What sort of 'soul' is the Egyptian ka?

Oddly enough, all this latest talk in my commentbox of the European Urnfield culture and Etruscan cremation practices led me to thinking about Egyptian beliefs about the afterlife. I suppose that lacking comprehensive resources on the Etruscan concept of death and afterlife, something which I doubt exists yet, I look to the Egyptians for even a shade of insight, particularly since there are so many connections to be made. (Remember that the Etruscans used scarabs as tomb offerings directly proving that they were affected by Egyptian beliefs in some form or fashion.)

Now, as I understand it, the Egyptian spirit is often said to be divided into three parts: the ka, the ba and the akh . These names are modern Budgisms based on the vowelless spelling of the Egyptians but I gather that they were pronounced more like *kuˀ, *baˀ and *ˀaḫ respectively at around 1500 BCE. I notice the Wikipedia under Egyptian soul currently claims there are five components of the Egyptian soul but then piles on even more concepts in its schizophrenic, multi-author account. Added is the sheut 'shadow' (or rather *šawīt), the ib 'heart' (*ˀib) and the ren 'name' (more accurately *rin), helping to thoroughly confuse the reader rather than elucidate. I consider these last three related but incidental to the fundamental Egyptian notion of 'soul'. In the following I want to explore a new idea that came to me.

The ka is sometimes translated 'soul', sometimes 'image' and sometimes 'double'. Many other translations are also attempted. Yet I wonder recently if this odd translation confusion exists because this term wasn't just referring to the spirit itself as modern theologians might understand it but instead to the containment of the spirit in a vessel, a vessel such as a body or statue. If so, this gives a whole new nuance to the *Ḥáˀat-Kuˀ-Patáḥ, a famous temple in the city of Memphis. It would then suggest that this temple wasn't just housing the 'soul of Ptah' (*kuˀ Patáḥ) but that it was so named because it was believed, at least by its cult leaders, to be the temple in which the very body *and* soul of Ptah was present. This gives me a vision of a grand building within which an impressive, monumental statue of Ptah sits, around which rituals were enacted by its devout priesthood as if he were the living article, all for the wonder of the entranced commonfolk. Such a subtle notion of the word *kuˀ would emphasize this particular temple's central importance to the worship of this god of death and resurrection. Why go to just any temple of Ptah when one can experience Ptah's holy spirit 'in the flesh'? It would certainly have been great temple marketing if my perception here is realistic.

Also by reconceiving of the ka as the 'soul' specifically when contained in its physical manifestation rather than just 'soul' proper, contrastively then the ba must be the spirit itself, particularly when it was separate from the *ẖīˀat or body. It would be the ba, I think, that is most appropriately translated as the English-speaking notion of 'soul'. So maybe we can get away with thinking of the ba as a *subset* of the ka rather than on a par with it.

This theoretical structure of an ancient belief system leads me to wonder if the akh was in turn yet another subset (of the ba, that is). In the resulting reinterpretation, the soul really isn't divided into three parts, or even five. Instead the soul is composed of three layers with the akh, as 'life-force' or 'will', being the innermost of the three metaphysical strata. In this ontology, the 'will' (akh) is a component of the 'soul' (ba) which is further merely an ingredient in the union of body and soul (ka).

I'll have to explore this idea further and see what evidence is in its favour or against it.

Taking away another root

Ever since I began distinguishing between well-grounded Proto-Indo-European (PIE) roots and mistaken ones built on historical confusion and an unhealthy denial of a non-IE Mediterranean-centered language family^[1], I can't stop finding more issues to write about! I sometimes wonder if the methodology used by Indo-Europeanists truly contrasts that by a Nostraticist or Proto-Worlder, but I try hard to suppress this rebellious, philosophical notion. I'm convinced now that many beloved PIE roots are in reality nothing more than a mirage built on a package of loanwords diffused from an Etrusco-Rhaetic language in the bustling Po Valley of early 1st millennium BCE. The previous reconstructed etyma that I've suspected openly on this blog show specific distribution patterns that imply such a geographical origin. I dare now suspect another cherished root: PIE *h₁em- 'to take'.

First of all, Julius Pokorny's 1950's work is online showing his pre-Laryngeal-Theory reconstruction, *em-, justified by evidence found in Celtic, Italic, Baltic, Slavic and Anatolian. At first blush, my skepticism appears too sensational. Yet we can first quickly peck away at the falsely segmented Hittite word *u-emiyami based on *actual* wemiyami 'I find'. The word has been connected with a quite different root, *gʷem-(ye-) 'to come', which happens to be a more convincing etymology.^[2] The habit of spelling the syllable we with two symbols, ú and e, instead of just one was a normal Hittite practice, it turns out.^[3]

So now we're just left with comparanda from Celtic, Italic, Baltic and Slavic - precisely the language groups easily accessible by trade from the Po Valley around 1000 BCE. The meaning of the verb also lends itself well to the language of trade and we may note the coincidence of Latin emere 'to buy' (hence caveat emptor 'buyer beware').

For the purposes of this revision, an Etrusco-Rhaetic verb *em 'to take' is in order to provide the source for the surrounding Indo-European forms. Etruscan evidence is rather easy to find. It's long been noted, thanks to the numerals expressed in the Liber Linteus, that numbers between 10 and 100 whose last digit is higher than 6 are conveyed by subtractive terminology. So while 13 in Etruscan is ci-śar (simply a compound of 'three' followed by 'ten'), 17 on the other hand is ci-em zaθrum, literally meaning 'three minus 20' much like Latin duo-de-viginti '18', literally '2-from-20'. Yet how do we analyse the grammar of this expression? Considering that the infinitive is expressed as the bare verb stem itself as in many other languages, this element em may very well not just mean 'minus' as the Bonfantes suggest but more specifically 'taking away from'. Thus ci-em zaθrum is literally 'taking away three from (ci em) twenty (zaθrum)'. Just like a proper SOV language is supposed to do, the infinitive verb is placed after the noun phrase and thus we know that 'three' is what is being taken away. Its simple preterite form, eme, is written twice in the continuous script found on a cup (TLE 366): nacemeuruiθalθileniθaliχememesnamertanśinamulu. In both these instances, in fact, the verb happens to follow two very common adverbs, nac 'when' and iχ 'thus', guaranteeing that I've properly divided these words.

We now need to explain what this root is doing in a non-IE language and the easy answer would be to blame it on loaning from an IE language into Etruscan, perhaps from a language like Umbrian (cf. emantur) or Latin (cf. emere). Yet lacking other evidence outside of this Po Valley trading circle, how can I be sure that this IE root even exists? Lacking further evidence, Occam's Razor guides me to the simplest answer, that it was my aforementioned Etrusco-Rhaetic root *em that spread throughout the northern territory into surrounding Indo-European languages. In a manner of comical speaking, these early IE peoples would not only have taken in novel goods from afar through exchange, but also would have taken the very verb 'to take' as well!


NOTES
^[1] Surely at the very least it's undeniable that the non-Indo-European language, Minoan, was positively influential on local languages up to 1400 BCE. Study of such a Mediterranean language family is abundantly warranted yet so entirely neglected.
^[2] In Hittite historical phonology (1999), p.297 (see link), Kimball must assume a compound dependent on a preverbal particle *u preceding an Anatolian root *em-, however note the original etymology given that doesn't require such a faith-based word-slicing in The Classical journal, vol 31 (1936), p.452 (see link) and Sturtevant, A comparative grammar of the Hittite language (1933), p.90 (see link) where the verb is derived squarely from *gʷem-ye-, a construction further attested outside of the Anatolian branch. As per the next note, the division of we- into u-e- based solely on spelling constitutes unnecessary assumption.
^[3] Melchert, Anatolian historical phonology (1994), p.25 (see link).