The PIE *to-participle in my subjective-objective model

I'm exploring an interesting idea involving participles, related to my previously concluded model of PIE conjugation involving two dimensions contrasting subjective with objective (ie. the source of hi- and mi-class respectively) and progressive with non-progressive. I had come to this model in order to explain how Anatolian-Tocharian dialects relate to the other dialects which I call Core IE. As with all things theoretical, I don't know whether I'm completely legitimized by the existing facts in exploring this idea, so take this all with a grain of salt and take thrill in the cerebral journey.

In a nutshell, if my quadripartite system distinguishes four sets of endings exemplified in the 1ps with *-mi (objective progressive), *-m (objective non-progressive), *-h₂ór (subjective progressive) and *-h₂e (subjective non-progressive), then it stands to follow that there may likewise be four non-finite forms, participles, corresponding to each of the four categories I describe. From the evident participle suffixes, we then seem to be inevitably led to the following system:

	objective	subjective
progressive	*-ónt-	*-m(h₁)nó-
non-progressive	*-tó-	*-wós-

However, Szemerényi informs us in Introduction to Indo-European linguistics (1996)^[1] that the PIE *tó-participle is just not present in Tocharian and Anatolian. From this absence of evidence, it's understandably concluded that the participle hadn't yet formed. Only once Anatolian and Tocharian parted ways would the emerging Core IE dialects create this new participial form.

This status quo account is admittedly very persuasive... as long as one forgets to question how such a suffix can be formed from known PIE grammar specifically with the required semantics to make it the prevailing participle form by far, above all other possible thematic suffixes like *-nó-, *-mó- and *-ló- among others which are also occasionally used. Why did all of the Core IE dialects agree to this one suffix with *-t-? I have a defiant answer: Maybe it had been a participle ending right from the start and that the Anatolian-Tocharian area were motivated to chuck this one ending away. But then, why?

Falling back on my recent insights on how Anatolian-Tocharian emerged out of my model (note too my later relabelling of "eventive" as "progressive" in this model), I realize one interesting motivation for a conjectural loss of this participle. Notice that my theory suggests that Anatolian-Tocharian dialects were developing tense out of a tenseless system, making the former progressive marker *-i a present tense marker. In effect, the four-way system of old was reshaped into a three-way system of mi-class, hi-class and middle. That means that one of the participles had to go, and guess which one! So these dialects must have ended up with a mi-class participle in *-ónt-, a hi-class participle in *-wós- and a middle participle in *-m(h₁)nó-. There would have been no longer any room for a *tó-participle in this particular evolution as it would only duplicate the function of one of the other three, hence a loss specifically in Anatolian and Tocharian of a now-redundant element.


NOTES
^[1] Szemerényi, Introduction to Indo-European linguistics (1996), p.323 (see link): "The suffix -to- is widespread in all IE languages except Anatolian and Tocharian." Well that was pretty straight-forward, wasn't it?

A thought on the real name for the land of the Minoans

On page 844 of Bromiley's The International Standard Bible Encyclopedia (1988), an intriguing and researched summary of the true name for the land of the Minoans is to be discovered:

"In an Ugaritic text concerning the abode of Kothar-wa-Ḫasis, the god of artisans, the word kptr occurs: kptr ksu ṯbth ḥkpt arṣ nḥlth, 'Caphtor is the throne of his sitting, Ḥkpt the land of his inheritance' (UT, `nt VI:14-16). The passage seems to preserve a memory of a connection with Crete as the home of their crafts; Ḥkpt may be another name for Crete or one of its regions. Economic texts from Mari speak of Kaptara, and an Akkadian text from Ugarit refers to ships arriving from Kapturi. C. H. Gordon had raised the question whether the words kpt-r and ḥ-kpt may include some morphological elements, a preformative ḥ- and a sufformative -r, leaving kpt as the basic word (Ugaritic Literature [1949], p.23 n.1), and relating this to Egyptian kft-yw. But the persistance of r in Hebrew, Akkadian and Ugaritic forms, plus the fact that final -r could become -yw by phonetic decay (see CAPHTOR II), rather support kptr/kftr as the original word."

So we can conclude with some degree of confidence that the name sounded something like *Kʰaputar, give or take some variation. For the sake of argument, I will suggest this specific form and what follows is a whole lot of speculation. I personally think of speculation as a necessary tool in the learning process, used to invigorate new paths of research and I always strive to improve my arguments with facts and evidence or to eventually abandon them, whichever logic guides me to do. However, for those overly evidentialist sorts that find any speculation without evidence too distasteful to even express, you may be spared reading further.

I'm very interested in the Minoan language and have been pursuing a hunch for years that the likeliest relationship it has is to the 'Etrusco-Cypriot' languages (Etruscan, Lemnian, Rhaetic, Eteo-Cretan and Eteo-Cypriot) whose epicenter lies in Western Anatolia and Cyprus, formerly known as the kingdoms of Alashiya, Arzawa and Assuwa. I also have a hunch that by 1400 BCE, Minoan had become a dead language but still used in ritual while, in the everyday world of the commoners, a mix of Greek and Etrusco-Cypriot languages survived on in Crete.

If I take the name *Kʰaputar for granted, I'm reminded of a plural suffix *-r that I see in the Minoan Libation Formula sometimes marking the word *una 'libation' (written syllabically as U-NA-; in the sequence U-NA(-RU)-KA-NA-SI, *una(-r) kana-si '(we) bear libation(s)', compare Etruscan -r [animate plural], un 'libation' and cenu 'brought'). Without this ending, we're tentatively left with a singular word *kʰaputa. Wild imagination may lead one to see similarity between it and the Latin-derived word 'capital' however this leads to another interesting mystery: Where does Latin caput 'head, summit' come from?

Some etymologists try very hard to make caput a Proto-Indo-European word.^[1] However, it's unclear to me why anyone would be so determined to force the word to be PIE given the meager basis. At most, they're reduced to label it vaguely as a 'regional term' or an 'Italo-Germanic innovation' which only skips over the problem of how the word came to be. However, let's try a new idea. Let's suppose for a moment that this odd word is entirely non-IE and a loan from a theoretical Old Etruscan word *χapuθ (> (?) Late Etruscan *χafθ), lent also at some point to Germanic (hence Old English heafod). At this juncture, I think many readers here might predict where I'm going with these crazy ideas.

Is it just possible that the word for 'head, summit' in both Minoan and Etrusco-Cypriot during the mid 2nd millenium BCE was originally *kʰaputa? From there, *Kʰaputar 'The Summits'(?) would become the word for the entire Minoan region, perhaps in connection to the Horns of Consecration, a very sacred and prominent symbol undoubtedly related to the Egyptian aker symbol representing the sun both emerging from and setting into the two horizons.

Now naturally, these are so far just a delightful multiplication of hypotheses and fun wordgames to toy with while passing the toke around. I recognize that it remains incumbent on me to prove them with tangible evidence if I'm ever to insist this hypothesis to others. So, as always, I'll just have to see where these ideas take me.


NOTES
^[1] Mallory/Adams, Encyclopedia of Indo-European culture (1997), pp.260-261: kaput (see link).

Libyan's interpreter at UN collapses from sheer monotony

Okeydokey... this vaguely linguistics-related news story is kind of amusing and delightfully odd:

AFP: 'I can't take it!' cried Kadhafi interpreter: report

Tsk, tsk.

Let's not confuse the topic of history with a history of history, yes?

I've been on a "Proto-Indo-European kick" for some time now and so maybe it's time to welcome some new subjects. Variety is the spice of life, they say. My mind has been returning to the issue of the Minoans and the stinking mysteries that authors almost seem to encourage by drowning the reader in what-ifs and no conclusive framework. I think I now have another opinionative plaint against the style of many history books.

Rodney Castleden in The Knossos labyrinth: A new view of the 'Palace of Minos' at Knossos (1990), for example, explains in detail about the archaeology of Minoan civilization. I have no issue with the author, nor with the book per se. All this information appears to be excellent and informative in a way. Yet it's also irritating. I think I find myself being short with it because I have a subconscious objection against reading pages and pages of text that mix known facts about the Minoans, which I appreciate and love since it's my passion afterall, with the history of history itself, or rather, the minutiae of modern academic politics that I abhor with a vehement passion. For me, what historians have theorized through the eons, particularly the theories that no longer hold true, is unimportant and quite frankly egotistical because it reeks of a petty glorification of individuals' personal careers. I don't care about academics and their own histories to be brutally honest. Just the history of the Minoans, thanks.

So at the end of such texts where an author is too timid to offer a structure of his own, the reader is left confused with a myriad of multiple, mutually conflicting ideas from various historians and archaeologists, past and present. But... that's not what I read these books for. I want a structure. My detail-famished mind needs to know step-by-step what happened, or at least what likely happened given current knowledge, regarding the Minoan people themselves.

Oh well, I guess it just goes to show that one needs to be an active participant in reading and to develop one's own testable models to make sure that one understands what was read. Relying on authors for coherent models is in the end partly my fault as a reader. Time to get more pro-active!

Glen Gordon's ancient, twisty path??

I just found a National Post article called Ugly Duckling: Glen Gordon's ancient, twisty path. Naturally my brittle ego was both shocked and titulated. I just had to take a teensy peek inside and ascertain what dirty, little gossip a leading Canadian newspaper had to say about me. Lol!! I'm all about the guilty pleasures.

But alas, the article concerns a shabby home on a Toronto street called Glen Gordon Road and is a sequel to another header: Ugly Duckling: Get some support for this saggy bottom. Geewiligers! Is this a hint that I need to get back in the gym? How unflattering and rude, I must say. Hahaha.

Minoans, Greeks, the Po Valley and Arretium

Something I googled up the other day makes me start thinking again about the precise extent of Minoan trade. Clark, Prehistoric Europe - The economic basis (1966), page 186, reads:

"Although there is no direct proof that the rich copper deposits of Etruria began to be worked for the Minoan market, it is suggestive that the earliest metal equipment of Italy should show Early Minoan influence, notably in the daggers with mid-rib and rivetted tang found in cemetaries of the Remedello culture of the Po Valley and in the central Italian Rinaldoni culture."

This piece of text reinvigorates my hunch that the very reason why Etruscans, when considered to be the offshoot of peoples with like languages in Western Anatolia and surrounding islands (ie. Eteocretan, Eteocypriot, Minoan), ended up in Western Italy by the first millenium BCE was because trading routes and mining sites were in some way already established in the Po Valley, Western Italy and Sardinia during late Minoan times.

I came across this while reopening the case concerning the etymology of Arretium, a town in NE Etruria, which I can confidently say is unanalysable in the Etruscan language, despite Arretium being purportedly founded by the Etruscans themselves... but this is a slightly separate issue. For that, another quote seems interestingly apt, but in part related to my first quote above, from Voyage to the other world - The legacy of Sutton Hoo (1992), ed. by Kendall/Wells, p.31:

"By the fifth century B.C., true Celtic art was born - called by archaeologists the Early La Tène style after its type-site in Switzerland. The inspiration for this art again seems to have come through northern Italy and the Adriatic. This process is probably due to a shifting of trade routes between temperate Europe and the Mediterranean as trade from Massalia began to decline and the focus of Greek trade shifted to the Po Valley."

In other words, a historical continuity of trade between Northern Italy and Greece via the Adriatic since Minoan times. So this is in part why I'm now pursuing a hunch that Arretium could perhaps be a Greek name in the end, namely from Erythrion, a name built on the word erythros 'red' (< PIE *h₁reudʰ-) which could presumably be inspired by copper mining in the area. Interestingly, Erythrion is attested elsewhere as a personal name. The name would then also presumably pass into Old Etruscan as *Aritium with the -r- dropped to naturalize pronunciation for Etrurian ears and predictably shortened to Aritim as attested.

Ancient Etruscan ointment discovered in Italy

It turns out ancient ointment made of moringa oil has been preserved and discovered near Chiusi. This shows yet more interesting trade with Egypt, this time involving aristocratic cosmetics. Neat! Read more in the link:

Ancient Etruscan ointment discovered in Italy (Jul 10, 2009; Rossella Lorenzi, Discovery News)

Interesting quirks of a PIE subjective-objective model

Before I amble around in endless prehistoric details yet again, I found some faults with a few of the things I said previously. First, I realize that 'to share' is the likelier original semantics of *deh₃- to account for the seemingly contradictory 'give' and 'take' meanings in the daughter languages (and so my original suggestion of 'trade' is, well, relatively crap). Whatever the case, it doesn't appear to have strictly meant 'to give' in the beginning. Second, I discovered my misunderstanding of ablaut in the mediopassive such that a Narten stem should display an accented, shortened vowel in the middle while a non-Narten stem should display zero-grade^[1] (therefore, *h₁és-h₂or 'I sit' and *gʰu-h₂ór 'I spill', not **h₁ḗs-h₂or and **gʰeu-h₂ór). Whoops! Well that's the whole point of my blog... to share, to err and to learn. Hopefully on the side of learning more than on the side of erring, of course.

Anyways, as per my previous model, there are interesting quirks that seem to automatically surface when I personally take on the goal to finally account for both the mi-class/hi-class contrast in Anatolian with the durative-aorist-perfect system of Core IE dialects. I'd like to remind new readers here that my exploration is not a frivolous one since Indoeuropeanists themselves still to this day admit to their lack of a coherent model to explain everything^[2]. While my previous contrasting of eventive and non-eventive explains the use of the two sets of personal endings of the verb to an extent, there are still a few cases such as the would-be sigmatic aorist (eg. *bʰērst 'he carried (at one time)') where a term like non-eventive is totally inadequate, if not wrong altogether, but it was the closest concept I could find to explain the pattern.

So here's the slightly revised model that might better reflect the functional symmetries between the objective and subjective using a progressive/non-progressive contrast (ie. progressive here is meant to signify an action that's specifically realis present continuous while non-progressive would cover everything else). My model remains effectively the same as before, just with more descriptive labels hopefully:

	objective	subjective
progressive	*-mi *-si *-ti	*-h₂ór *-th₂ór *-ór
non-progressive	*-m *-s *-t	*-h₂e *-th₂e *-e

The mediopassive-perfective connection

It appears that there's a symmetry between objective verbs marked with *-i (> durative, mi-class) and those unmarked by it (> aorist, past) on the one hand, and subjective verbs marked with *-r (> middle) and those without (> hi-class, perfect-stative) on the other. This implies that just as the presence or lack of *-i in the objective is determined by its progressiveness, so too in the subjective between what would later be described as middle and hi-class in Anatolian morphology^[3]. Thus for example objective *stḗu-mi 'I invoke' and *dí-deh₃-mi 'I am sharing' (progressive) would oppose *stḗu-s-m̥ 'I have invoked' and *déh₃-m̥ 'I share' (non-progressive) just as subjective *h₁és-h₂or 'I am seating myself' (progressive) would oppose *h₁és-h₂e 'I am seated' (non-progressive). In other words, rather than reconstructing a separate voice, we might simply place the middle in a different aspectual category under the subjective distinct from the *r-less "proto-perfect" and further treat it as the marked "presentive" form of the subjective.

Changes to the realis mood

Another interesting aspect of this model is the prediction that while the so-called "secondary" endings (ie. the objective non-progressive endings in this model: *-m, *-s, *-t) came to dominate the function of irrealis in the Core IE dialects, the original grammar must have allowed for both objective and subjective non-progressive endings. So presumably if *h₁i-yéh₁-n̥t 'they should go' is the optative of an objective verb like *h₁y-énti 'they go', then theoretically *ḱéi-ih₁-th₂e 'you should lie down' (rather than later *ḱéi-ih₁-s) would have originally been the optative of *ḱéi-th₂or 'you lie down'.

Further, the emergence of *-o-h₂(e) in the 1ps subjunctive (and later in the derivative thematic present of Core IE dialects) could add to this account since the specific replacement of original objective subjunctive *-o-m with a subjective marker would be due to the fact that the 1ps is the only person where speaker and subject are the same and the only person subsequently of which one's own potential actions are spoken with the utmost authority. Over time, this suppletiveness in the subjunctive would begin to merge into a single paradigm hinting at a nuance between potentiality stated with authority (marked in the subjective in the 1ps only) against potentiality stated as hearsay (marked in the objective for all other persons). It's a mind-boggling issue and I'll have to think more on this.


NOTES
^[1] As per the middle forms suggested in Sihler, New comparative grammar of Greek and Latin (1995), p.133 (see link) and in Indo-European perspectives - Studies in honour of Anna Morpurgo (2004), p.506 (see link).
^[2] Fortson, Indo-European language and culture (2009), p.103 (see link): "The close similarity of the perfect endings to the middle endings has generated much research and controversy; precisely what the connection between them is remains unclear."
^[3] Jay Jasanoff says something similar, that forms with *-r functioning as present indicatives once opposed the "perfect" without this ending, in Hittite and the Indo-European verb (2005), p.58 (see link).

More on a PIE subjective-objective model

Maybe I'm beating a dead horse, but I don't care. The more I ponder on an underlying tenseless, two-dimensional system for PIE verbs, the more I see many advantages to it as well as some unexpected quirks. My previous description of my model was very tedious, I have to admit, but I think I can now sum my model up more concisely as a conjugation model that takes into account the definiteness of object (subjective vs. objective) and the definiteness of event (eventive vs. non-eventive). These two dimensions together give us four basic categories from which I think we can more easily derive both the system of early Anatolian-Tocharian dialects and that of the remainder, the "Core IE" dialects. I'm going to talk more here about the theoretical evolution of the conjugation system in both these groups from the following system.

	objective	subjective
eventive	*-mi *-si *-ti	*-h₂ór *-th₂ór *-ór
non-eventive	*-m *-s *-t	*-h₂e *-th₂e *-e

The above model sides with simplicity to show an overall pattern. Surely experientials (ie. later sigmatic aorists) would belong in the "objective eventive" despite being unmarked by continuous *-i. However, despite these minor details, a quadripartite scheme with the above characteristic endings as above seems to hold true in general and may help to understand how things evolved later on.


Evolution into Anatolian-Tocharian group

Essentially, it looks like the following general semantic changes help to alter the above system in these dialects:

‣ eventive → presentive
‣ non-eventive → preterite
‣ subjective eventive → middle
‣ subjective non-eventive → hi-class present & preterite
‣ experiential *CēC-s-t replaces *CoC-e in 3ps hi-class past

	mi-class	hi-class	middle
present	*-mi *-si *-ti	*-h₂ei *-th₂ei *-ei	*-h₂ór *-th₂ór *-ór
past	*-m *-s *-t	*-h₂e *-th₂e *-st

Evolution into Core IE group

A different path is indicated for this group:

‣ objective eventive → durative present
‣ objective non-eventive → both preterite & aorist
‣ subjective non-eventive → perfect-stative
‣ subjective eventive → a seperate middle voice
‣ middle present *-r optionally exchanged with *-i from active voice
‣ middle past deletes final consonant in endings by analogy with pattern in active voice

	durative	aorist	perfect-stative	middle
present	*-mi *-si *-ti	*-m *-s *-t	*-h₂e *-th₂e *-e	*-h₂ór/*-h₂ói *-th₂ór/*-th₂ói *-ór/*-ói
past	*-m *-s *-t			*-h₂ó *-th₂ó *-ó

New thought: A 2D matrix of eventive/non-eventive and subjective/objective

I've been remodeling the PIE verb system for my own personal kicks to try and unite a number of various ideas (ie. Jasanoff's theories, the durative-aorist-perfect model, active-stative, and subjective-objective) into a single coherent model that explains everything much clearer than what I'm finding in journals and books. Forgive me if I'm reinventing the wheel but as far as I know this wheel hasn't been invented yet.

My cursed snag to my subjective-objective theory remains that subjectives are expected to yield atelic verbs (ie. verbs without completed goal) or imperfectives (ie. continuous actions or states)^[1]. Even though the middle clearly should be placed under a subjective category in my hypothetical remodeling, and even though the subjective on which the middle is built acts in many respects like a former subjective (for example, its curious habit of accomodating verbs of state), subjective can't turn into a perfective in any direct way according to known structural linguistics! We oddly expect the total reverse: subjectives yielding imperfectives and objectives yielding perfectives. This is why I had a migraine a couple of days ago, frenetically diagramming to make this irritating paradox finally unfold into clarity. I think I've come to an interesting idea that takes a page from the Ancient Egyptian verbal system: eventive versus non-eventive.

An eventive verb, as the name implies, refers to a specific event in time while the non-eventive by contrast focuses more on the action or state in a more generalized context. It's the difference between "I ate the baby (last night)" (eventive) and "I eat babies (in general)" (non-eventive). By drawing a two-dimensional grid between eventive & non-eventive on the one hand and subjective & objective on the other, we end up with four main categories in which to place the earliest verbs of Common Proto-IE. This is perhaps not a good synchonic model for PIE but it seems to be explanatory as a seed for its eventual evolution in many of the Core IE dialects. Observe:

Objective non-eventive	*h₁és-m̥ 'I am' *déh₃-m̥ 'I share'[2] *bʰḗr-m̥ 'I carry (often)'
Objective eventive	*bʰḗr-mi 'I am carrying' *bʰḗr-s-m̥ 'I have carried (once)'
Subjective non-eventive	*wóid-h₂e 'I know' *ḱónk-e 'it is hung' *stestóh₂-e 'it is stood' *gʰónh₁-h₂e 'I am born'
Subjective eventive	*h₁és-h₂or 'I sit' *gʰu-h₂ór 'I spill'

Objective non-eventive

This category is the origin of root aorists and imperfective past in Core IE while becoming the mi-class preterite in Anatolian. This is because the pan-PIE choice to make the continuous aspect a present tense marker made non-continuous verbs automatically a past by contrast. This change also hints at why past tense came to be less marked than the present, curiously opposing normal language tendencies (ie. the shift occurred rapidly as PIE broke up). Telic verbs (ie. Narten presents with *ē-vocalism) were free to be marked in the continuous, suggesting an impending goal, or to remain unmarked to convey realized ones. The former leads to a durative-turned-present and the latter leads to a momentaneous-turned-past.

Objective eventive

This category was largely characterized by telic verbs marked in the continuous aspect (later Narten presents). However, the antecedent of sigmatic aorists (ie. those verbs marked in *-s- with lengthened root vowel) which originally expressed a past experience were by definition eventive as well. An experiential form, parallel to Mandarin guo (过), can easily yield explicit past tense "sigmatic aorists" in Core IE dialects, while forming special 3ps sigmatic past forms for an originally tenseless hi-class as evidenced by Anatolian and Tocharian.

Subjective non-eventive

This category denoted states where the subject was affected (passives), but also often verbs of emotion and of thought which are clearly subjective, as well as what would later be termed reduplicated perfects. The development of reduplicated perfects with built in punctual meaning directly out of a "stative" requires the brunt of explanation. My hunch here is that reduplication was restricted to verbs like 'stand' signifying states resulting from prior action. An expression like 'it is stood' qualifies as such while 'I have gone' is less focussed on state than on the completed action itself. This latter comparison explains how forms specialized in suggesting resultant states from actions (PIE *stestóh₂e 'it is stood up' → 'it has come to be stood up') can gradually expand to take on resultant actions as well (post-IE *bʰebʰórh₂e 'I have come to be carrying' → 'I have carried') and thus become a "perfect" with punctual aspect now emphasizing active result over mere resulted state. The connection then between subjective and perfect is indirect and subtle.

It should also be noted that only part of this single category yields perfects while the rest of the subjective umbrella is the mother of middles and inherently non-punctual states. So the subjective simply doesn't become perfective wholesale and there's no need for me to worry about the crosslinguistic tendency I mentioned above which refers more to developments as a whole, not in part. Phew!

Also, forms like *wóidh₂e 'I know' which never ever show reduplication in the later perfect hint at their original meaning and usage: 'I know' (stative) → 'I have come to know' (inchoative) → 'I have known/seen' (perfective past). Since 'to know' is not an action and since reduplication expresses a resultant state from an *action* as outlined above, naturally there can be no reduplicated forms possible for these stative verbs.

Subjective eventive

This category is composed of middles and are marked as such with special endings in *-r. They are actions accomplished through a medium ("mediopassive"), through one's own effort ("sitting"), or involuntarily like "spilling" or "sneezing". This category appears to be the least molested by change in the various branches, so I think I'll stop typing now.

There, now we'll see if this post stands the test of time...


NOTES
^[1] Gülzow, The Acquisition of Intensifiers Emphatic Reflexives in English and German Child Language (2006), Studies on language acquisition, v.22, p.45, table 6 (see link).
^[2] Perhaps this verb is better placed in the category of subjective non-eventive instead. Note Hewson/Bubeník, From Case to Adposition: The Development of Configurational Syntax in Indo-European Languages (2006), p.100 (see link): "In this context we should remind ourselves of the archaic transactional meaning of *dō- (< *deh₃-) as shown by Hittite dahhi 'I take' (< *deh₃-h₂e) vs. general IE 'I give'." Jasanoff shows a slightly different interpretation in The role of o-grade in Hittite and Tocharian verb inflection (1992), published in Reconstructing Languages and Cultures, p.140 (see link): "Hittite has a 3 sg. dāi, which can be referred to an underlying *doh₃-e and taken as the continuant of a pre-Proto-Indo-European 'middle' with the meaning 'gives/gave to oneself'."

UPDATES
(12 Sep 2009)

Just realized that a better translation of what must have been a bidirectional sense underlying *deh₃- is not 'to trade' but 'to share'. Thus, a semantic shift from 'to share (my own property)' → 'to give' (as in Core IE dialects) and 'to share (another's property)' → 'to take' is clearer. It also works well with what grammatically hints at a non-eventive and atelic verb, non-eventive because it avoids the continuous marker and atelic because it shows no Narten-grade which would explicitly mark it as telic.

Also corrected the forms *h₁ḗs-h₂or and *gʰeu-h₂ór to *h₁és-h₂or and *gʰu-h₂ór respectively, as per the middle forms suggested in Sihler, New comparative grammar of Greek and Latin (1995), p.133 (see link) and in Indo-European perspectives - Studies in honour of Anna Morpurgo (2004), p.506 (see link).

(05 Sep 2009)

Added footnote #2 concerning the changing semantics of *deh₃-.

(09 Sep 2009)

Added footnote #1 concerning theories on the typical evolution of transitive verbs.