The basis of coming and going
« previous post | next post »
The protean particle zhī 之 (3 strokes, classifier / radical 丶) has more grammatical functions than you can shake a stick at, e.g.:
(literary) genitive or attributive marker
indicates that the previous word has possession of the next one
indicates that the previous word modifies the next one
particle indicating that the preceding element is specialized or qualified by the next
(archaic) particle infixed in a subject-predicate construct acting as a nominalizer or indicating a subordinate clause
(literary) the third-person pronoun: him, her, it, them, when it appears in a non-subject position in the sentence
—
(adapted from Wiktionary, with illustrative quotations for each type)
What I find especially remarkable about zhī 之 is that it began as a verb which is usually rendered as "go". The oracle bone glyph (ca. 1200 BC) used to write 之 is this 
, which depicts a foot on a base line ("beginning place").
What made me think of 之 a few days ago is the "-basis" of "diabasis", "anabasis", and "katabasis" about which we puzzled in this post. That may seem like quite a stretch, but there are some striking similarities between the two that caught my attention, though, I caution, by no means do I wish to imply there is any sort of genetic relationship between 之 and "-basis", only a curious correspondence between them that may belie a fundamental / basic trait of human cognitive patterning. We will proceed slowly, taking this one step at a time, with the understanding that we are walking on two legs, which are — for the sake of this argument — the Indo-European and the Sinitic.
When we think of "basis", what comes to mind is something that serves as a bottom or foundation. Without rigorous historical linguistic analysis, we would not readily think that the Greek root whence the word comes means "a going, a step; a stand, that whereon one stands", and that derives from the verb bainein βαίνειν "to go, walk, step" which, in turn comes from the PIE root *gwa- "to go, come". Though "go" and "come" both have to do with movement through space, they are very different — indeed opposite — matters, so I would refer to bainein as a contranym. Whether it (and its derivatives) means "come" or "go" in a given utterance is determined by context and / or its prefixes: "anabasis" ("go up"), "katabasis" ("go down"), and "diabasis" ("go across", cf. "dialect" ["speak across"!]).
basis
From Latin basis, from Ancient Greek βάσις (básis), from Proto-Indo-European *gʷémtis, derived from Proto-Indo-European *gʷem- (whence also come). Doublet of base.
*gwā-, also *gwem-, Proto-Indo-European root meaning "to go, come."
It forms all or part of: acrobat; adiabatic; advent; adventitious; adventure; amphisbaena; anabasis; avenue; base (n.) "bottom of anything;" basis; become; circumvent; come; contravene; convene; convenient; convent; conventicle; convention; coven; covenant; diabetes; ecbatic; event; eventual; hyperbaton; hypnobate; intervene; intervenient; intervention; invent; invention; inventory; juggernaut; katabatic; misadventure; parvenu; prevenient; prevent; provenance; provenience; revenant; revenue; souvenir; subvention; supervene; venire; venue; welcome.
It is the [recte The] hypothetical source of/evidence for its existence is provided by: Sanskrit gamati "he goes," Avestan jamaiti "goes," Tocharian kakmu "come," Lithuanian gemu, gimti "to be born," Greek bainein "to go, walk, step," Latin venire "to come," Old English cuman "come, approach," German kommen, Gothic qiman.
Now, back to the other leg (rather I should say "foot") of our "come-go" diversion, the Sinitic one.
Cf. Burmese စ (ca., “to start; to begin”) (Schuessler, 2007, p. 613).
Also, as a verb in Classical Chinese / Literary Sinitic, zhī 之 implies "to go, head towards; reach, arrive" (reddit). Graphically, the character used to write Middle Sinitic tsyi, Old Sinitic /*tə/ (Baxter-Sagart), /*tjɯ/ (Zhengzhang) shows a foot with a horizontal plane under it indicating the place whence the foot comes. Cf. IE "basis" for coming and going.
South Coblin:
As to 之 as a verb, my impression of it is that it is a verb of motion whose destination must always be present in the utterance, either literally or by implication through the use of some sort of concomitant grammatical construction, as in phrases such as 其所之* or 子安之**？, etc. As to directionality with regard to the speaker or other relevant point in the utterance, I must admit that I have never investigated that question. One could determine it by collecting and comparing text examples.
*VHM: qí suǒ zhī ("where he / she / it / one is heading")
**VHM: zǐ ān zhī ("where are you headed?")
"Coming" and "going" are difficult for L2 speakers to master in English. I suspect that the same is true for other languages as well. So much depends upon the perspective from which one views the movement.
Selected readings
- "Diabasis" (11/15/24)
- "cactus wawa: the strange tale of a strange character" (11/1/14)
- "Cactus Wawa revisited" (4/24/16)
- "No character for the most frequent morpheme in Taiwanese" (12/10/13)
- "Mixed literary and vernacular grammar" (9/3/16)
[Thanks to South Coblin, Axel Schuessler, and Michael Carr]
David Marjanović said,
November 19, 2024 @ 9:18 am
I'm not saying this is wrong; but if it's right, it's part of a stage quite a bit older than PIE, which didn't tolerate vowel-final roots and appears not to have had what used to be reconstructed as "*a" at all. PIE did, however, have a suspicious trio of *gʷem-, *gʷeh₂- (pronounced with something like [a]!) and possibly *gʷew-, all with similar meanings somewhere around "go, walk" and similar grammatical behavior; likewise *drem-, *dreh₂-, *drew-, all "run", again all with similar grammatical behavior.
Judging from Wiktionary, basis is an abstract noun in *-ti- with its declension straightened out within Greek (nominative *gʷémtis, genitive *gʷmtéys > Pre-Greek nom. *gʷḿtis > gʷátis > básis). I think nom. *gʷéh₂tis, genitive *gʷh₂ətéys would have the same Greek outcome, though (assuming the same leveling of the declension).
Chris Button said,
November 19, 2024 @ 9:22 am
I don't think this is related. Burmese "c" reflects OC ts-/dz-.
I think 之 is related to 止.
Chris Button said,
November 19, 2024 @ 9:49 am
Actually pre-Old Burmese tj- gave "c" စ-, but we'd need evidence for the medial j in the Old Chinese form, which doesn't seem possible in this case.
Victor Mair said,
November 19, 2024 @ 10:20 am
@Chris Button
Burmese "c" reflects OC ts-/dz-.
For zhī 之, Middle Sinitic tsyi, Old Sinitic /*tə/ (Baxter-Sagart), /*tjɯ/ (Zhengzhang)
I think 之 is related to 止.
I've often thought of that too — zhǐ 止 — but semantically it puts a whole different light on the matter. Oracle bone (OB) zhǐ 止 does depict the same "foot" as OB zhī 之, but it lacks the horizontal plane of the latter, which means "go", whereas the former, zhǐ 止, means "stop". Phonologically there are also problems:
zhī 之 — Middle Sinitic tsyi, Old Sinitic /*tə/ (Baxter-Sagart), /*tjɯ/ (Zhengzhang)
zhǐ 止 — Middle Sinitic tsyiX, Old Sinitic /*təʔ/ (Baxter–Sagart), /*kjɯʔ/ (Zhengzhang)
Coby said,
November 19, 2024 @ 11:22 am
Re "Coming" and "going" are difficult for L2 speakers to master in English.
I am not sure that this true for L2 speakers in general, but it certainly is for native Spanish-speakers. Spanish venir can only be directed toward the speaker, so that a sentence like "I will come to your house" would have to be iré (not vendré) a tu casa. On the other hand, Catalan allows venir in such a case (vindré a casa teva), and that creates difficulties for native speakers of one language for whom the other is an L2 (but perhaps not for native bilinguals, of whom there are many in Catalonia).
Chris Button said,
November 19, 2024 @ 11:28 am
@ Victor Mair
In my "Pulleyblankian" reconstruction, the rhyme is -əɣ. You cannot have Old Chinese -jəɣ (which no doubt existed in pre-Old Chinese) since that would be equivalent to Old Chinese -əj. For the same reason, -jək and -jəŋ don't exist in OC either since they are equivalent to -əc and -əɲ.
Incidentally, the spreading of the palatalization to the velar coda (i.e., -ɣ/-k/-ŋ becoming -j/-c/-ɲ) accounts for why the "front vowel hypothesis" (and separately the "rounded vowel hypothesis"), which is championed by Baxter-Sagart and Zhengzhang, fails to disrupt the traditional rhyme categories for rhymes with velar codas. But that is another matter.
In any case, if the Burmese form does not go back to ts-, then it needs to go back to tj-. But that is not possible on the basis of the Chinese evidence.
Separately, I suspect 之 might not have gone back to t- anyway …
I reconstruct 止 as ᵏlǝɣɁ, which supports the velar in Zhengzhang's form. I suspect that 之 may therefore reflect ᵏlǝɣ rather than tǝɣ (it is phonetic in 寺 EMC zɨʰ OC lǝɣ for example, which can't have had a t- onset). But I need to think this through a little more.
For the semantics, compare the possible relationship between 之 and 止 with the relationship between "hither" and "here" in English. The further etymological relationship with "he" and "it" then also compares nicely with the third-person pronominal use of 之.