c-AFRASIAN-3_methodology.htm

PROTO-LANGUAGE PHONEMES

in IE and Afrasian

(Hieroglyphic Egyptian and Arabic)

(Nostratic Hypothesis)

(118)T[H]SA-F[H]A-¿E-P[?]FE ("stand-up-(imperfective)=standing-like-=erect+digit=erect-digit=finger ("penis")"); E: Db' {for **Dwjb (this word uses Gardiner : 1970 D50, ‘(vertical) finger', to indicate the medial /wj/)}, ‘finger, thumb, toe, digit'; A: ?aSba¿-un {S-w-y-b-¿}, ‘finger'; (IE: **(s)te/e:ibh-{s-mobile + **theweyebh-}, ‘**finger, penis', listed under *ste/e:ibh-, pole, stick, stiff, push together'); RATIONALE: For an extended explanation of the methodology that was employed to reach the reconstructions contained in this entry, click here.

a. In order to obtain the most convincing sets of cognates, we first scan the English entries for a given concept in Egyptian, Arabic, and IE dictionaries:

1) We choose the word ‘finger' for investigation.

b. We learn that, for ‘finger', the Egyptian word for it is Db'; the Arabic is ?aSba¿un; the IE word is *penk^we in many IE-derived languages. Almost unbelievably, Pokorny suggests that the word underlying English ‘finger' is based on an IE word meaning ‘five'. One would think that ‘fingers' came before ‘counting fingers'.

1) Of course, this derivation is ridiculous for several reasons:

a) Several IE number terms have main or variant forms ending in -w [or -u derived from it] (*dwo:(u), ‘two'; *k^wetru, ‘four'; *penk^we, ‘five' (if ^w can be equated with w), **se(i)pu- (in OHG sibun), ‘seven'; ok^to:(u), ‘eight'; (e-)new(en) (cf. Welsh and Cornish naw, ‘four' for the segmentation), ‘nine'; dek^u-, ‘ten'. Since IE is related to Egyptian through Nostratic, it will not surprise us to learn that all Egyptian numbers from 1 to 10 (and several beyond 10: djjjw, ‘fifty') end in -w. It is, therefore, most probable that the -w of IE numbers is a formant for numbers.

1)) An additional consideration that suggests an analysis of penk^we into penk + -w(e) is the Old Indian derivative: páñca-, ‘five'. The normal reflex of IE k^w in Old Indian is k while the normal reflex of k^ is sh. It is obvious that páñca- has arisen from **pán(t)sha-, in which the /sh/ has first palatalized the preceding -n- to ñ-, and then has infixed a -t- for ease of pronunciation: **páñ-t-sha-; and the resulting cluster /tsh/ is normally written as c in Old Indian. Therefore, the best reconstruction for ‘five', based on these considerations, should be **penk^-we-.


b) If we realize that IE *penk^we is probably better analyzed as **penk^- + -**w(e), it becomes probable that **penk^-, ‘finger', preceded **penk^-we-, ‘five'; and that 'five' is 'finger(s)-number'.

c) Derived forms like Armenian hing, ‘five', and OHG finf, both ‘five', as well as OHG fingar, ‘finger', suggest that, at least, one involved root should be emended to **peink^-.

d) Now an IE root that might be the basis for **peink^-, ‘finger', is found listed as *1. peig-, ‘scratch, decorate with color (perhaps, originally, ‘tattoo')'. This root has the alternate form *peik- but a look at some of its derivates (Tocharian B pink-, ‘write', Old Indian piMsháti, ‘decorates', suggest strongly that the best reconstruction is **peink-, and piMsháti makes the further emendation to **peink^ likely.

1)) This is true because velar nasals are very unstable in IE — so unstable, that few IEists have entertained the possibility that these phonemes, /ng/ and /nk/, should be reconstructed for IE; and have gone to great length to explain velar nasals in verbs as the result of metathesis: g/kn becomes . This seems even less likely in the word for 'finger'.

e) Therefore, we suggest that the composition of IE penk^we, ‘five', is **peink^-, ‘finger', and the number formant -w(e); and that **peink^-, ‘finger', should be interpreted as a derivation from **peink^-, ‘scratch', so that a ‘finger' should be understand as a ‘scratcher'.

c. However, it is certainly clear that this IE root for ‘finger' has nothing to do with either Egyptian Db' or Arabic ?aSba¿un. Contrary to the mistaken claims of the glottochronologists, very few early roots simply disappear; they are retained but re-defined. In order to determine the form of the IE root, for which we will look, we must do our best to reconstruct the Pontic-Nostratic root that will underlie the the Egyptian, Arabic, and IE form for which we will look.

1) Arabic roots are most frequently triconsonantal. Arabic ?aSba¿un has four consonants (?-S-b-¿), not counting the -n, which is simply the indefinite article as a suffix. While quadriconsonantal roots exist in Arabic, they are rare; and a look at the Arabic dictionary reveals that lexicographers have listed ?aSba¿un as a derived form of the root S-b-¿, ‘to make'; and we cannot be sure if the verbal meaning, which could be derived from ‘activity of the fingers', is secondary, and ?aSba¿un, ‘finger', is primary; or S-b-¿, ‘to make', is primary, and ?aSba¿un is secondary, meaning something like: ‘the objects with which we make things'.

However, by the table of correspondences we have developed, ?aSba¿un would correspond to PL T[H]O-P[?]/P[?]F(V)-¿(V).

2) Fortunately, the Coptic word derived from Egyptian Db', 'finger', exists as te:e:be in Sayidic-Fayumic. By the procedures we have originated, the presence of Coptic e: indicates an earlier Egyptian ai. We are fortunate to have the Egyptian Db.t, ‘brick', which appears in Sayidic Coptic as to:o:be, and is conjectured to be cognate with Arabic Tûbun (T-w-b), ‘baked brick', which can be related by our correspondence tables to PL T[?]O-F/F[H](V)-P[?]F(V).

a) Now PL T[?]O-F/F[H](V)-P[?]F(V) corresponds to IE d-w-bh-, and, with the normal voweling of IE roots, we should expect an IE **deubh- in a related meaning but no such IE root exists. However, we have one more chance to find it. When s-mobile is combined with any voiced stop of spirant, it is de-voiced. If s-mobile + **deubh- exists as a compound, it will appear as **steubh-.

Under *stew6-, ‘thicken, form itself into a ball', we find Greek stú:pho:, ‘draw together, make dense, hard', which would represent IE **stu6bh-. This is certainly an appropriate semantic range for a ‘baked brick'. Accordingly, we posit an IE root **deweHebh-, ‘**harden, dried/baked brick', attested in the form of **steu(6)bh-, and emend the reconstructed form to T[?]O-F/F[H](V)-?/H/¿/HH(V)-P[?]F(V).

Since Egyptian Db.t, 'brick, is once written >db.t, and we know that, from an early date, and were confused in the Egyptian hieroglyphic script, we should emend the Egyptian form to **dwjb.t. However, since Egyptian d can only correspond to T[?/H]A/E, we will emend the PL form to T[?]A/E-F/F[H](V)-?/H/¿/HH(V)-P[?]F(V); and assume that the Arabic T is the result of velarization (emphaticization)of t in the presence of the following sequence of uw.

3) In view of Sayidic Coptic as to:o:be, which represents an Arabic root of T-w-(?/h/H/¿)-b, and incidentally validates our hypothesis of Coptic o: representing a root medial aw; and from the pattern established by the Sayidic Coptic o:o:, we will now hypothesize that the e:e: of Sayidic Coptic te:e:be, ‘finger', analogously represents a root medial ay, which should show up in IE as -y-.

a) As we mentioned earlier, Arabic ?aSba¿un would correspond to PL T[H]O-P[?]/P[?]F(V)-¿(V), which, in turn, would correspond to IE t-b/bh-H. If the PL root had a medial -y- as the Coptic reflex of Egyptian Db' strongly suggests, and if Db(') is cognate with ?aSba(¿)un, then we might hypothesize that the PL root had the form T[H]O-¿E-P[?]F(V). While Arabic b can be either PL P[?] or P[?]F, Egyptian b can normally only be PL P[?]F or P[H]F, which is decisive for P[?]F.

b) PL T[H]O-¿E-P[?]F(V) would result in IE **teibh-. While there is no **teibh- with an appropriate semantic range, there is *steibh-, which includes Lithuanian sti:bis, ‘erect penis'. It is well-known that, in languages around the world, ‘finger' is often euphemistically used for ‘penis'. The reconstructed root *steib(h)- has the meanings of ‘pole, stick, stiff, push together'. Derived languages display meanings like ‘staff, rod', etc. The semantic range seems to encompass ‘tubular' and ‘push', which could well relate to ‘finger'.

However, we have one important consideration to consider at this point. Egyptian D, according to the tables of correspondence we have constructed, can only represent PL T[H]SA, T[H]SE, T[?]SA, or T[?]SE, the last of which is tempting because T[?]SE has been determined by us to have the base meaning of ‘finger' (cf. Arabic daiyun, ‘hand'; PL T[?]SE-¿E, "hand-like").

But the Arabic points strongly at an initial aspirate (T[H]O = Arabic S), which inclines us to favor T[H]SA/T[H]SE. The range of meanings we have established for T[H]SE ("porcupine, spread out/open, bristle out, frightened, spiny, fuzzy, move slowly") do not seem like a probable base but those for T[H]SA ("ass, rear up, lift up, stand up, thick, excessive, kick, trample, prance") seem much more appropriate.

Therefore, we will hypothesize that PL T[H]SA, in the sense of ‘lift up', is the probable initial of the word we are seeking. T[H]SA appears in IE only with s-mobile as *sta:-, ‘stand, place', which, according to modern notation would be **stheH₂-.

However, as Old Indian tiSThati clearly shows, the IE root should be reconstructed as **stha:-, with the Old Indian Th reflecting not a laryngal/pharyngal (H) but the normal complex reflex of PL T[H]S. The cerebralization (better, velarization) of ST is the result of the absorption of w, which is attested as *sta:u-. As we have indicated in our table of correspondence, an IE long vowel can be occasioned by a laryngal/pharyngal but also by an aspirated voiceless stop. Therefore, *sta:u- is the normal expected reflex of PL S[H]O + T[H]SA-F[H]A, s-mobile + **thaw-.

Now, in neither Egyptian or Arabic is the PL S[H]O attested in this word (Egyptian s; Arabic s).

So, if the words for ‘finger' include the element T[H]SA, ‘stand', this will appear as st(h)- in IE but D in Egyptian.

We can interpret the composition T[H]SA-F[H]A as "stand-(imperfective)", with the resulting meaning of ‘standing'. And under IE *sta:-, we indeed find *st(h)a:u- with exactly this meaning: e.g., Lithuanian stovùs, ‘standing (of water)'. But, as we have seen throughout this work, PL ¿E is frequently added to stems to produce an adjective. We can see this theoretical T[H]SA-F[H]A-¿E in Lithuanian stóviu, ‘stand', and can reconstruct IE **stha:wi-, ‘**erect'.

Since P[?]FE means ‘digit', and the sign for Db', is, in fact, a ‘vertical', an ‘erect' finger, let us look at whether it may be possible that the Db segment of Db' should be an Egyptian reflex of T[H]SA-F[H]A-¿E-P[?]FE, "erect-finger", in fact the exact description of the Egyptian determinative.

If we suppose that Db- is a reflex of T[H]SA-F[H]A-¿E-P[?]FE, we can represent it in Egyptian symbols as **Dwjb-. But the rules abstracted from the relationships we have studied, this would result in a theoretical Sayidic Coptic form **j/to:e:b-, which we can readily understand might have assimilated to **j/te:e:b-, comparing favorably with the form we actually find attested in Sayidic Coptic: te:e:be. The reflex of Egyptian D is either j (/dzh/) or t in Sayidic Coptic. An assimilation of two originally different vowels (o:e:) would explain the doubled vowels in the attested Coptic form (e:e:).

Before we return to investigate the Arabic form further, let us inquire as to whether there many be an IE expressions of the theoretical T[H]SA-F[H]A-¿E-P[?]FE which can be identified.

By the rules we regularly employ, T[H]SA-F[H]A-¿E-P[?]FE would be represented by IE **t(h)V(:)weyebh, or, with normal reduction: **t(h)wibh. Now IE *twi/i:bh- does exist in the meaning ‘tubularly hollow'. However, the derived Latin ti:bia, ‘shinbone', suggests that ‘hollowness' may not be an essential part of the meaning of this stem. And, since we now that PL T[H]SE means ‘(porcupine) quill/spine', and these are ‘hollow', it looks very much as if IE *twi/i:bh- represents both T[H]SA-F[H]A-¿E-P[?]FE and T[H]SE-F[H]A-¿E-P[?]FE.

In the article on *twi/i:bh-, Pokorny concludes the entry with: "gehört kaum zu sti/i:b(h)-, ‘Stange' (oben S. 1015), das festes s- hat."

Now, this is the entry, *ste/e:ib(h)-, ‘pole, stick, stiff, push together', that we had cause to investigate earlier (see above). And when we see Latvian stìebrs, ‘rush', and others, it looks very probable that *ste/e:ib(h)- and *twi/i:bh- (from T[H]SE-F[H]A-¿E-P[?]FE) are variants of the same root, with absorption of the w when s-mobile is employed, to prevent an initial consonant cluster of **stw; conversely, T[H]SA-F[H]A-¿E-P[?]FE seems to be present in *ste/e:ib(h)- and *twi/i:bh- as well, although it is probable that ‘ti:bia' represents T[H]SA-F[H]A-¿E-P[?]FO, "erect-leg", rather than T[H]SA-F[H]A-¿E-P[?]FE, "erect-digit".

On the basis of these considerations, we feel it is not inappropriate to reconstruct PL T[H]SA-F[H]A-¿E-P[?]FE, ‘erect-digit', and then: ‘finger', reflected in Egyptian **Dwjb-. The final ' of the attested Egyptian form Db' is the formant for tools T[?]SO, which may account for the -t in IE forms like Lithuanian stiêbtis, ‘to pull one's self up to full height'. The existence of ' in Db' is probably attributable to confusion was Db'.w.t , ‘signet, seal', where a formant indicating a tool would be appropriate, and the meaning of ‘tool for pushing together' might roughly describe the function of a ‘signet'.

It remains to be seen whether the Arabic root we have extracted (S-b-¿) can be related to **Dwjb, or better: can it be related to PL T[H]SA-F[H]A-¿E-P[?]FE. Now this word would have appeared in Pontic-Nostratic as t^hsawayap^?fa. If the stress-accent caused the deletion of stress un-accented vowels, we might expect something like t^hswi(/ya)p^?f. If the consonant cluster t^hsw needed to be simplified, it might have been simplified to t^hw with deletion of the s, which would have made our root t^hwaip^?f so that an original t^hsawa would have had the same form as a syllable derived from PL T[H]SO, namely t^hwa. The Arabic reflex of t^hwa (and, we will assert t^hwa) is S, so that T[H]SA-F[H]A-¿E-P[?]FE would appear in Arabic either as **S-y-b (t^hswayap^?f), which appears not to exist in Arabic, or as **S-b (t^hswip^?f), which is probably seen in Sabba, ‘cast, mold', which may have a distant relationship to IE *ste/e:ib(h)-, ‘pole, stick, stiff, push together'; and, in this context, it is interesting to note Arabic qaSaba-tun, 'reed').

There are a number of ways in which **S-b, ‘**finger, could have been made into a verb. A device would be to add ¿, which we have identified as a perfective element derived from PL ¿A; and Saba¿a exists, meaning ‘point out with the finger'. Once this verb had been formed, and was in general use, from the proclivity to prefer triconsonantal roots, the natural next step would have been to rename the ‘finger', the ‘pointer(-out)': ?aSba¿-un.

To return to the Egyptian form, we remember that the Sayidic Coptic form is te:e:be. I believe that the most economical explanation of the final -e in this word, and the frequent Coptic final -e in words that terminate in -.t in hieroglyphically written Egyptian, is to assume that an unindicated j preceded the terminations, which produced phonetically /ai/, which became final /e/ after the disappearance of final -t.

4) To summarize, will suggest that there was a PL root, T[H]SA-F[H]A-¿E-P[?]FE, "erect-digit=finger",

a) that is represented in Egyptian by **Dwjb, presently glossed as Db‘. This accounted for the **Dwjb segment of the word, which was additionally expanded by -j and ', which accounts for the final -e of the Sayidic Coptic form te:e:be;

b) that is represented in Arabic as **S-b, presently glossed as S-b-¿, which was additionally expanded by -¿, which accounts for the final -¿ of the Arabic form ?aSba¿-un

c) that is represented in IE as **theweyebh-, presently glossed as *(s)te/e:ib(h)-, which was additionally expanded by s-mobile which accounts for the initial s- of the IE form *ste/e:ib(h)-. The initial cluster stwe- was simplified to ste with the deleted w sporadically lengthening the vowel: ste:-. The meaning ‘finger' is not attested for this root but the meaning ‘penis' is, and we assert, in view of the established relationship between AA and IE, and the fact that the forms reflected in the Arabic and Egyptian forms would yield IE *ste/e:ib(h)- as discussed above, that the primary meaning was ‘finger', which was extended to ‘penis' but then subsequently lost.

5) As a final comment, we hypothesized above that another PL form, for which we could anticipate a very similar outcome in the three languages under discussion, was T[H]SE-F[H]A-¿E-P[?]FE, ‘porcupine(-plural)-like-digit=quill=hollow=reed'. We saw above in Latvian stìebrs, ‘rush', that this form, seen better in *twi/i:bh-, ‘pipe-like, hollow', apparently could also be modified by s-mobile (*ste/e:ib(h)-).

But, it may be of some interest to know that this word, also, is probably represented in Egyptian by **Db.w (incorrectly glossed Db3.wDb(3).w is glossed as ‘mat of leaves(?)', which may be a close guess for ‘**mat of floating rushes/reeds'. According to this analysis, this would have represented Egyptian **Dwjb, also. Unfortunately, there is no simple Arabic reflex that we can find but qaSaba-tun, 'reed', with the expected S-b is suggestive since we would expect medial semivowels to be suppressed when forming a triconsonantal root; however, the process of prefixing a semantic element to the root like q(anû, '[Assyrian:] pipe') which does not presently occur in Arabic, is otherwise unknown to me.

One of the things we try to do when making these comparisons is identify and differentiate other words in the affected languages that appear, at least on the surface, to be homonymous. One such in this context is Egyptian Db3, ‘stop up, block'. We will be able to suggest an Arabic cognate after a short discussion of the proposed IE cognate.

As in Db3, ‘reeds', I believe Egyptologists have incorrectly assumed a final 3 on the word Db3, ‘stop up, block', because it is a part of another similar word (‘repay, clothe, adorn') written with the same sign but with the addition of the sign for 3. But even if Db3 were the correct reading for the meaning ‘stop up, block‘, it will not be considered in the comparisons offered because it is a verbal element (from RE, ‘apply') which does little more than verbalize a nominal concept in cases where the root is already at least two or three consonants. Because Db(3) is written with the same sign (+3) as Db(3).w, ‘floats', we must assume that it had the same or very similar medial elements. Since we believe that we have identified the medial elements of Db(3).w, ‘floats' as wj, we will, therefore, provisionally reconstruct this word as Dwjb(3).w.

One possibility for a PL equivalent of this is T[?]S(A/E)-F(V)-¿E-P[?]F(V). This would be equivalent to IE **dheweyebh-. However, there is a possible Coptic reflex of Db(3) in Sayidic to:p, ‘stitch, stop, caulk', and the interesting compound mehto:p, ‘needle, peg' (meh = ‘fill'). The most frequent Egyptian equivalent of Coptic p is p rather than b (PL P[?]A/E).

It would be preferable, of course, to be able to cite a cognate from the standard literary language but dialectal Tâba-tun (T-w-b), ‘stopper', is a substantial match. Arabic b can correspond to both Egyptian p and b.

Although no IE **dhwibh- can be found, there is *dheubh-, ‘peg', which has derivatives like MHG tübel, ‘block, peg, plug, nail'. This is the exact semantic range for which we have been looking, and we can provisionally assume that IE *dheubh- is likely to be the corresponding IE term. In addition, IE has *dheubh-, ‘deaf', listed incorrectly in Pokorny as a derivative of 4. dheu-, ‘smoke, whirl', which is almost certainly wrong. Nothing would be more natural than to characterize a deaf person as having his ears ‘stopped up, plugged'. Interestingly, this form corresponds with Egyptian Db(3)- exactly if the PL initial syllable is T[?]SA or T[?]SE, the only possibilities if the Egyptian D is to be assigned its normal values, the T[?]SO suggested by the Arabic form. At the very least, the agreement between the IE and Arabic forms, however, suggest that the medial sequence was F(V) rather than F(V)-¿E; and if Coptic to:p is related, it also suggests one medial semivowel.

Since all the related Egyptian forms are transliterated as D, we will provisionally assume that the Arabic form is aberrant, and that the underlying PL word is T[?]SA/E-F(V)-P[?]F(V). Since a ‘plug' certainly is related ideationally to ‘finger', we will identify the final element as P[?]FE, ‘digit'; and because the most frequent employment of F(V) is as a sign of the imperfective, we will hypothesize: T[?]SA/E-FA-P[?]FE. Now a ‘plug' can be looked at something that ‘keeps something in' or, alternatively, something that ‘lets something out'. PL T[?]SE means ‘release', and if the word is reconstructed with this element, we have T[?]SE-FA-P[?]FE, "releasing-digit=release-plug". T[?]SE-FA-P[?]FE would produce IE *dheubh-, Egyptian **Dwb(3), and Coptic to:p, if Egyptian b has, in this instance, the rare Coptic reflex p.

This leaves only the aberrant Arabic Tâba-tun (T-w-b), ‘stopper', to understand and hopefully to explain. With the correspondences we have established, T-w-b would normally reflect PL T[?]O-F(A)-P[?]F(E) (but not T[?]SE-FA-P[?]FE). The simplest explanation, though it would take other examples from the same dialect to substantiate it, is that in this dialect, a form like the expected **dâba was pronounced **tâba (but we might also consider a dialect that, like Greek, will not permit a sequence of two aspirated stops: **thaph becomes **taph), and in a derived root-form like **tauwaba, the initial t was velarized (emphaticized) to T. This is not an ideal explanation but, on the strength of the available evidence, the most plausible explanation I can construct. It seems rather unlikely that Tâbatun is not related to dheubh- and **D(w)b(3) in some manner.

Apparently, for the Egyptian, the result of an earlier medial sequence (o:e: from wj) was close enough to o: (from w) that the same sign could be employed, at least in the stage of Egyptian at which we find it.

Finally, we may look at Egyptian Db3, 'repay, replace, restore', which has the Coptic to:o:be, 'repay, requite'. The Coptic o:o: suggest a medial segment of two semivowels like wj in to:o:be, 'brick', discussed above. There, we concluded that Db.t, 'brick', represented Egyptian **Dwjb.t.

In preparation for a discussion of Db3, 'repay, replace, restore', we will need to return to the topic of **dwjb.t, 'brick', which showed us that, because of the Egyptian confusion between D and d, on the strength of Arabic Tûbun, it was reasonable to assume that a word spelled with Db could actually represent **db

Above, we partially reconstructed a form T[?]A/E-F/F[H](V)-?/H/¿/HH(V)-P[?]F(V) to account for **Dwjb.t, 'brick', and Tûbun, 'brick'. We will now attempt to specify a little more closely the undefined second medial element.

IE has a root *da:u, 'burn', which, however, has derivatives in Latvian (zhaut, 'dry, smoke'), which point to an alternate form of *da:u- or a second root grouped for convenience under that heading: *de:uti-. This is certainly the semantic range we would expect for 'dried/fired brick'. On this basis, we will hypothesize that the root-form above contained the basis T[?]A-FA-HHE, "seep-(imperfective)-go-out-from=dried-out", corresponding to IE *de:u-/*da:u- (**deweH₄-), yielding Egyptian **dwjb.t and Arabic **t-w-H-b.

Now we do have Arabic tâba (t-w-b), 'repent, come back to God'. Based on what we have seen, and the Coptic vocalism, I believe this Arabic form could well be a development from **t-w-y-b. The Arabic form this would have had before contraction, is tawayaba. The velarization (emphaticization) of the hypothetical t of **tûbun might simply be a result of u being selected as the form vowel (**tuwaHaba); and I will hazard the conjecture that the underlying form for 'brick' is T[?]A-FA-HHE-P[?]FE, "seep-(imperfective)=seeping(-and-contracting)-go-out-from=dried-out-foot=dried-brick".

In the case of Db3, 'repay, replace, restore', I hypothesize that the underlying form was T[?]E-FA-¿E-HE-P[?]FE, "spin-around-(imperfective)=turning-around(-and-returning)-like=returned+go-across from+foot=return"; Egyptian **dwjjb. This basal idea seems to be able to faciliate the Egyptian and Arabic nuances rather well.

This explanation has the advantage of reconciling the two Coptic forms of to:o:be under almost identical forms for Egyptian: **dwjb and **dwjjb.

The IE form ste/e:ib(h), discussed above in conjunction with 'finger', has another grouping of meanings, apparently restricted to Latin, that are of interest in this context. Under the subheading *2. ste/e:i/p-, we have Latin stips, ‘money, gift, contribution', and stipendium, ‘income, stipend, salary, pay'. I suggest that the basal idea here is a "return" for services rendered, and that this root is a reflex of T[?]E-FA-¿E-HE-P[?]FE (discussed above).

I hypothesize that the divergent p of *2. ste/e:i/p- in Latin stips derives from an IE form **deweyeH₄(e)bh-, in which the precursor of Latin devoiced the final bh under the influence of the preceding H₄.

6) Some well-meaning critics have asserted that the composition of these cognate comparison sets is simply a matter of observing superficial similarities.

Since the process described above is that followed essentially for every cognate entry, the description of the critics of the method employed here, whether or not they agree with the results of those methods, is simply inaccurate.

the latest revision of this document can be found at
HTTP://WWW.GEOCITIES.COM/Athens/Forum/2803/c-AFRASIAN-3_methodology.htm

Patrick C. Ryan * 9115 West 34th Street - Little Rock, AR 72204-4441 * (501)227-9947
PROTO-LANGUAGE@email.msn.com