A Big Garden from a Little Root.

I was rereading one of my favorite Pasternak poems, Зеркало (Mirror) (that webpage has the original Russian alongside an abridged translation by Peter France and Jon Stallworthy), when I stuck on an unusual word I had hurried over before in my attempt to make sense of the whole thing: саднят [sádnyat]. I had scribbled ‘smart, burn’ above it in my copy, which was all I really needed in context, but what kind of verb was it, and where was it from? Before proceeding with those issues, though, I’ll quote the relevant section from Jean Marie Schultz, “Pasternak’s ‘Zerkalo’” (Russian Literature XIII [1983]: 81-100), as a sample of how much work it is to figure out what’s going on in his early poetry:

Here the verb “sadnit'” compacts two distinct sensations, one tactile and one olfactory. First, with its meaning of “to smart” or “to burn”, “sadnit'” indicates the feeling that an abrasion might produce; thus, the verb conveys the sensation the trees (if personified) might be expected to have as their sap flows out over the broken limbs. Second, that the trees burn the air with their sap relates possibly to the very pungent odor that pine resin from newly broken limbs has as it fills the air, particularly after a rain (VIII:iii) when all smells are intensified.⁴

fn 4: Sap is a tree’s natural antiseptic, and the burning sensation produced by the application of an antiseptic to a wound is well known. However, it must be remembered that this is a humanly perceived feeling so that we have here, as throughout the poem, the human experience imposed upon a seemingly impersonal description. Furthermore, the allusion to the “medicinal” function of sap prefigures the medicine, “lekarstvo”, introduced in the next stanza. Likewise, the underlying evocation of the sap’s odor here also works toward the development of a sub-motif revolving around scent (V:iii).

OK, so what’s the story with the verb? Well, Wiktionary provides help with its usage;
it occurs in collocations like но́гу са́днит [nógu sádnit] ‘my leg is sore’ and на се́рдце са́днит [na sérdce sádnit] ‘my heart hurts.’ But what’s really interesting is the etymology, given in Russian Wiktionary: it’s derived from Old Russian садьно ‘wound,’ which in turn goes back to Proto-Slavic *saditi, from Proto-Indo-European *sodéyeti, causative of *sed- ‘to sit.’ The English verb sit is from that root, but cast your eyes down that page and see how much else is! The thematic root present *séd-e-ti gives Proto-Celtic *sedeti, which with a couple of prefixes gives us Welsh eistedd and hence eisteddfod; the -ye- present *sédyeti gives Greek ἕζομαι (as well as sit); *séd-os ~ *séd-es gives Welsh hedd ‘peace’; *sod-ó- gives Proto-Slavic *xodъ and Greek ὁδός; *sōd-o- gives Proto-Germanic *sōtą ‘soot’ (“reflecting the nature of soot as accumulated particles that sit on surfaces”); *sōd-u-s gives Proto-Slavic *sadъ ‘grove; garden’ (hence Russian сад, which also features in the poem); *sod-yo-m ‘seat’ gives Old Irish suide and Latin solium; *sed-lo- ‘seat’ gives Proto-Germanic *setlaz; *ni-sd-ós ‘nest’ (with zero grade) gives Proto-Balto-Slavic *nísda (leading to Russian гнездо) and Proto-Germanic *nestą (leading to nest)… well, I could spend all day lost in the web of connections. In Russian alone, the root is the ultimate source of посадить/сажать ‘to seat, plant,’ сиделка ‘(sick-)nurse,’ седло ‘saddle,’ село ‘village,’ сажа ‘soot,’ досадный ‘annoying,’ наседка ‘brood-hen,’ население ‘population,’ осадки ‘precipitation,’ осада ‘siege,’ председатель ‘chairman, president,’ расселина ‘crevice, fissure,’ сосед ‘neighbor,’ ссадина ‘scratch,’ усадьба ‘farmstead, country estate,’ and всадник ‘rider, horseman’ (as well as many others). This is the kind of thing that made me want to be a historical linguist.

Comments

  1. Alas, I don’t see my favorite, Irish sídh ‘fairy mound’ (as in banshee) listed or linked to among the derivatives!

    (I was curious about the genesis of this etymology. See p. 245 of Patrick Sims-Williams ‘The evidence for vernacular Irish literary influence on early mediaeval Welsh literature’, in Dorothy Whitelock et al., eds (2011) Ireland in Early Medieval Europe : Studies in Memory of Kathleen Hughes, visible to LH readers on Google Books here, I hope. It seems this root etymology for Old Irish síd was first proposed by John Morris-Jones in Taliesin (1918), volume 28 of Y Cymmrodor, on p. 238, n. 1, available here.)

  2. David Eddyshaw says

    Hmph. Can’t say I’m sold on Welsh hedd “peace” being related to sídh “fairy mound.” And Morris-Jones had some weird notions about etymologies sometimes.*

    However, the GPC entry for hedd actually cites an Old Irish síd, síth “peace”, so I suppose it works formally, at least, even if the semantics seem pretty dubious. “The strange combination of meanings ‘tumulus’ and ‘peace’ must have its roots in Celtic mythology”, eh? Not buying that without some evidence of an actual myth. You could explain away pretty much any kind of semantic mismatch with that kind of handwaving.

    But GPC does go on to say of Old Irish síd, síth “peace” that it might be the same word as síd, síth “gorsedd y Tylwyth Teg.”

    * This Siδi can hardly be the “Welsh equivalent” of Irish síd, as M-J blithely asserts, if by that he means “cognate”: that just doesn’t work. A loanword from Irish seems not impossible, though. There are quite a few of those in Welsh.

  3. David Eddyshaw says

    I like the idea that Welsh heddlu “police” might etymologically be “fairy host”, though.

    I don’t understand how Old Irish síd (in either sense) can come from the PIE *sed- root. Welsh hedd is straightforward: it goes back to *sedos without any trouble, and “sit” to “peace” is semantically doable, via “stasis” or the like. But how do you get the Old Irish /i:/?

    I suppose Latin sedes shows the same ablaut, though.

  4. David Eddyshaw says

    I see that the long vowel in the first syllable of sedes is itself somewhat mysterious:

    https://en.wiktionary.org/wiki/sedes#Latin

  5. But how do you get the Old Irish /i:/?

    From the “long e grade” of the root, according to Morris-Jones; click on Xerîb’s last link.

  6. Here’s a question, just in case someone more knowledgeable in Old Russian will turn up. Where саднить got it’s suffix -н- ? Tsyganenko thinks that there was a noun садьно (wound) which later got verbalized, but was itself a nominalization of садить like болеть->больно. But that doesn’t make sense. It would mean that a wound just sits there. There might be a generalization from a specific wound which comes from sitting on a horse. Thus the meaning would be “a wound acquired from sitting”. Presumably there is a Ukrainian word садно with this meaning. But there is also a word ссадина (abrasion) with apparent etymology с-садина (something that was sitting is removed) is it a parallel development?

    These roots are not only produce a large garden, but are also very tangled….

  7. @D.O.: sadit’ is from the PIE causative *so(:)d-eye- “to set”, not from the intransitive “sit”. So the idea would rather be that a wound is something that is put by someone else on the body. For a similar idea, compare colloquial German jemandem eine setzen “to slap someone”, literally “to set one on someone”.
    @DE: the common idea behind the fairy hill and the peace meanings could be that of “settlement”.

  8. Hans, makes sense (though a little bit strange), thank you!

  9. PlasticPaddy says

    @hans, de
    I don’t know about Wales, but in Ireland “settlement” would be unusual for sídh, because the living are advised to stay well away. Seat (of some Godawful Power you never want to meet) would work. Sort of like a badger sett where the badger can turn you in to a bird /small animal or make you die in agony.

  10. @PP: I was thinking about a settlement of fae.

  11. PlasticPaddy says

    @hans
    I don’t really know this all that well, but I think the fairies are not supposed to live in settlements in this world, but in the Otherworld, which can be accessed through holy places like the mounds. The fairy mounds are uninhabited; things emerge from the Otherworld when the mound is disturbed by a human or when the fairies become restless for another reason.

  12. David Eddyshaw says

    From the “long e grade” of the root, according to Morris-Jones; click on Xerîb’s last link.

    Yeah, I saw that: but that is a peculiar formation (see the Wiktionary entry on Latin sedes.) Why does it have the lengthened grade?

  13. David Eddyshaw says

    I also agree with PP that getting to “fairy mound” from “sit” is pretty difficult semantically. And the Irish word never seems to be followed by a word in the genitive meaning “fairies”, or to be used for any other kind of “settlement”, so the equation with the -sedo-/-sido- element of Gaulish names looks difficult. On the contrary, the word by itself without modifiers means, specifically, “fairy mound.” Or even just “fairy” collectively.

    This is a separate issue from the long-vowel thing, though, as that also turns up in the homophone “peace”, which surely must be connected with Welsh hedd somehow. But the Welsh word doesn’t have the long vowel …

  14. Neither de Vaan nor Ernout/Meillet see any difficulty with the lengthened grade, and neither does Sihler’s Comparative Grammar. Otoh, the Wiktionary article has problems: The Gen. Plur. is wrong (it should be sedum not sedium), and so is the IPA for the pronunciation.

  15. You missed one descendant: пизда, originally, “that upon which one sits”! See
    https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/p%C3%ADsdeh%E2%82%82.

  16. That’s a good one!

  17. @PP: My point doesn’t depend on síd meaning exactly “settlement”; “settlement” was rather to illustrate that a word meaning a physical location where someone or something resides, can also mean an agreement and is not so far from “peace” as it may seem.
    @DE: IE languages frequently have words derived with different suffixes and ablaut grades from the same root meaning similar things; a classic example is German Bund, Verbindung, Verband, which all have “union” as one of their meaning.

  18. David Marjanović says

    *sod-ó- gives Proto-Slavic *xodъ

    That’s very difficult to accomplish, as explained at the link.

    “The strange combination of meanings ‘tumulus’ and ‘peace’ must have its roots in Celtic mythology”, eh?

    As in “rest in”? That’s more Christian than “Celtic mythology”, though.

  19. David Eddyshaw says

    @Hans:

    Sure, but the long-vowel grade is a bit different from the PIE *e/o/zero thing which your example illustrates.

    https://en.wikipedia.org/wiki/V%E1%B9%9Bddhi

    What I’m wondering about is just what kind of derivational formation this *se:dos is. The Wiktionary entry implies that it’s a thematic remodelling of an older athematic *se:ds, which doesn’t match up very well with what WP says about the vrddhi grade. If it is a “possessive derivative”, as the WP account would imply, how come?

  20. David Marjanović says

    What I’m wondering about is just what kind of derivational formation this *se:dos is.

    To me it looks like a reduplicated noun, like *kʷeʷklos “wheel” from the verb root *kʷelh₁- “turn”, with dissimilation + compensatory lengthening, but this presupposes that 1) the syllable boundaries were *kʷe.kʷlos but *ses.dos, even though *-st- would almost definitely have given *-.st-, and 2) the allophones [s] and [z] were nonetheless not considered different enough to block dissimilation. And *seːdos still doesn’t give us seːdeːs.

    However, there’s little evidence that vrddhi goes back all the way to PIE (as opposed to being a later reinterpretation of long vowels as part of the ablaut system). The best or only example, “egg”, has also been explained as a reduplicated *h₂oh₂ujóm from *h₂áwis “bird” (*h₂o-h₂wj-ó-m, *h₂éwj-s), though I don’t know what the *o-grade is doing in the reduplication syllable.

  21. The development of *se-sd- => Latin se:d- would be entirely parallel to ni:dus from *ni-sdo-, wouldn’t it?
    @DE: I don’t think we know sufficiently on how derivation worked in PIE and in the early stages of the branches to treat lengthened grades with so much suspicion. We know that both Germanic and Sanscrit took vrddhi and ran with it, but that doesn’t mean that Celtic and Italic couldn’t have indulged a bit as well.

  22. David Marjanović says

    would be entirely parallel to ni:dus

    Oh. Yes, of course! There are other cases, too, IIRC.

  23. David Eddyshaw says

    I did wonder about reduplication as the explanation of Old Irish síd, but unfortunately *sesdos would produce Old Irish *set (with t for /d/), to judge by Old Irish net “nest.”

  24. David Eddyshaw says

    Welsh hud “magic” and Old Norse seiðr “magic” are from the same PIE root

    https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/seyt-

    but I can’t see any way of getting Old Irish síd “fairy mound” out of that one, despite the tempting semantics. There’s a variant síth, but the vowel is wrong for a reflex of either PIE *ey or *oy. There doesn’t seem to be an actual Irish cognate of Welsh hud (I think it would have been *soíth.)

  25. What I’m wondering about is just what kind of derivational formation this *se:dos is.

    Peter Schrijver (1991) The Reflexes of the Proto-Indo-European Laryngeals in Latin, p. 376, would get the lengthened grade *-ē- of the root continued in Latin sēdēs and Old Irish síd from a root noun, where such a lengthened grade is unproblematic:

    12. sēdēs ‘seat, abode’ may reflect a root noun in view of its lengthened grade root. This root noun would explain the derivation of sēdare ‘to calm, make sit’ (Steinbauer 1989, 142). However, Skt. sadhás-tham ‘seat, abode’, sádhiṣ- (ntr.) ‘id.’ seem to point to *sed-H-(e)s-, so that it seems possible that sēdēs reflects an original h₁-stem (thus Kuiper 1948, 23-35). It seems feasible that a root noun *sēd- and a h₁-stem *sed-eh₁- merged in sēdēs, the inflection of the two differed only in the Nsg. after the loss of the laryngeals and the shortening of the Asg. *-ēm to -em. The difference in meaning between *sēd- and *sed-eh₁- is obscure.

    I see no compelling reason to identify sēdēs with the s-stem OIr. síd ‘peace’. The latter may be compared with Gr. ἕδος, Skt. sadas- ‘seat’ < *sedos. The -i- of síd is probably secondary (cf. W. hedd ‘peace’, with short -e-), taken from the root noun.

    More recently, Stefan Höfler (2014) ‘Notes on three “acrostatic” neuter s-stems.’, Indogermanische Forschungen, vol. 119 (issue 1), pp. 293–338 (https://doi.org/10.1515/if-2014-0016) offers extensive discussion of the previous literature on formation and ablaut grade reflected in Latin sēdēs and Old Irish síd. Since his paper is a bit technical, and it sits behind access barriers, I excerpt for LH readers Höfler’s own account of the *-ē- and the polysemy of síd, which resembles Schrijver’s:

    Because of these irregularities, it has been proposed that sēdēs should be regarded as a remodeled root noun. This seems to be an attractive solution, since it could explain the inflectional behavior and also the vowel length. The starting point would be a root noun *sēd-s, gen. sg. *sĕd-e/os, whence with leveled root ablaut in favor of the strong stem *sēd-s : *sēd-e/os, resulting in Latin †sēs(s) : sēdis. Why the nom. sg., then, was transformed to sēdēs, is an open question….

    Höfler applies solution to the problem of the formation and polysemy of Old Irish síd as Schrijver, after discussing alternative theories:

    The remaining option, therefore, is to compare OIr. síd with Lat. sēdēs, Umbr. sersi and Lep. siteś, and somehow trace it back to a root noun. Admittedly, this is not the most elegant solution, but in view of the alleged inner-Celtic parallel, its likelihood might increase a little. The regular outcome of an already leveled root noun *sḗd-s : gen. sg. *sḗd-os might have been †: †síd (parallel to : ríg m. ‘king’ < *(h₃)rḗg-s : *(h₃)rḗg-os), while the regular standard s-stem *séd-os : *séd-es-os would have led to †sed : †side.

    It now appears feasible to assume that these two words merged into one paradigm at some point within Proto-Irish as some instance of étymologie croisée. One could hypothesize that the possible Scharnierform was the dat. sg. in phrases such as ‘in (the) seat’ and ‘in peace’, which would have produced †í síd for the root noun and †í sid for the s-stem in (classical) Old Irish. Since the two forms differed only in vowel length, it probably would not have been too unreasonable to confound them and eventually fuse them into one lexeme.

    This bold assumption would, then, also be able to explain the two very different meanings ‘fairy mound’ and ‘peace’: One could suppose that the root noun carried the semantics ‘seat, residence’ (thus still Lep. siteś?) > ‘seat, residence of fairies’ > ‘fairy mound’, whereas the s-stem had allegedly developed the specialized meaning ‘peace’ already in common (insular) Celtic times, whence also Welsh hedd ‘id.’ < *sĕd-os.

    This account may seem quite arbitrary at first, but after a thorough look through the attested Old Irish s-stems one will note that as a category, they are a rather heterogeneous group. Beside a few inherited words with parallels in other IE languages, there are a number of s-stems that can be traced back to PIE roots, but without s-stem parallels elsewhere, and also quite a few neuters without any etymological links at all, suggesting that the two latter groups received their s-stem inflection only in Celtic or Irish times. But more interestingly, there might be one or two instances of étymologies croisées within the squad of s-stem nouns that could perhaps support our audacious assumption of *sḗd-s × *séd-es– → *sḗd-es– (OIr. síd). The first example is the s-stem ond (gen. sg. uinde, uinne) ‘stone’ which might owe its peculiar o-vocalism to an analogical influence of, or a merger with a thematic noun that regularly had an o-grade in the root, just as it is proposed for Lat. pondus n. ‘weight’ after *pondus m. (see above note 28), which might be etymologically identical with it (as if from *pend-os ‘heaviness’). We could therefore project a cross between *pénd-es– × *pónd-o– → *pónd-es– (OIr. ond)…

    3.4 As we may now conclude, there seems to be no need to project a long-vowel s-stem *sḗd-os for PIE. ON sætr is morphologically and semantically best analyzable as an inner-Germanic vṛddhi-derivative *sēd-es-o-, whereas OIr. síd most likely represents a cross between the regular s-stem *séd-os, as in Ved. sádas-, Gk. ἕδος, ON setr and Welsh hedd, and the root noun *sḗd-s, continued most probably by Lat. sēdēs, Umbr. sersi and Lep. siteś.

    This account is an advance on what is in Höfler’s dissertation (available here) on the same general topic. On the eventual development of one of the meanings (« ‘seat, residence’ (thus still Lep. siteś?) > ‘seat, residence of fairies’ > ‘fairy mound’ »), note also Irish forad ‘mound, fort, residence’ (probably a virtual Celtic *u̯o-ro-sed- ‘under-forth/completely-sit-’) and Welsh gorsedd ‘mound of earth, tump, knoll, hillock; barrow, tumulus, grave’ (Morris-Jones: *u̯or-en-sed- ‘over-in-sit-’, cited in the GPC; Hamp: a *u̯or-uks-sed- ‘over-up-sitting’ or the like; Patrick Sims-Williams following Loth: perhaps a Celtic *u̯or-sed- compounded only after original *-rs- had become *-rr-).

    As an addendum to my first comment, the first to associate Old Irish síd ‘peace’ with any lengthened-grade forms in Latin like sēdēs seems to have been Henri d’Arbois de Jubainville, ‘Latin sedare “apaiser, calmer,” et irlandais sid “paix”.’, MSL 15 (1908–1909), p. 375 here.

    From the synchronic perspective, the polysemy of síd ‘peace’ and ‘fairy mound; faerie’ reminds me a bit of the polysemy of Russian мир (that is, миръ and міръ) and its congeners.

  26. The strange combination of meanings ‘tumulus’ and ‘peace’ must have its roots in Celtic mythology

    I went looking briefly for the roots of the sídh in Irish myth and and texts featuring combinations of the notions of ‘tumulus, fairy mound’ and ‘peace’. From the Mesca Ulad (edition and translation available here), the account of the final settlement between to the Milesians and the Tuatha Dé Danann:

    Ó do-ríachtatar Meic Míled Espáine Hérind tánic an gáes timchell Túathi Dé Danann. Cu ru léiced Hériu ar raind Amairgin Glúnmáir meic Míled. Úair is samlaid ro baí side rígfili ⁊ rígbrithem. Cu ru raind Hérinn dar dó ⁊ co tuc in leth ro boí sís d’Hérind do Thúaith Dé Danann et in leth aile do Maccaib Míled Espáine da chorpfhini fadéin. Do-chuatar Túath Dé Danann i cnoccaib ⁊ sídbrugib cu ra accallset Sída fo thalmain dóib. Barfhácsat cúicfhiur díb ar comair cacha cóicid i nHérinn ic mórad chath ⁊ chongal ⁊ áig ⁊ urgaile etir Maccu Míled. Bar-ácsat cúiciur díb ar chomair cúicid Ulad int shainruth.

    When the sons of Míl of Spain reached Ireland, their wisdom circumvented the Túatha Dé Danann. Ireland was left to the division of Amorgen Glúnmár, son of Míl, for he was a kingly poet and a kingly judge. He divided Ireland in two and gave the half under the ground to the Túatha Dé Danann and the other half to the sons of Míl of Spain to his own kin. The Túatha Dé Danann went into the hills and fairy regions, and they dug[?] the fairy-mounds underground for them. They left behind, for each fifth of Ireland, five of their number to increase battles and conflicts and strife and struggle among the sons of Míl. They left behind five of their number for the fifth of Ulaid in particular.

    This translation from Grigory Bondarenko (2010) ‘Northern Ireland Autochthons And Otherworlds In Celtic And Slavic’ in Dunja Brozović Rončević et al., eds., Celts and Slavs in Central and Southeastern Europe : Studia Celto-Slavica III, available here (his translation of the phrase ra accallset sída fo thalmain dóib ‘they dug the fairy mounds (síde) underground for them’, follows the DIL under ad-claid). The paper has extensive discussion of this passage and of the whole topic of the síd in myth, with bibliography on the broader Indo-European context on page 285. Note also the text, with translation, of De gabáil in t-ṡída on pages 292ff. Reading these two texts, I do not feel any especial resonance between the notions of ‘peace’ and ‘tumulus’ in the use of the word síd, but then, I am not a native speaker of Middle Irish.

  27. David Marjanović says

    a root noun *sēd-s, gen. sg. *sĕd-e/os

    Oh, that would have vowel length in the nom. sg. through Lachmann’s mad-cackling law – in Latin, but not in Celtic. Hm.

  28. David Eddyshaw says

    Belated thanks, Xerîb. Very interesting. (I’m also gratified to see that my amateur feeling that there was something fishy about the long vowel in síd was justified.)

  29. Oh, that would have vowel length in the nom. sg. through Lachmann’s mad-cackling law – in Latin, but not in Celtic.
    Lachmann is not needed here, lengthened grade would be normal for non-neuter root nouns in the Nom. Sg. In PIE.

  30. David Marjanović says

    …True, probably through Szemerényi’s law getting out of hand.

Speak Your Mind

*