10. Numerals

     (In the process of being written)

Notes:

1. I have departed from IPA in the representation of mid vowels, and follow the long established academic practice when describing Vulgar Latin and the development of Romance languages, namely:

   [e]and [o] denote any front unrounded or back rounded mid vowel respectively, when the degree of height is either unknown, indeterminate or irrelevant;

   [ẹ] and [ọ] denote specifically high versions (IPA [e] and [o]);

   [ę] and [ǫ] denote specifically low versions (IPA [ɛ] and [ɔ]).

2. Also [ə] denotes:

   - generally any reduced non-defined centralized vowel without reference to height;

   - in Late Britainese, a unstressed non-phonemic low-mid to near-low ([ɜ] to [ɐ]) central vowel, depending upon regional pronunciation.

10.1 Cardinal numbers

10.1.1 Numbers 1 through to 10

All the Romance languages without exception derived their cardinal numbers 1 through to 10 from Vulgar Latin which, for the most part, is like Clasical Latin (without, of course, the final --m of septem, novem and decem). The important difference is that 4 is derived from *quattro and 5 from cīnque. The numbers in Britainese, therefore, are:

one	ún [in]
two	dous [duz]
three	trais [trəi̯z]
four	quatr [kwatr̩]
five	cinq [t͡ʃiŋk]
six	sais [səi̯z]
seven	set [set]
eight	oit [wit]
nine	nov [nov]
ten	dege [ded͡ʒ]

Notes:

• ún has an acute accent to distinguish it from the unstressed indefinite article un "a, an";
• In earlier Britainese 1 and 2 has feminine une ['ynə] and doues ['du.əz];
• quatr is pronounced [kwatr] before a vowel or silent h, e.g. quatr cans [kwatr̩kanz] "four dogs", but quatr homns [kwatromn̩z] "four men";
• 5 is written cinq rather than cinc due to the influence of cinquant "50" (see below).

10.1.2 Numbers 11 through to 19

Except for Romanian, which has unsprezece, doissprezece, treisprezece &c. "one on ten, two on ten, three on ten, &c." that are clearly calques of the corresponding Slav numerals, all other Romance languages have forms derived from Vulgar Latin and there is no reason to suppose that Britainese would not have done so also.

The Latin septendecim, dudēvīgintī, ūndēvīgintī "17, 18, 19" did not survive in Vulgar Latin and were replaced by forms meaning "ten (and) seven, ten (and) eight, ten (and) nine" respectively.

The other numerals were derived from Vulgar Latin 11 to 16 (but we find that Old Spanish se(d)ze (← *sedce ← sēdecim) = "16" has given way in modern Spanish to diez y seis and to Portuguese dezaseis); there is no reason to suppose that Britainese would have behaved similarly). In the Vulgar Latin of Gaul and the Iberian peninsula, the unstressed ending -decim [-dekɪ(m)] gave way to *[-dkʲe] (cf. subsection 4.4.2.1 on the Consonants page). Also, we should note that the initial u- of undecim "11" seems to been short [ʊ] in some dialects, including those of Gaul and the Iberian peninsula (cf. Wiktionary undecim)

The numbers in Britainese, therefore, are:

eleven	oundge [und͡ʒ]
twelve	dodge [dod͡ʒ]
thirteen	tredge [tred͡ʒ]
fourteen	quatordge [kwa'tord͡ʒ]
fifteen	chindge [kind͡ʒ]
sixteen	sedge [sed͡ʒ]
seventeen	dedgeset [dəd͡ʒ(ə)'set]
eighteen	dedgeoit [dəd͡ʒ'wit]
nineteen	dedgenov [dəd͡ʒ(ə)'nov]

10.1.3 The decades

In Romanian the decades "twenty, thirty, ..., ninety" are douăzeci, treizeci, ..., nouăzeci, ("two-tens, three-tens, ..., nine-tens"); but in all other Romance languages they are derived from Vulgar Latin equivalents of the Classical Latin numbers. The latter were all formed with the element -gint- [gɪnt] ← Proto-Indo-European *-dḱm̥t- "decade":

• 20 = vīgintī. Initial vī- may be for an earlier *dvī-; what is certain is that the final -ī was an old dual ending, but in Latin the word was an indeclinable adjective. A contracted form vinti is attested and the evidence of the Romance languages is that this was the Vulgar Latin form. Dalmatian venč, Italian and Corsican venti, Romansh ventg and Romagnol vènt reflect the pronunciation [vɪnti], but many others, e.g. Occitan and Old French vint reflect [vinti] where the final [i] has clearly influenced the pronunciation of the first vowel.
• 30 = trīgintā [triːˈɡɪntaː] where, as in the remaining decades, the final -ā was originally a neuter plural ending; but just like 20 and all the other decade words, it is an indeclinable adjective. The Romance languages show that in Vulgar Latin the word was contracted to ['trɪnta] .
• The decades 40, ..., 90 all ended in -āgintā ^†; in Iberian Proto-Romance it seems to have been contracted to *-inta [ɪnta], presumably under the influence of ['trɪnta], but elsewhere the Vulgar Latin contracftion was *-anta. Note also that everywhere the initial quadr- of quadrāgintā "40" became *quarr- in Vulgar Latin.

† As well as octāgintā "80" Classical Latin also had the alternative octōgintā; this did not survive in Vulgar Latin.

In modern French we find that 70 is soixante-dix "seventy-ten", 80 is quatre-vingts "four-twenties" and 90 is quatre-vingt-dix "four-twenty-ten", ; this vigesimal counting is often attributed to Celtic influence. We must, therefore, consider whether similar Celtic influence caused similar vigesimal numerals in Britainese. In English in our timeline we do indeed find examples of vigesimal counting; one well-known example are the opening words of Abraham Lincoln's Gettysburg address: "Four score and seven years ago ...". Perhaps the most well-known is the King James Bible translation of Psalm 90, verse 10: "The days of our years are threescore years and ten; and if by reason of strength they be fourscore years, yet is their strength labour and sorrow; for it is soon cut off, and we fly away." The Hebrew, however, merely has שִׁבְעִ֪ים šiḇ-‘îm "seventy" and שְׁמ֘וֹנִ֤ים šə-mō-w-nîm "eighty" respectively.

In the Middle Ages we find more extensive examples in French of counting by scores or twenties, e.g. deus vins, trois vins &c., just as there was in English of that period. But both Old and Middle French had words for 70, 80 and 90 inherited from Vulgar Latin: se(p)tante = 70; oitante/ (h)uitante and the semi-learned octante = 80; nonante = 90. Quite why vigesimal forms from 70 to 99 have ousted these older forms in standard French of is unclear. The older decimal system is still persists:

• septante, huitante, nonante in the Swiss cantons of Vaud, Valais and Fribourg, and in Acadia in Canada;
• septante, octante, nonante in Belgiumß, Democratic Republic of the Congo, and Rwanda.

As for Britain, it should be noted that Welsh now generally uses a strictly decimal system, the traditional vigesimal numbers being used mainly for telling the time and giving one's age up to about 30; thus while the French are happy to say quatre-vingt-dix-neuf "four-twenty-ten-nine" for 99, the Welsh have abandoned pedwar ar bymtheg a phedwar ugain "four on fifteen and four score" in favour of naw deg naw "nine ten nine"! Also despite the undoubted influence of the King James Bible ("Authorized Version") counting in scores or twenties has never become standard in English.

In short, I am sure counting in twenties would have been known in Britainese in the Middle Ages, as it was in our timeline in England and France, but that the evidence of English and Welsh in our island in our timeline suggests that the Classical and Vulgar Latin decimal system would have prevailed; and thus the decades in Britainese are:

twenty	vint [vint]
thirty	trent [trent]
forty	quarant [kwe'rant]
fifty	cinquant [t͡ʃiŋ'kwant]
sixty	saissant [səi̯'sant]
seventy	setant [sə'tant]
eighty	oitant ['wi'tant]
ninety	nonant [nu'nant]

Notes:

• novant [nu'vant] is sometimes found for "ninety" under the influence of nov "nine". But this usage is considered substandard.
• In French, English and Welsh 45 is respectively quarante-cinq, forty five, pedwar deg pump, that is, the units numeral follows the decade numeral without "and" or any other particle. However, in French we find et "and" used if the unit is "one", e.g. 31 = trente et un. However, English has just thirty one and Welsh tri deg un. I am sure that in BART Britainese will follow the English and modern Welsh of our timeline, e.g.
  31 = trent un; 45 = quarant cinq; 86 = oitant sais; 99 = nonant nov.

10.1.4 Hundreds, thousands and more

The Latin centum "100" and mīlle "1000" will become cent [t͡ʃent] and mil [mil] in Britainese. Both of those words were indeclinable adjectives in Latin and remain unchangeable in Britainese.

In Classical Latin 200, ..., 900 were declinable plural adjectives: ducentī, trecentī, quadringentī, quīngentī, septingentī, octingentī, nōngentī.. Some probably survived in Vulgar Latin for a time as we see in Spanish quinientos and Portuguese quinhentos ← Latin accusative quīngentōs. But the formation 'five' + 'hundred' generally became the norm, e.g. Catalan cinc-cents; French: cinq cents; Italian cinquecento; Occitan cinc cents. As in our time line we also have English five hundred and Welsh pum cant, I have no doubt that in BART Britainese will have cinq cent - written as two words and, as in Italian, English and Welsh, no plural ending on cent.

In British English we have "and" following "hundred" before other numerals, e.g. "two hundred and thirty four", but neither French deux cents trente quatre nor Welsh dau gant tri deg pedwar require any conjunction or particle, neither will Britainese dous cent trent quatr.

In Britainese mil will behave in the same way as cent, thus e.g.
2026 = dous mil vint sais
74,813 = setant quatre mil, oit cent tredge
528,149 = cinq cent vint oit mil, cent quarant nov
 

Latin had no word for million, billion &c. So here we enter a more speculative realm. It is quite possible that in BART millione appeared in early Italian and was adopted into other languages, becoming millioun [mi'ljun] in Britainese. It is possible also that around the 15th century we find billion and trillion being used in France to mean 10¹² and 10¹⁸ respectively and then the subsequent confusion between the 'long scale' of the original use and the 'short scale' of those who came to use the two words to mean 10⁹ and 10¹² respectively. I discuss this in greater detail when dealing with TAKE numerals. As the details of the history of BART have not been developed, it is pointless to pursue this further here (TAKE, however, is an artificial language developed in our timeline, not in BART or any other timeline).

10.1.5 Zero

Latin had no word for the mathematical concept zero and there is no Roman numeral denoting it. The closest we have in Latin is nihil "nothing", often contracted to nīl; it did not survive in Vulgar Latin and there are, therefore, no inherited Romance forms of the word. However there have been direct borrowings in many languages; perhaps the most well-known is "nil" used in (Association) football to mean no score. But such usage is outside of our consideration here (Was soccer played in BART?).

Latin also had the adjective nūllus (masc.), nūlla (fem.), nūllum (neut.) = "no, not any." The adjective was also part of Vulgar Latin and has survived in many Romance lasnguages, see Wiktionary nullus. We find in a manuscript from 525 AD Dionysius Exiguus using to assist in date-reckoning a masculine [sic] noun nulla = "zero"; it was used extensively by Bede in his writings and elsewhere in Medieval Latin. It survives as an invariable masculine noun in Italian with the meaning "nothing."

Although forms derived from nūllus do, as we have seen, survive in some Romance languages, the Romance languages generally have different words forthe adjective "no" and for the indefinite pronoun "nothing". This will be true of Britainese also. But it is very likely early Medieval Britainese had null [nyl] meaning "zero"; later when the pronunciation shifted to [nil] it may well have become confused by those who knew Latin with the Latin nīl, but I think null [nil] would have prevailed.

In our timeline in the early 13th century the Italian mathematician Fibonacci coined a Latin word zephirum from Arabic صِفْر (ṣifr) "nothing; cipher" to denote "zero." This is supposed to given Italian zero whence early Spanish zero (written cero in modern Spanish) and French zéro and hence zero in English and many other languages. It is not improbable that there would have been a borrowing from Arabic صِفْر in BART as Arabic numerals replace the old Roman ones, as surely they would have one. But whether it would have given rise to zero must remain speculative. It is safer to stick with null [nil].

10.2 Ordinal numbers

10.2.1 In early Britainese

The Latin ordinals from 1st through to 5th seem to have been preserved in Vulgar Latin and appear in early Britainese as:

first	prim [prim], primair [pri'mai̯r]
second	autre ['au̯trə], segond [sə'gond]
third	terce [tert͡ʃ]
fourth	quart [kwart]
fifth	chint [kint]

For 6th, 7th, 8th and 9th were found:

• seist ← sextu(m) as well as seisme ['sei̯mə] on analogy with with setme ['setmə], see below; cf. Spanish sesma besides siesta, both now surviving as feminine nouns with specialized meanings,namely, a traditional unit of length, equivalent to about 13.9 cm, being a sixth of a "vara" and an afternoon nap taken about the 6th hour (i.e. noon).
• setme ← *settmu ← septimu(m)
• oitme ['oi̯tmə] ← *octmu cf. Old French uitme
• novme ['novmə] ← *novmu cf. Old French neufme

These are given in the masculine singular. Unlike descriptive adjectives, these always came before the noun just as ordinals do in our timeline in French, English and Welsh (except for cyntaf = 1st); this remains so in all periods of Britainese.

In early Britainese we find deime ['dei̯mə] ← *decmu ← decimu(m); but this soon gave way to degeme ['ded͡ʒmə] under the influence of dege and by analogy with ordinals 7th to 9th. Hence also oundgeme, dodgeme, tredgeme ... 11th, 12th, 12th etc.

For ordinals of decades the Latin endings -gintī, -gintā both became -gēsimus; these did not survive in Vulgar Latin and ordinals derived from these which are found in some Romance languages are learned borrowings. The ending -ēsimus of centēsimus "hundredth" and millēsimus "thousandth" did persist in some regions. It is found as -esimo in Italian where starting with 11th we drop the final vowel of the cardinal number and add the suffix –esimo / –esima e.g. undicesimo "eleventh", ventiduesimo "twenty-second", centesimo "hundredth".

In Old French the vocalization of the suffix is somewhat odd; it generally appeared as -isme, e.g. onzisme "eleventh", quinzisme "fifteenth". Possibly these were influenced from disme "tenth" ← *decmu ← decimu(m). Soo we also find vintisme "twentieth", quarantisme "fortieth", centisme "hundredth." In the later 12th century the s became silent and was sometimes omitted in spelling. From the 13th century onwards -ie(s)me, taken from dialect, appears and ousted -i(s)me leading to modern French -ième.

In early Britainese we find -esme ← *-esmu ← -ēsimu(m), e.g. vintesme "20th", cinquantesme "20th", centesme. "100th."

10.2.2 In later Britainese

With the dropping of final unstressed [ə] the suffix -me that we found on the ordinals 6th through to 15th became simply -m [m̩] and it was extended gradually to the ordinals 2nd to 5th thus, dousm [duzm̩], traism [trəi̯zm̩], quatrm [kwatrm̩], cinqm [t͡ʃiŋkm̩]; of the two variants of Early Britainese 1st, only primair [pri'mai̯r] survived.

The suffix -esm of the decade numerals and hundred did not persist but gave way to the simpler -m, thus e.g. vintm "20th", cinquantm "20th", centm. "100th."

Thus in modern Britainese, apart from primair "1st", all other ordinals are formed by suffixing -m [m̩] to the cardinal number. In the case of compound numbers the suffix is, like English -(e)th and French -ième, added to the last numeral only, e.g. vint traism "23rd", quatr cent trent cinqm "435th", dous mil vint saism "2026th." They would not, of course, normally be written out in full but rather as: 23m, 435m, 2026m respectively.

There remains one question to answer. Does Britainese retain primair in compounds just as English retains first or does it merely suffix the ordinal ending to un in compounds? i.e. Is 21st in modern Britainese vint primair like English "twenty first" or is vint unm like French vingt-unième?

If we look Welsh, our other 'control language', its modern colloquial version offers no help since from 11th onwards it merely uses the modern decimal system after the noun, e.g. "the twenty first child" is yr emyn dau deg un (cf. English "hymn 21"). Literary Welsh, however, is quite clear: while "one" is un, "first" is cyntaf, e.g. un bachgen "one boy" ~ y bachgen cyntaf "the first boy", in compound forms we use unfed not cyntaf, e.g. un ar bymtheg (one one fifteen) "sixteen" ~ unfed ar bymtheg "sixteenth"; un ar ar hugain "twenty one" ~ unfed ar hugain "twenty first."

Both French and Welsh have a word unrelated to their word for one to denote "first" meaning the ordinal that precedes all others in series, but suffixes the ordinal ending to un (French unième, Welsh unfed) for "first" in compound ordinals, thus:
21st in Britainese is vint unm [vin'tinm̩].