Old Latin

From Infogalactic: the planetary knowledge core
Jump to: navigation, search

<templatestyles src="Module:Hatnote/styles.css"></templatestyles>

Old Latin
Prisca Latinitas
The playwright Titus Maccius Plautus wrote using Old Latin.
Native to Roman Republic
Region Italy
Era Developed into Classical Latin during the 1st century BC
Latin alphabet 
Official status
Official language in
Regulated by Schools of grammar and rhetoric
Language codes
ISO 639-3 None (mis)
Linguist list
Glottolog oldl1238[1]
Expansion of Rome, 2nd century BC.gif
Expansion of the Roman Republic during the 2nd century BC. Very little Latin is likely to have been spoken beyond the area coloured green, nor was it ubiquitous within it.
This article contains IPA phonetic symbols. Without proper rendering support, you may see question marks, boxes, or other symbols instead of Unicode characters.

Old Latin, also known as Early Latin and Archaic Latin, refers to the Latin language in the period before 75 BC, i.e. before the age of Classical Latin.[2] In New and Contemporary Latin, it is called prisca Latinitas ("ancient Latin") rather than vetus Latina ("old Latin"), as vetus Latina is used to refer to a set of Biblical texts.

The use of "old", "early" and "archaic" has been standard in publications of Old Latin writings since at least the 18th century. The definition is not arbitrary but these terms refer to writings with spelling conventions and word forms not generally found in works written under the Roman Empire. This article presents some of the major differences.

Philological constructs

The old-time language

The concept of Old Latin (Prisca Latinitas) is as old as the concept of Classical Latin, both dating to at least as early as the late Roman Republic. In that time period Cicero, along with others, noted that the language he used every day, presumably the upper-class city Latin, included lexical items and phrases that were heirlooms from a previous time, which he called verborum vetustas prisca,[3] translated as "the old age/time of language."

During the classical period, Prisca Latinitas, Prisca Latina and other idioms using the adjective always meant these remnants of a previous language, which, in the Roman philology, was taken to be much older in fact than it really was. Viri prisci, "old-time men," were the population of Latium before the founding of Rome.

The four Latins of Isidore

In the Late Latin period, when Classical Latin was behind them, the Latin- and Greek-speaking grammarians were faced with multiple phases, or styles, within the language. Isidore of Seville reports a classification scheme that had come into existence in or before his time: "the four Latins" ("Latinas autem linguas quatuor esse quidam dixerunt").[4] They were Prisca, spoken before the founding of Rome, when Janus and Saturn ruled Latium, to which he dated the Carmen Saliare; Latina, dated from the time of king Latinus, in which period he placed the laws of the Twelve Tables; Romana, essentially equal to Classical Latin; and Mixta, "mixed" Classical Latin and Vulgar Latin, which is known today as Late Latin. The scheme persisted with little change for some thousand years after Isidore.

Old Latin

In 1874, John Wordsworth used this definition: "By Early Latin I understand Latin of the whole period of the Republic, which is separated very strikingly, both in tone and in outward form, from that of the Empire."[5]

Although the differences are striking and can be easily identified by Latin readers, they are not such as to cause a language barrier. Latin speakers of the empire had no reported trouble understanding Old Latin, except for the few texts that must date from the time of the kings, mainly songs. Thus, the laws of the Twelve Tables from the early Republic were comprehensible, but the Carmen Saliare, probably written under Numa Pompilius, was not entirely (and still remains unclear).

An opinion concerning Old Latin, of a Roman man of letters in the middle Republic, survives: the historian, Polybius,[6] read "the first treaty between Rome and Carthage", which he says "dates from the consulship of Lucius Junius Brutus and Marcus Horatius, the first consuls after the expulsion of the kings." Knowledge of the early consuls is somewhat obscure, but Polybius also states that the treaty was formulated 28 years after Xerxes I crossed into Greece; that is, in 452 BC, about the time of the Decemviri, when the constitution of the Roman Republic was being defined. Polybius says of the language of the treaty "the ancient Roman language differs so much from the modern that it can only be partially made out, and that after much application by the most intelligent men".

There is no sharp distinction between Old Latin, as it was spoken for most of the Republic, and Classical Latin, but the earlier grades into the later. The end of the republic was too late a termination for compilers after Wordsworth; Charles Edwin Bennett said, "'Early Latin' is necessarily a somewhat vague term ... Bell, De locativi in prisca Latinitate vi et usu, Breslau, 1889,[7] sets the later limit at 75 BC. A definite date is really impossible, since archaic Latin does not terminate abruptly, but continues even down to imperial times."[8] Bennett's own date of 100 BC did not prevail but rather Bell's 75 BC became the standard as expressed in the four-volume Loeb Library and other major compendia. Over the 377 years from 452 to 75 BC, Old Latin evolved from being partially comprehensible by classicists with study to being easily read by scholars.


The Forum inscription, one of the oldest known Latin inscriptions. It is written boustrophedon, albeit irregularly. From a rubbing by Domenico Comparetti.

Old Latin authored works began in the 3rd century BC. These are complete or nearly complete works under their own name surviving as manuscripts copied from other manuscripts in whatever script was current at the time. In addition are fragments of works quoted in other authors.

Numerous inscriptions placed by various methods (painting, engraving, embossing) on their original media survive just as they were except for the ravages of time. Some of these were copied from other inscriptions. No inscription can be earlier than the introduction of the Greek alphabet into Italy but none survive from that early date. The imprecision of archaeological dating makes it impossible to assign a year to any one inscription, but the earliest survivals are probably from the 6th century BC. Some texts, however, that survive as fragments in the works of classical authors, had to have been composed earlier than the republic, in the time of the monarchy. These are listed below.

Fragments and inscriptions

Notable Old Latin fragments with estimated dates include:

Works of literature

Cato the Elder and his wife

The authors are as follows:


<templatestyles src="Module:Hatnote/styles.css"></templatestyles>

Old Latin surviving in inscriptions is written in various forms of the Etruscan alphabet as it evolved into the Latin alphabet. The writing conventions varied by time and place until classical conventions prevailed. The works of authors in manuscript form were copied over into the scripts current in those later times. The original writing does not survive.


Some differences between old and classical Latin were of spelling only; pronunciation is thought to be essentially as in classical Latin:[9]

  • Single for double consonants: Marcelus for Marcellus
  • Double vowels for long vowels: aara for āra
  • q for c before u: pequnia for pecunia
  • gs/ks/xs for x: e.g. regs for rex, saxsum for saxum
  • c for g: Caius for Gaius

These differences did not necessarily run concurrently with each other and were not universal; that is, c was used for both c and g.


<templatestyles src="Module:Hatnote/styles.css"></templatestyles>

Diphthong changes from Old Latin (left) to Classical Latin (right)[10]


Old Latin had a strong stress on the first syllable of a word until about 250 BC. All syllables other than the first were unstressed and were subjected to greater amounts of phonological weakening. Starting around that year, the Classical Latin stress system began to develop. It passed through at least one intermediate stage, found in Plautus, in which the stress occurred on the fourth last syllable in four-syllable words with all short syllables.

Vowels and diphthongs

Most original PIE diphthongs were preserved in stressed syllables, including /ai/ (later ae); /ei/ (later ī); /oi/ (later ū, or sometimes oe); /ou/ (from PIE /eu/ and /ou/; later ū).

The Old Latin diphthong ei evolves in stages: ei > ẹ̄ > ī. The intermediate sound ẹ̄ was simply written e but must have been distinct from the normal long vowel ē because ẹ̄ subsequently merged with ī while ē did not. It is generally thought that ẹ̄ was a higher sound than e (e.g. perhaps [eː] vs. [ɛː] during the time when both sounds existed). Even after the original vowel /ei/ had merged with /ī/, the old spelling ei continued to be used for a while, with the result that ei came to stand for /ī/ and began to be used in the spelling of original occurrences of /ī/ that did not evolve from /ei/ (e.g. in the genitive singular /-ī/, which is always spelled -i in the oldest inscriptions but later on can be spelled either -i or -ei).

In unstressed syllables, *oi and *ai had already merged into ei by historic times (except for one possible occurrence of poploe for populī "people" in a late manuscript of one of the early songs). This eventually evolved to ī according to the process described above.

Old Latin often had different short vowels than Classical Latin, reflecting sound changes that had not yet taken place. For example, the very early Duenos inscription has the form duenos "good", later found as duonos and still later bonus. A countervailing change wo > we occurred around 150 BC in certain contexts, and many earlier forms are found (e.g. earlier votō, voster, vorsus vs. later vetō, vester, versus).

Old Latin frequently preserves original PIE (Proto-Indo-European) thematic case endings -os and -om (later -us and -um).


Intervocalic /s/ (pronounced [z]) was preserved up through 350 BC or so, at which point it changed into /r/ (called rhotacism). This rhotacism had implications for declension: early classical Latin, honos, honoris (from honos, honoses); later Classical (by analogy) honor, honoris ("honor"). Some Old Latin texts preserve /s/ in this position, such as the Carmen Arvale's lases for lares. Later instances of single /s/ between vowels are mostly due either to reduction of early /ss/ after long vowels or diphthongs; borrowings; or late reconstructions.

There are many unreduced clusters, e.g. iouxmentom (later iūmentum, "beast of burden"); losna (later lūna, "moon") < *lousna < */leuksnā/; cosmis (later cōmis, "courteous"); stlocum, acc. (later locum, "place").

Early du /dw/ becomes later b: duenos > duonos > bonus "good"; duis > bis "twice"; duellom > bellum "war".

Final /d/ occurred in ablatives (later lost) and in third-person secondary verbs (later t).

Grammar and morphology


Latin nouns are distinguished by grammatical case, a word with a termination, or suffix, determining its use in the sentence, such as subject, predicate, etc. A case for a given word is formed by suffixing a case ending to a part of the word common to all its cases called a stem. Stems are classified by their last letters as vowel or consonant. Vowel stems are formed by adding a suffix to a shorter and more ancient segment called a root. Consonant stems are the root (roots end in consonants). The combination of the last letter of the stem and the case ending often results in an ending also called a case ending or termination. For example, the stem puella- receives a case ending -m to form the accusative case puellam in which the termination -am is evident.[11]

In Classical Latin textbooks the declensions are named from the letter ending the stem or First, Second, etc. to Fifth. A declension may be illustrated by a paradigm, or listing of all the cases of a typical word. This method is less frequently applied to Old Latin, and with less validity. In contrast to Classical Latin, Old Latin reflects the evolution of the language from an unknown hypothetical ancestor spoken in Latium. The endings are multiple. Their use depends on time and locality. Any paradigm selected would be subject to these constraints and if applied to the language universally would result in false constructs, hypothetical words not attested in the Old Latin corpus. Nevertheless, the endings are illustrated below by quasi-classical paradigms. Alternative endings from different stages of development are given, but they may not be attested for the word of the paradigm. For example, in the Second Declension, *campoe "fields" is unattested, but poploe "peoples" is attested.

First declension (a)

The 'A-Stem' declension. The stems of nouns of this declension usually end in –ā and are typically feminine.[12]

puellā, –ās
girl, maiden f.
Singular Plural
Nominative puell-ā puell-āī
Vocative puell-a puell-ai
Accusative puell-am puell-ās
Genitive puell-ās/-āī/-ais puell-om/-āsōm
Dative puell-āi puell-eis/-abos
Ablative puell-ād puell-eis/-abos
Locative Rom-ai Syracus-eis

A nominative case ending of –s in a few masculines indicates the nominative singular case ending may have been originally –s: paricidas for later paricida, but the –s tended to get lost.[13] In the nominative plural, -ī replaced original -s as in the genitive singular.[14]

In the genitive singular, the –s was replaced with –ī from the second declension, the resulting diphthong shortening to –ai subsequently becoming –ae.[15] In a few cases the replacement did not take place: pater familiās. Explanations of the late inscriptional -aes are speculative. In the genitive plural, the regular ending is –āsōm (classical –ārum by rhotacism and shortening of final o) but some nouns borrow –om (classical –um) from the second declension.[14]

In the dative singular the final i is either long[16] or short.[17] The ending becomes –ae, –a (Feronia) or –e (Fortune).[16]

In the accusative singular, Latin regularly shortens a vowel before final m.[17]

In the ablative singular, –d was regularly lost after a long vowel.[17] In the dative and ablative plural, the –abos descending from Indo-European *–ābhos[18] is used for feminines only (deabus). *–ais > –eis > īs is adapted from –ois of the o-declension.[19]

In the vocative singular, an original short a merged with the shortened a of the nominative.[17]

The locative case would not apply to such a meaning as puella, so Roma, which is singular, and Syracusae, which is plural, have been substituted. The locative plural has already merged with the –eis form of the ablative.

Second declension (o)

campos, –ī
field, plain m.
saxom, –ī
rock, stone n.
Singular Plural Singular Plural
Nominative camp-os camp-ei < -oi sax-om sax-ā/-ă
Vocative camp-e camp-ei < -oi sax-om sax-ă
Accusative camp-om camp-ōs sax-om sax-ā/-ă
Genitive camp-ī camp-ōm sax-ī sax-ōm
Dative camp-ō camp-eis < -ois sax-ō sax-eis < -ois
Ablative camp-ōd camp-eis < -ois sax-ōd sax-eis < -ois
Locative camp-ei camp-eis < -ois sax-ei sax-eis < -ois

The stems of the nouns of the o-declension end in ŏ deriving from the o-grade of Indo-European ablaut.[20] Classical Latin evidences the development ŏ > ŭ. Nouns of this declension are either masculine or neuter.

Nominative singulars ending in -ros or -ris syncopate the ending:[21] *agros > *agrs > *agers > *agerr > ager. (The form terr "three times" for later ter < *tris appears in Plautus.)

Many alternative spellings occur:

  • As mentioned above, the sound change -ei > -ẹ̄ > -ī leads to numerous variations, including the reverse spelling ei for ī. This spelling eventually appears in the genitive singular as well, although is earliest and the true ending; cf. populi Romanei, "of the Roman people."[22], which both spellings in the same inscription.
  • Likewise, the sound change -os > -us and -ōm > -om > -um affect the nominative and accusative singular, and the genitive plural.
  • One very early text has genitive -osio (the Proto-Indo-European ending) rather than (an ending appearing only in Italo-Celtic). This form also appears in the closely related Faliscan language.
  • In the genitive plural, -um (from Indo-European *-ōm) survived in classical Latin "words for coins and measures";[23] otherwise it was eventually replaced by -ōrum by analogy with 1st declension -ārum.
  • The nominative/vocative plural masculine -ei comes from the Proto-Indo-European (PIE) pronominal ending *-oi. The original ending -oi appears in a late spelling in the word poploe (i.e. "poploi" = populī "people") in Sextus Pompeius Festus.[24]
  • The dative/ablative/locative plural -eis comes from earlier -ois, a merger of PIE instrumental plural *-ōis and locative plural *-oisu. The form -ois appears in Sextus Pompeius Festus and a few early inscriptions.
  • The Praeneste Fibula has dative singular Numasioi, representing Proto-Indo-European *-ōi.
  • A number of "provincial texts" have nominative plural -eis (later -īs from 190 BC on[25]), with an added s, by some sort of analogy with other declensions. Sihler (1995)[24] notes that this form appears in literature only in pronouns and suggests that inscriptional examples added to nouns may be artificial (i.e. not reflecting actual pronunciation).
  • In the vocative singular, some nouns lose the -e (i.e. have a zero ending) but not necessarily the same as in classical Latin.[26] The -e alternates regularly with -us.[27]
  • The locative was a separate case in Old Latin but gradually became reduced in function, and the locative singular form eventually merged with the genitive singular by regular sound change. In the plural, the locative was captured by the ablative case in all Italic languages before Old Latin.[28]

Third declension (consonant/i)

The 'Consonant-Stem' and 'I-Stem' declension. This declension contains nouns that are masculine, feminine, and neuter. The stem ends in the root consonant, except in the special case where it ends in -i (i-stem declension). The i-stem, which is a vowel-stem, partially fused with the consonant-stem in the pre-Latin period and went further in Old Latin.[29] I/y and u/w can be treated either as consonants or as vowels; hence their classification as semi-vowels. Mixed-stem declensions are partly like consonant-stem and partly like i-stem. Consonant-stem declensions vary slightly depending on which consonant is root-final: stop-, r-, n-, s-, etc.[30] The paradigms below include a stop-stem (reg-) and an i-stem (igni-).

Rēgs –es
king m.
Ignis -is
fire m.
Singular Plural Singular Plural
Nominative rēg/-s rēg-eīs/-īs/-ēs/-ĕs ign-is/-es ign-eīs/-ēs/-īs/-ĕs
Vocative rēg/-s rēg-eīs/-īs/-ēs/-ĕs ign-is/-es ign-eīs/-ēs/-īs/-ĕs
Accusative rēgem rēg-eīs/-īs/-ēs ignim ign-eīs/-ēs/-īs
Genitive rēg-es/-is/-os/-us rēg-om/-um/-erum ignis ign-iom/-ium
Dative rēg-ei/-ī/-ē/-ě rēg-ebus/-ebūs
ign-i/-eī/-ē ign-ibus/-ibos
Ablative rēg-īd/-ĭd/-ī/-ē/-ĕ rēg-ebus/-ebūs
Locative rēgī rēgebos ignī ignibos

For the consonant declension, in the nominative singular, the -s was affixed directly to the stem consonant, but the combination of the two consonants produced modified nominatives over the Old Latin period. The case appears in different stages of modification in different words diachronically.[31] The nominative as rēgs instead of rēx is an orthographic feature of Old Latin; the letter x was seldom used alone (as in the classical period) to designate the /ks/ or /gs/ sound, but instead, was written as either 'ks', 'cs', or even 'xs'. The Latin neuter form (not shown) is the Indo-European nominative without stem ending; for example, cor < *cord "heart."[32]

The genitive singular endings include -is < -es and -us < *-os.[33] In the genitive plural, some forms appear to affix the case ending to the genitive singular rather than the stem: regerum < *reg-is-um.[34]

In the dative singular, -ī succeeded -eī and -ē after 200 BC.

In the accusative singular, -em < *-ṃ after a consonant.[33]

In the ablative singular, the -d was lost after 200 BC.[35] In the dative and ablative plural, the early poets sometimes used -būs.[35]

In the locative singular, the earliest form is like the dative but over the period assimilated to the ablative.[36]

Fourth declension (u)

The 'U-Stem' declension. The stems of the nouns of the u-declension end in ŭ and are masculine, feminine and neuter. In addition there is a ū-stem declension, which contains only a few "isolated" words, such as sūs, "pig", and is not presented here.[37]

senātus, –uos
senate m.
Singular Plural
Nominative senātus senātūs
Vocative senātus senātūs
Accusative senātum senātūs
Genitive senāt-uos/-uis/-ī/-ous/-ūs senāt-uom/-um
Dative senātuī senāt-ubus/-ibus
Ablative senāt-ūd/-ud senāt-ubus/-ibus
Locative senāti

Fifth declension (e)

The 'E-Stem' declension. The fifth declension in Old Latin is almost morphologically identical to the one of Classical Latin.

rēs, reis
thing f.
Singular Plural
Nominative rēs, reis rēs
Vocative rēs rēs
Accusative rem rēs
Genitive rēis, rēs rēsom
Dative reī rēbos
Ablative rēd rēbos

While the commonest ending in the nominative in both the singular and plural forms is '-ēs' (i.e. 'rēs, rĕī'), there have been recorded a few instances of either a shortened 'e' with the addition of a consonantal 'i', as in 'reis', or the abandonment of the nature of the 'e-stem' declension (i.e. 'res, rei').

The genitive in the singular functions as the second declension: 'rĕī' (the breve above the 'e' is the result of an approximant 'r' preceding a mid-open vowel). The genitive plural, in a like manner to the second declension, is formed primarily by '-ēsōm'

The dative is generally formed with an '-ei' in the singular, and an '-ēbos' in the plural.

The accusative, like all the other declensions, retains the labial 'm', shortening the quantity of the theme vowel.

The ablative singular is a predictable '-ēd.' The plural is like the dative.

The locative functions exactly in the singular as it does in the plural, with a short '-eis' as the 1st although there are no singular-based city names in the singular besides the occasional 'Athenseis'.

Personal pronouns

Personal pronouns are among the most common thing found in Old Latin inscriptions. In all three persons, the ablative singular ending is identical to the accusative singular.

Ego, I Tu, You Suī, Himself, Herself, Etc.
Nominative ego tu -
Accusative mēd tēd sēd
Genitive mis tis sei
Dative mihei, mehei tibei sibei
Ablative mēd tēd sēd
Nominative nōs vōs -
Accusative nōs vōs sēd
Genitive nostrōm,
-ōrum, -i
-ōrum, -i
Dative nōbeis, nis vōbeis sibei
Ablative nōbeis, nis vōbeis sēd

Relative pronoun

In Old Latin, the relative pronoun is also another common concept, especially in inscriptions. The forms are quite inconsistent and leave much to be reconstructed by scholars.

queī, quaī, quod who, what
Masculine Feminine Neuter
Nominative queī quaī quod
Accusative quem quam quod
Genitive quoius, quoios, -a, -um/om
(according to gender of whatever is owned)
Dative quoī, queī, quoieī, queī
Ablative quī, quōd quād quōd
Nominative ques, queis quaī qua
Accusative quōs quās qua
Genitive quōm, quōrom quōm, quārom quōm, quōrom
Dative queis, quīs
Ablative queis, quīs


Old present and perfects

There is little evidence of the inflection of Old Latin verb forms and the few surviving inscriptions hold many inconsistencies between forms. Therefore, the forms below are ones that are both proved by scholars through Old Latin inscriptions, and recreated by scholars based on other early Indo-European languages such as Greek and Italic dialects such as Oscan and Umbrian.

Indicative Present: Sum Indicative Present: Facio
Old Classical Old Classical
Singular Plural Singular Plural Singular Plural Singular Plural
First Person (e)som somos, sumos sum sumus fac(e/ī)ō fac(e)imos faciō facimus
Second Person es esteīs es estis fac(e/ī)s fac(e/ī)teis facis facitis
Third Person est sont est sunt fac(e/ī)d/-(e/i)t fac(e/ī)ont facit faciunt
Indicative Perfect: Sum Indicative Perfect: Facio
Old Classical Old Classical
Singular Plural Singular Plural Singular Plural Singular Plural
First Person fuei fuemos fuī fuimus (fe)fecei (fe)fecemos fēcī fēcimus
Second Person fuistei fuisteīs fuistī fuistis (fe)fecistei (fe)fecisteis fēcistī fēcistis
Third Person fued/fuit fueront/-erom fuit fuērunt (fe)feced/-et (fe)feceront/-erom fēcit fēcērunt/-ēre


  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.
  • Lua error in package.lua at line 80: module 'strict' not found.


  1. Lua error in package.lua at line 80: module 'strict' not found.
  2. Lua error in package.lua at line 80: module 'strict' not found.
  3. De Oratoribus, I.193.
  4. Book IX.1.6.
  5. Wordsworth 1874, p. v.
  6. Histories III.22.
  7. Lua error in package.lua at line 80: module 'strict' not found.
  8. Bennett & 1910 iii.
  9. Lua error in package.lua at line 80: module 'strict' not found.
  10. Allen (1897), p.6
  11. Lua error in package.lua at line 80: module 'strict' not found.
  12. Buck (1933), pp. 174–175.
  13. Wordsworth (1874), p.45.
  14. 14.0 14.1 Buck (1933), p. 177.
  15. Buck (1933), pp. 175–176.
  16. 16.0 16.1 Wordsworth (1874), p. 48.
  17. 17.0 17.1 17.2 17.3 Buck (1933), p. 176.
  18. Buck (1933), p. 172.
  19. Palmer (1988), p. 242.
  20. Buck (1933), p. 173.
  21. Buck (1933), pp. 99–100.
  22. Lindsay (1894), p. 383.
  23. Buck (1933), p. 182.
  24. 24.0 24.1 Sihler (1995), A New Comparative Grammar of Greek and Latin.
  25. Wordsworth (1874), p.56.
  26. Buck (1933), p.181.
  27. Lua error in package.lua at line 80: module 'strict' not found.
  28. Lua error in package.lua at line 80: module 'strict' not found.
  29. Buck (1933), p. 197.
  30. Buck (1933), pp. 185–193.
  31. Wordsworth (1874), pp. 67–73.
  32. Buck (1933), p. 185.
  33. 33.0 33.1 Bennett (1895), p. 117.
  34. Roby (1872), p. 162.
  35. 35.0 35.1 Allen (1897), p. 9.
  36. Gildersleeve (1900), p. 18.
  37. Buck (1933), pp. 198–201.

See also

External links

  • Lua error in package.lua at line 80: module 'strict' not found.