MVS 'eight' (Re: Behind the 8-ball)

[Koontz John E]

> Michael McCafferty asks:
> As many of you know, the Miami-Illinois term for "eight" /paraani/ is, as
> Bob Rankin pointed out in an article in IJAL several years ago, a
> borrowing from a Siouan language. Tutelo typically gets the nod.
>
> Can anyone suggest why this happened? No, not that Bob wrote it up, but
> that such a borrowing occurred. It's one of the strangest things. I
> imagine, since we're talking numbers, that it was borrowed probably in the
> process of trading. But does the number 8 have any mythological meaning?

An alternative analysis that I have offered in the past is that
Miami-Illinois paraare (later palaani) 'eight' might come from a source in
Mississippi Valley something like Ioway-Otoe.

The Tutelo form (per Oliverio) is pala'ani, in the system:
           Tu
'two'      noN'oNpaa
'three'    la'ani(N)
'six'      aka'aspee ~ akaaspe'e
'seven'    saako'omiNiN
'eight'    pala'ani(N)

Disregarding variation in vowel marking for length and accent, the
variants are essentially palali and palani.  This reflects l > n / __ VN -
the final vowel of 'three' is iN - an allophonic change not always
indicated in Hale's transcriptions.

Biloxi and Ofo have
           Bi           Of
'two'      noN'pa       nuN'pha
'three'    da'n(N)      ta'ni(N)
'six'      akaxpe'      akape'
'seven'    noN'pahudi   fa'kumi(N)
'eight'    dan'hudi'    pa'tani(N)

In Mississippi Valley we find:
           Te            OP            IO            Wi
'two'      nuN'pa        naNba'        nuN(uN)'we    nuNuN'p
'three'    ya'mni(N)     dha'bdhiN     da(a)'<ny>i   daani'
'six'      s^a'kpe       s^a'ppe       sa(a)'gwe     ha(a)kewe'
'seven'    s^ako'wiN     ppe'naNba     sa(a)'hmaN    s^aagoo'wiN
'eight'    s^aglo'gha    ppe'dhabdhiN  greeraa'briN  harumaN'k

The reconstructions supported by these forms are:

'two'      *nuNuN'pa
'three'    *raa'priN
'six'      *s^aa'kpe ~ *(a)kaa's^pe (with metathesis)
'seven'    *s^aakoowiN ~ *s^aakwaN  (Winnebago looks like a Dakota loan)
'eight'    ???

For 'eight' we find Dakota s^aglo'gha, which follows in the *s^aak-
series, but is otherwise unattested, Winnebago harumaN'k, which seems to
be something like 'lying on it (by hand)', and a collection of forms that
seem to be derived from 'three', i.e., presumably forms that indicate
something like 'five' (or a hand of fingers) + 'three'.  In Dhegiha and
Biloxi there are parallel forms for 'seven' based on 'two'.  (Note that
other Dhegiha doesn't always agree with Omaha-Ponca; Osage, for example,
has hki'etopa, apparently referring to a pair of fours.)

The formations for 'eight' forms based on 'three' differ, but Tutelo and
Ofo seem to have *pa or perhaps *hpa followed by their respective reflexes
of 'three', while Dhegiha has *hpe followed by its reflex of three.
Dhegiha also uses the *hpe-construction with 'seven'.  The meaning of
*(h)pa or *hpe is unknown, but it might perhaps be *hpa 'nose, head'
(pan-Siouan) or *hpe 'forehead' (MV only).  Speculatively finger gestures
unward or near the head might have been opposed to ones elsewhere to
indicate the upper five of the decade.

IO has an indiosyncratic formation with *kre + 'three'.  What is
interesting here is that the form of 'three' exemplified is not the form
of 'three' found in the simple form.  IO and Winnebago agree in having
reflexes of *RaaniN for 'three'.  The simplification of the medial cluster
*pr to *R is exhibited elsewhere in these languages in initials of noun
(cf. IO nyiN(iN), Wi niNiN for 'water') and medially in 'bean' (IO
uNnyiN<ng>e, Wi huNuNniN'k) vs. Da mniN 'water' or omni(N)c^a 'bean'. So
-raabriN in greeraa'briN seems a poor match with IO historical phonology.
It stands out from daa'<ny>i like tertiary or ternary or trinary stand out
from three in English.

I would suggest then that IO 'eight' is probably a loan from a loan from a
MV language where raabriN or something like it was the usual form for
'three'.  The big problem with this is that no other Siouan language has
the *kre- construction for 'eight' (or even 'seven').  This is either an
insurmountable problem or an interesting suggestion that we don't have a
full range of Siouan languages to work with, depending on your point of
view.

Whenever I'm out on this limb, I always ask myself, if greeraa'briN is a
loan, what might it have replaced?  Presumably something within the range
of 'eight' forms we know from other Siouan languages, and one of those
possibilities there is a form based on *hpa or *hpe + 'three', which, if
IO 'three' is any basis on which to judge, would have been something like
*hpaRaaniN or *hpeRaaniN.  Either of these forms seems to me to be as
likely a source of Miami-Illinois paraare ~ palaani 'eight' as the Tutelo
(or Ofo) forms, though, of course, there is little to chose between with
any of them as far as form, once you start mapping "r" sounds.

As to how such a form could have gotten into Miami-Illinois, the usual
scenario with Tutelo (or Ofo or a hypothetical early form of Biloxi or
some form of Proto-Southeastern) is trade interactions leading to MI
borrowing of the source numeral system, or, at least of some of its
numeral terms, with 'eight' ending up a fossilized relict of this
situation.  But if the source is MV Siouan we have a new scenario to
consider.  We know that the Michigamea at least among the Illinois tribes
spoke a rather different language, and we have some evidence that it may
have been a Siouan language.  In that case, perhaps, a Siouan form for
'eight' might be a relict of the fusion of Siouan-speaking groups into MI.
In which case, perhaps paraare or palaani may represent the Michigamea
form of 'eight', which, as I've shown, is rather like what a hypothetical
*hpa-based "regular" form of 'eight' might have looked like in early IO.