of æ and œ with e was commonly recognized in writing. Sometimes forms written with æ and œ coexist with forms with e; in other cases the form with e has superseded the diphthong in A-L. Consider the following:

• æon and eon, æther and ether, amœba and ameba, anæmia and anemia, anæsthesia and anesthesia, cæsura and cesura, chamæleon and chameleon, dæmon and demon, diæresis and dieresis, encyclopædia and encyclopedia, fæces and feces, fœtus and fetus, hyæna and hyena, prætor and pretor

The following words are usually spelled with e, though they originally had æ:

• ænigma > enigma, æquilibrium > equilibrium, æra > era, Æthiopia > Ethiopia, diarrhœa > diarrhea, mæander > meander, musæum > museum, œsophagus > esophagus, pæninsula > peninsula, præcentor > precentor, prædecessor > predecessor, præmium > premium, præsidium > presidium, tædium > tedium

In other cases, particularly names, the forms with the diphthongs are the only correct spelling, e.g. ægis, Cæsar, Crœsus, Œdipus, onomatopœia, pharmacopœia, Phœbe, phœnix, Piræus, sub pœna.

The sequences ei, æi, œi (distinguished in writing and pronunciation from ej, the vowel followed by a consonant, as in Sejanus) are sometimes retained in spelling preceding a vowel. In such cases the sequence is invariably pronounced as a simple vowel, sometimes i (as in meiosis, pronounced as if miosis) but more usually e: Cassiopeia, Deianira, Pleiades, onomatopœia, pronounced as if Cassiopea, Deanira, Pleades, onomatopea.

The result was a system of five vowels, a, e, i, o, u. These would subsequently split, according to their environment, into long, short, and (eventually) unstressed variants; and these variants would eventually also be altered in various dialects of A-L dependent upon neighboring sounds. However, in phonemic terms, A-L still has only five vowels, with multiple allophones.

In addition, there were the diphthongs, ai, oi, ui, au and eu. Of these, ai and au eventually monophthongized, eu merged with the open variant of u, and yi merged with the "long" i. Only oi and ui remained as true diphthongs, but both are extremely rare.

Allophones of a, e, i and o

The vowels a, e, i, o have three primary variations: long, short, and reduced; each of these may, in turn, have additional variations based on their phonetic environment, including whether they are stressed; in an open or closed syllable; their position in the word; and neighboring consonants. One of the most common environmental alterations of a vowel is due to the presence of a following r; such vowels will be called "r-colored".

Open and closed u

The pronunciation of the letter u does not depend upon stress, but rather upon whether the syllable in which it appears is open or closed. There are no "long" and "short" variants of either type of u, but there are reduced and r-colored variants of both types.

Diphthongs

Diphthongs in A-L are distinguished from simple vowels by having no long or short variants, regardless of position or syllable type. The only diphthongs that are at all common are au and eu. For variations in the pronunciation of the latter, see Open u. Au is, rarely, reduced in an unstressed syllable to [ ə ]: Augustus pronounced as if "Agustus". Such words may be pronounced with the full value of the diphthong, however.

Note that ui is often dissyllabic, as in fruc.tu.i, va.cu.i, tu.i. The words cui and huic were traditionally pronounced [ kaɪ ] and [ haɪk ] (Am. [ kai ], [ haik ], Au. [ kɑe ], [ hɑek ]).

Phonology

Since A-L developed for many centuries within English, yet without having a full complement of English sounds, its phonology taken in isolation is somewhat unlike that of most natural languages.

For the most part, establishing the identity of A-L phonemes presents little difficulty. The r-colored vowels [ɑɹ / ɑː/ aː] , [ ɝ / ɜ ] , [ ɔɹ / ɔː] are clearly conditioned by a following / ɹ /, and are allophones of / æ /; / ɛ /, / ɪ / or / ʌ /; and /ɑ/, respectively. The reduced vowel [ ə ] is likewise an allophone of the same five vowels in unstressed syllables, and the reduced r-colored vowel [ ɚ ] in American English is likewise an allophone occurring in unstressed syllables preceding / ɹ /.

The exact status of the "long vowels" is more difficult to ascertain. They certainly exist as conditioned alternants to the short vowels / æ /, / ɛ /, / ɪ /, /ɑ/, and / ʌ /; the limits on their distribution demonstrate that they cannot be the underlying forms. Their exact identity, given dialectal variants, is debatable, but they will here be considered to be [ ei ], [ i ], [ ai ], [ ɔu ], and [ ʲu ], the last being a form of [ u ] which palatalizes a preceding consonant.

However, many of these sounds can be demonstrated to exist as independent phonemes as well. / ʲu / is certainly a phoneme, when it appears as eu, in closed as well as open syllables. / ai / is marginally phonemic, as it exists in the closed syllable / haik / huic. / ei / is possibly phonemic when it appears as the diphthong ai, but due to the limitations of the data this cannot be demonstrated; in a maximally parsimonious system, it could be accounted for, even in these cases, merely as an allophone of / æ /. / ɔu / is at least marginally phonemic as it appears in the word post; and more certainly insofar as it provides a distinction between the endings -os [ ɑs ] and -ōs [ ɔus ] . The vowel [ i ] and its allophones might also be considered to be merely allophones of / ɛ / ; however, their sporadic (if somewhat uncertain) appearance in initial unstressed and final syllables suggests a basis for considering them phonemic. The creation of apparent minimal pairs due to degemination (e.g. penna [ pɛnə ] vs. pœna [ pinə ], collum [ kɑləm ] vs. colum [ kɔuləm ]) could also be used to make a case for the existence of / i / and / ɔu / as independent phonemes.

In addition, the vowel / ɒ / and the diphthongs / ɔi / and / ʲui / are unquestionably phonemic in A-L.

The table below reflects an analysis of A-L phonemes in which each of the "long vowels" has phonemic status.

Vowel alone

The first class consists of vowels alone, i.e. -a, -e, -æ, -i, -o, -u, -y. In this class, the vowels are generally long, but -a is always [ ə ]^[1]. In British Received Pronunciation, -e and -æ are [ ɪ ] (but in many other varieties of English [ i ]).

In the words mihi, tibi, sibi, by an old tradition, the final i was pronounced like final e above (i.e., as if spelled mihe, tibe, sibe).

A late and purely academic pronunciation distinguished final -ā from -a by pronouncing the former like "long a", [ eɪ ]: for instance, Oxford professor A. D. Godley rhymed "day" with Rusticā. That this was not the usual pronunciation can be told from such forms as circa, infra, extra, in absentia, sub pœna, all of which have an originally long final vowel: circā, sub pœnā, etc. This use is distinct from the older tradition (in use in the 17th-18th centuries) had made all final a's "long", regardless of their Latin length.

Vowel plus consonant cluster

The second class consists of vowels consonant clusters such as ns, nt, nx, ps, x. In this class, the vowels are always short. The vowel u may be reduced to [ ə ].

Vowel plus consonant

The third class consists of vowels followed by the consonants l, m, n, r, s, t. The treatment of these endings is inconsistent. Generalizations include:

1. All vowels are reduced before final r, creating American [ ɚ ], British and Australian [ ə ]: Cæsar, pauper, triumvir, Mentor, sulfur, martyr.
2. All vowels are reduced to [ ə ] before l: tribunal, Babel, pugil, consul.
3. Except sometimes before t, a is reduced to [ ə ] before any of this class of consonant: animal, memoriam, titan, atlas.
4. All instances of u are reduced to [ ə ] before any of this class of consonant: consul, dictum, locus.

The remaining endings are: -at, -em, -en, -es, -et, -im, -is, -it, -on, -os, -ot. Of these, -em, -im, -is, -it, -on, -ot have two possible pronunciations, one with a short vowel and one with [ ə ]. Final -es and -ies are alike pronounced [ iz ]^[2]. Final -eus may be pronounced containing the diphthong eu, or as if it were semivowel e followed by the ending -us. However, even when pronounced as two syllables, -eus still counts as a single syllable for the purpose of determining vowel length; the syllable preceding the ending is considered a penult.

This last pronunciation of -os is the expected one; however, in the masculine accusative plural, where the ending is historically -ōs, the academic prescription was the pronunciation Am. [ ɞus], Br. [ əʊs], Au. [ əʉs].^[3] Such an ending is not, of course, commonly met in isolated vocabulary items and proper names.

Stage I

Latin from the period when its orthography and grammar became standardized through to the pronunciation changes of Late Latin, while it was still a living language. Changes that took place in this period included:

• the merger of f and ph as [ f ]
• the change in pronunciation of v (formerly [ w ]) to [ v ] and of j (formerly [ j ]) to [ dʒ ].
• the merger of i and y as [ i ]
• the merger of e, æ and œ as [ e ]
• the change of non-initial, unstressed, prevocalic [ i ] to [ j ]
• the loss of distinctions of vowel length (merger of all long and short vowels)
• the palatalization of t to [ ts ] before [ j ]

Stage II

Latin spoken in the context of Gallo-Romance and French from approximately the 6th to the 11th-12th centuries. During this period, Latin became a primarily written language, separated from the ordinary spoken language of the people. While it escaped many of the changes of pronunciation and grammar of Gallo-Romance, it did share a few of the changes of the spoken language. This was for the most part a period of stability. Changes in this period included:

• the palatalization of c and g to [ ts ] and [ dʒ ] before front vowels
• the voicing of intervocalic s to [ z ]
• the fronting of u to [ y ]
• the restoration (based on spelling) of the vowels [ i ] and [ e ] from [ j ]

Stage III

Latin spoken in the context of English from the 11th/12th centuries to the present. This last stage provides the greatest and most complicated number of changes. It starts with the displacement of the native pronunciation of Latin under the Anglo-Saxon kings with that used in the north of France, around the time of the Norman conquest in 1066. The English and French pronunciations of Latin were probably identical down to the thirteenth century, but subsequently Latin as spoken in England began to share in specifically English sound changes. Latin, thus naturalized, acquired a distinctly English sound, increasingly different from the pronunciation of Latin in France or elsewhere on the Continent.

Some phases of development in this third stage can be reconstructed: