Re: VMs: Arguments against a code book?

[Jorge Stolfi <stolfi@xxxxxxxxxxxxx>]

  > [Nick Pelling:] I guess I'd be a ~little~ more convinced if there
  > wasn't half a millennium between the Italian and Latin texts you
  > had to hand.

True... however, my impression is that the spelling and vocabulary of
Italian, in particular, have changed surprisingly little since the
middle ages. I learned only modern Italian, mainly from reading
magazines, and modern Portuguese, at school and in life; yet I can
understand the "Divina Commedia" of Dante (1265 - 1321) much better
than the "Lusíadas" of Camões (~1524 - ~1580).

A while ago I spent quite a bit of time preparing Culpeper's Herbal
(1652) for analysis, only to find that its statistics were not much
different from H.G.Well's "War of the Worlds" (~1900) and Agatha
Christie's "Poirot at Styles".

On the other hand I have recently got hold of the "Towneley Plays"
(Middle English, ca 1460; was it here that I got the hint?), and they
are strikingly different from those "modern" english texts in many
regards. The TPs were meant to be recited from stage-wagons to
common folk on the streets, and perhaps for that reason they use very
few long words:

  [Evil angel:]
    Thou art so fayre vnto my syght,
    thou semys well to sytt on hight;
    So thynke me that thou doyse.
    
  [Good angel:]
    I rede ye leyfe that vanys royse,
    ffor that seyte may non angell seme
    So well as hym that all shall deme.

Incidentally, that feature reminds me of an obscure funny manuscript... 8-)

But you are right, I should get some Italian text of the 1500's.
However it is surprisingly hard to find electronic texts that are (a)
in prose, (b) not translations, (c) not in modernized spelling, (d) at
least as long as the VMS, and (e) reasonably well proof-read.
Suggestions are gratefully welcome...

  > I stand by my assertion that Italian and Latin are ~reasonably~
  > close, perhaps even close enough for the same basic
  > writing/encoding system to be used (with only a few changes)
  
I would guess that the biggest change from Latin to Italian, as far as
statistics are concerned, was the loss of the Latin declension system.
Italian must use prepositions where Latin would use gentives,
accusatives, etc. Moreover, the number of possible endings for the
same stem was drastically reduced, and word order became more rigid.

Another big change was the introduction of the definite article, which
is perhaps the most easily detectable syntactic feature of Romance (and
Germanic) languages.

Until the 19th century, it seems, many linguists would claim that the
Chinese language "had no grammar" -- apparently because it has no
articles, no gender or number inflections, no verbal inflections, and
does not even distinguish clearly between verbs, nouns, and
adjectives. Moreover it seems that succintness has always been highly
valued in Chinese culture, even more so than in English culture (but not
in Portuguese culture, as you must have noticed 8-), and old Chinese
texts tend to be very "telegraphic":

  人 參 一 名 人 銜 一
  名 鬼 蓋 味 甘 微 寒
  生 山 谷 補 五 藏 安
  精 神 定 魂 魄 止 驚
  悸 除 邪 氣 明 目 開
  心 益 智 久 服 輕 身
  延 年
  
  ren2 shen1 other name ren2(?) xian2(?) other name gui3(?) gai4(?)
  taste sweet little cold grow hill valley help five viscera calm xxx
  spirit stabilize ethereal-soul corporeal-soul stop fright heart-flutter
  eliminate evil energy brighten eye open heart sharpen wits long use
  lighten body extend year
  
This is the whole "ren2shen1" = "ginseng" entry from the Shennong
Bencao Jing, perhaps 1st century BC. Not a single wasted word, I would
say... Contrast it with Culpeper's:

  Common Nightshade hath an upright, round, green, hollow
  stalk, about a Foot or half a yard high, bushing forth
  into many Branches, whereon grow many green Leavs, somwhat
  broad and pointed at the ends, soft and full of Juyce,
  somwhat like unto Bazil, but larger, and a little unevenly
  dented about the edges at the tops of the Stalks and
  Branches, come forth three or four or more white Flowers
  made of five smal pointed Leavs apiece, standing on a
  Stalk together, one above another with yellow pointels in
  the middle, composed of four or five yellow threds set
  together which aftewards turn into so many pendulous green
  Berries of the bigness of smal Pease, full of green Juyce,
  and smal whitish round flat Seed lying within it.  The
  Root is white and a little woody when it hath given Flower
  and Fruit with many smal Fibres at it; The whol Plant is
  of a waterish insipide tast, but the Juyce within the
  Berries is somwhat viscuous, and of a cooling and binding
  quality. [... lots more ...] 

I very much doubt that one could detect any hint of syntactic
structure in an un-punctuated Chinese text by studying word
colocation. I suspect that Latin would not be much better in that
regard. But in Romance or Germanic languages one could probably
discover the articles, perhaps also the number/gender agreements.

All the best,

--stolfi


______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list