PERSPRO3A.htm (2/10/98)

published in

Dhumbadji!, Vol. 1, No. 4, Winter 1994


IE -l/-n Nouns

by Patrick C. Ryan

9115 West 34th Street

Little Rock, AR 72204

(501) 227-9947

April 1994


1. Introduction

Proto-Language as used here designates that language from which all of the world's languages are descended at a point in time just prior to the dispersal of evolving humanity, leading to ethnic and linguistic differentiation.

2. The Problem

A phenomenon that has successfully resisted explanation since the inception of Indo-European studies is the odd variation of final -l in the nominative-accusative case stem-form opposed by -n in the oblique cases stem-form of a few important nouns.

Probably the most notable of these is (P)IE sa/a:wel, 'sun`, which has the oblique case stem-form swen-.

Another probable example of the same phenomenon (though other interpretations may be possible) is to be found in k^emel, 'heaven` (opposed to k^emen-).

This specific type of variation is so rare that it is not really discussed in Brugmann (1888: I-V), which is so complete in most other regards, though, of course, -r/-n variations are described1.

3. The Hypothesis

I will demonstrate that the Proto-Language had an animate definte article, nh, which underlies PIE final -l in sa/a:wel, and that the Proto-Language also had an inanmate definite article, n, which underlies PIE final -n in swen-. In the derived languages, these articles became pronouns of the third person singular in a neutral social context for active verb constructions.

4. Discussion

General Phonology

I substantially agree with Bomhard (1994) in his latest reconstruction of Proto-Nostratic phonemes. However, I assume that the nasals of Proto-Language (from which Proto-Nostratic was evolved through Proto-Pontic), appeared in unaspirated (m, n, ng) and aspirated forms (mh, nh, ngh), the aspirates leading to Proto-Nostratic sequences of nasal + long vowel (mV:, nV:, and (n)k[h]V:), using Bomhard's notation.

This contrast of aspirated against unaspirated phonemes was fundamental to Proto-Language, and was based on the association of aspiration with animate concepts and lack of aspiration (glottalization) with inanimate concepts, circumstances which I have discussed at length elsewhere2.

Bomhard, having now added a dorsal nasal (ng) to his inventory of Proto-Nostratic phonemes, has informed me that he is not yet convinced that a dorsal nasal could also occupy a syllable-initial position. If it could not, it would be the only phoneme reconstructed for Proto-Nostratic (or PIE) that could not occupy any position an oddity of itself. If Bomhard rejects ng in initial positions, surely a reflex of it (e.g. g / k[h] / k' / G / q[h] / q') should be reconstructed for Proto-Nostratic.

Also, Bomhard posits Proto-Nostratic l as well as additional laterals of the type tl. If Bomhard means simply a voiceless l la the Nahuatl spelling convention, I could, perhaps agree since I believe his Proto-Nostratic l/tl could only have developed from Proto-Language from or as allophones of r(h) or n(h) through Proto-Pontic.

The reflexes of these Proto-Language nasals are as follows:

4. a. Proto-Language Definite Singular Elements


Proto-Language nh, 'wave`


A Proto-Language word of the form nha, meaning 'wave`, is indicated by the Egyptian sign for n (Gardiner 1973: Sign N35: "ripple of water") (n), which contrasts with the sign for 'nose` (na) [see below]:

Egyptian *n, 'wave'


Sumerian offers some support through na-3, *flow [??] (Jaritz Sign 159, which pictures a cultivated field with water for irrigation (Jaritz 1967:50). But cf. also Sumerian nab/p, 'river, ocean` (PL nh + p?f, 'leg, place`).


The reconstruction of the vowel -a for nha is based primarily on the PIE reflexes: e.g. 1. la:-, 'roar of waves, swing back and forth (motion of waves)`; lab- (for *la:b-), 'wet a little, lap` (cf.Egyptian np(3), 'be wet`); laku- (for *la:ku-), 'collection of water`; la:ma/a:, 'wet place, marsh`; and, lat- (for *la:t-), 'wetness, marsh` (cf. Egyptian nt, 'water`), among the obvious choices.

Proto-Language n, 'nose, stone`



Egyptian has a series of demonstratives on the base n, which are construed as neuters. They are nn, nw, nf, n3 (-i and -ii formants, which occur with the masculine base p and feminine base t, are not attested with n).

The form nw is written the 'jar` ([i]nw) + chick (w) + a sign which has been misinterpreted as an earlier form of the 'adze`-sign, which is

Egyptian *(i)nw, 'adze'

The sign which is written (often in addition (!) to the 'adze`-sign) is

Egyptian *n, 'nose'

which is a 'nose in profile (my characterization)`, albeit in a horizontal orientation; accordingly, we assume Egyptian *na, '*nose`, and hence, Proto-Language n, 'nose`.


Egyptian n meaning 'stone` can probably be identified in inr, 'stone`, which is cognate with PIE ond-, 'stone` (cf. in Old Indian dri-, 'stone`; and Akkadian nar, 'stele', a borrowing from Sumerian na ru-2-a, 'erected stone'). See also below under Usage.



It does not seem possible to identify Proto-Language n, 'nose`.


We have na-4, 'stone` (Jaritz 1967: Sign 453). The sign (Jaritz 1967: Sign 970) which reads nig-2, 'thing, something`, also reads ni-3, which may represent Proto-Language n, 'stonelike (thing)`. See also below under Usage.



Proto-Language n, 'nose`, occurs in the compound PIE nas-, 'nose`, and had originally the meaning 'nasal passage`, which Pokorny confirms. The reconstructed PIE form is derived from Proto-Language "n-so (PL s, 'skin`), a designation of its outward form.

This Proto-Language word occurs in numerous compounds listed in Pokorny without final -s under sna:-, 'flow` (Germanic sna/ub-, sna/up-, sna/ud-, sna/ut-, sna/ug-, sna/uk-). Since s- is a recognized prefix with a rather neutral (or undefined meaning), it seems clear that, from the variety of root formatives (b p d t g k) all combined with na or nu (from *nau; PL nf, 'group of noses`), that the main nominal/verbal idea of 'nose` resides in na.


This meaning can, perhaps, be seen in PIE ond-, 'stone` (cf. in Old Indian dri-, 'stone`, which is cognate was Egyptian inr, 'stone`; and monu-, '(hu)man-stone(s : dual), testicle(s), male` (see below under Usage). Although only a lexicographic mention, we do have Old Indian na, meaning 'jewel, pearl`, which compares with Sumerian na-4, 'stone, jewel'. Notice also Old Indian nu, 'weapon` ('stones?`), also lexicographic only.

Phonological Reflexes

The reflexes of PL -nh(a) are

and of PL -na are


In Proto-Language, -nha, conveyed a definite animate singular nuance (cf. Sumerian na-6, 'man`; ni-4, 'lady`); -na conveyed a definite inanimate singular nuance. Both, however, retained their original meanings of 'wave` and 'nose/stone` as well. When they were used as adjectives (classifiers), they stood with stress-accent before the nouns which they modified or without the stress-accent after the nouns by which they were modified.

1. "nh mho, 'the (known) human`

2. "mh nha, 'humanoid one (animate)'

3. "mh na, 'humanoid stone = testicle`

4. "ph nha, 'mousey one (animate)`

At a later stage of Proto-Language, compounds were formed by the combination of two syllables, with the stress-accent of the first syllable producing elision of the final vowel but retention of the glide:

1. b. (")nhmhw(o), 'name`

2. b. (")mhwnh(a), 'the human`

3. b. (")mhn(a), 'testicle, male`

4. b. (")phnh(a), 'the mouse`

Combined, these latter can probably be seen in PIE en(o)mN, 'name` (cf. Hittite la-a-ma-an, 'name`), Sumerian nam, 'something, status of ...` (1. b.); Egyptian m(i)n, 'someone` (2. b.); and PIE monu-s, 'man, human`, Egyptian *mn, 'male` (the 'phallus`-sign, reads mn); *mn, 'bull`; Sumerian man, '*male` (Jaritz 1967: Sign 822 pictures two testicles, reads man (for *mn [??]), and means 'comrade, brother`) (3. b.); and PIE pel-, '*mouse`6.

The simplex, Proto-Language mh can be seen in Egyptian m, 'who, what?` and Sumerian mu-9, '(hu)man`.

An early compound of three elements is Proto-Language ?e-"mn (PIE me:n-) from *?e-"m-n(a), which meant 'the called out (measured) thing`, used for 'moon` as the measurer of time (from Proto-Language ?e, 'away from`, and m, 'tongue, converse`)7.

Evidence in the Derived Languages


Since Proto-Language nh and na both evolved into Sumerian na/, it was not possible to use na to distinguish definite singular animate and inanimate concepts. Therefore, a selection was made so that PL nh as -n- appears for the animate third person prefix in verbal forms and in the third person personal pronoun an-e and in the possessive -a-ni (and in the nominal animate plural [see Final Comments below]); PL p?, 'piece, one (indefinite inanimate) as -b- appears in lieu of n for the inanimate third person prefix in verbal forms and in the possessive -bi.

This interpretation is further supported by na-6 (Jaritz 1967: Sign 611), '(hu)man`; and ni-4 (Jaritz 1967: Sign 921), 'lady` from PL nh, '(hu)manoid`.

We also have Sumerian an(a) (Jaritz 1967: Sign 831), 'one`, which reflects PL ?n(a), 'one here (inanimate)`; and ana-3 (Jaritz 1967: Sign 1), 'one`, PL ?nh(a), 'one here (animate)`.

That a selection was made is made virtually certain by the forms of the Sumerian interrogatives: a-na, 'what?`, and a-ba, 'who?`, which represent Proto-Language ?a-n and ?a-ph (from ph, 'flea, feeder, (married) male`), compounded with the PL interrogative ?a, 'forehead, place of focus, here?`8.


Though it is not generally acknowledged by Egyptologists, the simplest explanation for the so-called Egyptian genitival particle is that it is a definite article which follows and agrees with the noun that precedes: -n (feminine -n.t; plural -(i)nw). It is, therefore, in the normal post-ancient position for an Egyptian nominal modifier.

Thus, a phrase like pr-n z, 'the house (of) (the) man`, is somewhat analogous to Arabic baitu-r(l)-rajuli.

There is an acknowledged relationship between Arabic and Egyptian through their common parent, Afro-Asiatic.

The l of the Arabic article, ?al, can only correspond to demonstrative forms in Egyptian n since only r and n can represent l in Egyptian; and no demonstrative-formative/stem containing r exists in Egyptian. Egyptian 3 never corresponds to Proto-Semitic or Proto-Indo-European l, whatever some may have supposed.

The element n is readily identifiable as a demonstrative adjective element in Old Egyptian in masculine pn / feminine tn / dual and plural ipn/itn (also iptn), 'this/these`. In these variously compounded demonstratives, p (Proto-Language ph, 'flea, feeder, (married) male`) bears the masculine idea (or possibly p, 'buttock, piece` [gender indifferent {??}])while t (Proto-Language th, 'accompanier, large group, herd`) provides the feminine one.

In these combinations, n may be replaced by -w9, -f10, -311, -i12, and -ii13, with some exceptions.

The prefixed i-element is the Egyptian reflex of the Proto-Language indefinite plural elements H, 'many` / , 'much`, which are used to designate a few relic plural forms in oldest Egyptian.

Since Egyptian n reflects both Proto-Language nh and na, it is not possible to specify exactly which it represents.

Egyptian n also serves as a base for demonstrative pronouns In Egyptian contrasting with p/t, which are mostly used for demonstrative adjectives.

One of the forms is usually transcribed nn. If nn represents a form parallel to pn and tn, it would be redundant unless n has some other meaning in addition to those described above.

Ordinarily, we assume that doubled letters in Egyptian represent two similar syllables, but with the 'wave`-sign, we already have the precedent that

Egyptian mw, 'water'

is usually read mw, and means 'water`, which almost certainly is related to Arabic yammun, 'sea`, and should be emended to read *im.w, which we analyze as derived from the now familiar indefinite plural elements H, 'many` + mh, 'human, wanderer, pond` + fh, 'wolf(-pack)`, animate definite plural14.

I will show that the doubled 'wave`-sign

reads not nn but *in, and thus corresponds in form to theoretical *ip and *it, representing Proto-Language , 'many` + n, 'stone, nose, one (inanimate definite singular)`.

Firstly, the doubled 'wave`-sign is also used to signify one of two Egyptian negatives: nn, which contrasts with n.

Coptic, the last spoken stage of Egyptian, knows two related negatives: n and an. While it is possible that an older *nan could develop into an, it is more natural to think that an developed from a form which would normally have been written in Egyptian as *in.

Secondly, though the negative was early written n-n, it also was written with a sign which is probably the doubled form (though connected) of 'forearm`, which latter is:

The doubled sign is described as 'arms in gesture of negation`, and has the form:

Since this sign was also used for the preposition n, which had the recorded early form of *i-n, with other prepositional analogs (ir, im), it seems relatively probable that it represented *in not *nn as presently widely held 15.

Further support for the idea that the doubled 'wave`-sign reads *in rather than nn is found in the fact that the demonstrative pronoun, nn, in Middle Egyptian, is frequently written with a special doubled sign representing 'two rushes with shoots` which singly is used almost exclusively as a determinative:

Egyptian (i)nn

Aside from the demonstrative pronoun, the most familiar words using this doubled sign are nn(ii), 'be weary`, and nn.t, '(lower) heaven`.

In Arabic, we have ulla, 'be sick, ill`, aly?un, 'sky`, ?al, 'be unable`, and ?ul, 'these (masculine/neuter)`.

Sequences of Afro-Asiatic l and ?l would both result in Egyptian in; and it seems probable that the Arabic words above are cognate with the Egyptian *in, 'weary, inert`, *in.t, '(lower) heaven`, *in, 'not`, and *in, 'this (better '*these`)`16.

Though plurals in PIE have ceased to be (or never were) formed by a prefixed H, 'many`, or , 'much`, it appears that the earliest Egyptian (dual-)plural, indicated simply by doubling or tripling of word-signs, was formed with prefixed i-, and so was roughly analogous with Arabic "broken" plurals.

An interesting example of this oldest method of forming plurals is found in Proto-Semitic yad-, 'hand`, which represents Proto-Language , 'much` + ts?, 'finger` (cf. plural ?aydin). "Broken" plurals with initial ?ay- may be dissimulations from an earlier a- but other "broken" plural with ?a- are unexplained unless they are copies of the stress-accentual pattern which has been separated from its proper morpheme.

continue to Part Two?

additional copies of this file are available at


Patrick C. Ryan * 9115 West 34th Street - Little Rock, AR 72204-4441 * (501)227-9947