- Dec 11, 2013View Source
---In email@example.com, <grzegorj2000@...> wrote:
>> Let me start from the 1st Shift / Grimm's "Law" (rule inOf course, but this *toccare may be a Germanic loanwods in turn. The
>> fact). It is often treated as exceptionless but in real we
>> should have doubts. For example, the English "touch" and
>> "take" look like cognate of the Latin tangere with *t-
> English <touch> is a borrowing of OFr <tochier>, from VLat
> *toccare, so it’s irrelevant.
geminate -cc- seems to make it probable.
DGK: I believe so. The VL verb could have been borrowed from a Gmc. denominal or intensive verb, or formed directly from a Gmc. noun *tokka-, with the *-o- resulting from /a/-umlaut of earlier *tukka-. This is phonetically compatible by Kluge's Law with PIE *duk-nó- from *deuk- 'to lead, draw, pull' (IEW 220-1). However, since the meanings of obvious Gmc. derivatives of *deuk- do not stray far from the basic sense, this explanation of *tukka- is semantically difficult. Pulling and striking are entirely different operations.
A clue is provided by one of the principal meanings of Spanish _tocar_, 'to play (a musical instrument)'. This includes stringed instruments such as the guitar and the harp. The harp is a specifically Germanic instrument, and in English one speaks of plucking the harp. Plucking is a form of intensive pulling. One plucks feathers from a bird by sharply pulling them out.
One may thus suppose that a Gmc. phrase such as *tokko:naN harpo:nuN 'to pluck the harp' found its way into VL as *toccare harpam. (In fact the loanword _harpa_ was used by Martianus Capella ca. 500.) The VL phrase was first understood as 'to make the harp sound', and this meaning of *toccare was extended to other musical instruments and bells sounded for various purposes. One sounds a bell by striking it, but not hard enough to damage it. The resulting semantic development for *toccare, 'strike without damaging' > 'strike lightly, stroke' > 'touch' is straightforward.
All hypothesis like those of "echoic" origin of this word (and many others)
seem to be less probable (the fact that authors of etymological dictionaries
copy them one after another - is no proof yet). If we took under
consideration that the human language must have developed from nothing many
thousands of years ago, all words must have been echoic to more or less
degree originally. So, such a hypothesis cannot be treated as the zeroeth
one, and must be proved or positively argumented just like others. The
echoic / onomatopoeic etymology means only that the etymologist, who
investigates a given word, actually resigns, surrenders.
DGK: In general I agree. The gold standard of onomatopoeia is 'meow'. Phonetically similar words are so widely distributed in the world's languages that neither common inheritance nor recent borrowing provides a plausible explanation. Multiple independent formation after the sound is the only reasonable answer for 'meow'.
However, as we move away from names of specific sounds produced by human or animal vocal organs, the plausibility of onomatopoeia drops off quickly. A word like English _bang_ is not onomatopoeic, but merely suggestive. Naive native speakers have no way of distinguishing the two types, and some etymologists fall into the same trap. Indeed we must deal with the difficult area of phonesthemics, in which the suggestive power of a sound-segment encourages retention of words containing it over others. In Romance, *beccare, *toccare, and a few other verbs with *-cc- denote various types of striking. In my view at least, these verbs have etymologies like ordinary verbs, but the presence of *-cc-, which became popularly suggestive of striking (perhaps first with *beccare), encouraged not only the retention of striking verbs in *-cc-, but also the semantic drift of verbs in *-cc- from non-striking to striking (as with *toccare if my etymology is correct). The result gives etymological tyros (and, sadly, some veterans) the IMPRESSION of onomatopoeic formation.
Anyway, it is also the semantic proximity which can vote for relation of
"take" and "touch". Below I am going to give another possible etymology
(other than "echoic").
DGK: Here I must disagree. By itself, semantic proximity (which is disputable between 'take' and 'touch' anyway) has no vote in deciding whether words with a vague phonetic similarity have a common origin. We have all heard of "false friends".
> Yes, of course (except that I can't accept take = tango &c. becauseI agree, but there is proximity here, and that is why I have given both
> phonological irregularity must prevail on semantic proximity,
> otherwise we could never demonstrate that a given hypothesis isLatin tangere "to touch" comes from IE *tH2g- (de Vaan, Etym. Dict. of
> false) [...]
Latin..., p. 606). But its undoubtful (I hope) cognates may have the meaning
"take". This is the proof that there is semantic proximity between "touch
They are especially Greek tetagō'n "having seized" (does it mean "having
touched" or "having taken" rather?) and Latin tagāx "thievish" (does a thief
touch or take his loot rather?).
Another kind of a proof is that Gothic tekan means "to touch" (taítōk "I
touched") while Icelandic taka means "take" and tók = took (the English verb
is of Nordic origin). So, the Gmc. *tekana-/*takana- yielded both "touch"
and "take" meanings. We cannot separate them from one another. The meaning
"touch" is probably original, as "*nema-" was used for "take", also in OE.
DGK: I addressed these words in message #71332 (23 Sep 2013), and I will recapitulate here, incorporating my recent correction about Dybo's Law bleeding Austin's, which affects 'take'.
1. Greek _tetagó:n_ 'having seized' (by the ankle, so neither 'touched' nor 'taken' is appropriate) requires PIE *teh2g- (or *teh2g^-, *teh4g-, *teh4g^-). Kortlandt (NOWELE 36:59-65, 2000) attempted to get ON _taka_ and Go. _tekan_ from the same root, but his derivation requires multiple ad-hoc steps and a glottalic treatment which mocks the concept of a phoneme; it cannot be taken seriously.
2. Ringe (From PIE to PGmc 80) robotically referred Go. _tekan_ and ON _taka_ to a "post-PIE" *deh1g- ~ *dh1g- 'touch'. As is typical of this author, no explanation is given why the root must be "post-PIE" and not simply PIE, or what if anything it continues from PIE, or what its source might be if not PIE. Now, most scholars take the Go. simplex _tekan_ as primary, but I note that forms of the prefixed verb _attekan_ are thrice as numerous as those of _tekan_ in extant remnants of the Gothic Bible. Also in message #71332 I provided evidence that Kluge's Law preceded Grimm's, so that the geminated tenues produced by KL were unaffected by GL, and the same is to be expected for geminated tenues resulting from sandhi. Thus it is highly plausible that Gothic inherited the PREFIXED stem *atte:k-, continuing pre-Grimm's Law *atte:g-, and this reflecting *ad- (PIE *h2ed-) plus *te:g- (PIE *teh1g-) 'to touch' with trivial sandhi. Go. _attekan_ in this view is cognate with its semantic match, Lat. _tango:_ (generalized zero-grade nasal pres. *th1-n-g-), _tetigi:_ (gen. zero-grade perf. *te-th1g-). Go. _attaitok_ continues the gen. /o/-grade perf. *te-tóh1g- with analogical -t- replacing the *-d- expected by Verner's Law (as in _saiso_ 'sowed' for *sezo:, ON _sera_) after reanalysis of the stem as *at-te:k-. This reanalysis led to extraction of the Go. simplex forms. OE _þaccian_ and Old Dutch _thak(k)olo:n_ 'to touch lightly, pat, stroke' are probably based on a Gmc. fem. abstract *þakko: 'touch', PIE *th1g-néh2. But the prefixed verb would have produced its own Gmc. abstract *attakko:, with reanalysis yielding *takko: 'touch'. This alternative noun is probably the base of MLG _tacken_ 'to touch'.
3. ON _taka_ in my opinion is best explained through the /w/-extension of PIE *deh3-, surface-true *doh3-, originally 'to take' (as in Hittite _da-ah-hi_ 'I take'), not 'to give' as generally elsewhere in IE, where the semantic shift 'take' > 'take (for someone else)' > 'give (to someone else)' occurred. Various forms of the root with a /w/-extension are found in Indo-Iranian, Greek, Italic, and Balto-Slavic (IEW 225-6), although no simple thematic present is attested.
I theorize that at some stage in the history of PGmc, a thematic present representing PIE *doh3-w- was formed by analogy with some non-present forms, but its usage became restricted to the first person singular in ritual contexts: 'I (solemnly) take', (formally) PIE *dóh3woh2. Due to the accent, Dybo's Law did not delete the laryngeal, and *h3w was fortited yielding Early PGmc *tókWo: 'I take'. The labial component was regularly lost before */o(:)/ giving *tóko:, later *táko: when */o/ shifted to */a/. Then the verb was generalized beyond the 1sg. pres. ind., and due to its /a/-vocalism was assigned to Class VI.