8

Is the etymology of the word "semen" (eng. "seed") in Latin connected to the hebrew word שֶׁמֶן "shemen" (eng. "ointment")?

I've just read a peculiar article that attempted to make this connection:

The word for "ointment(s)" here in the Hebrew is SHEMEN meaning "ointment, oil, fat, cream, fertility". This word is almost certainly the origin for the Latin word "semen" meaning "seed", from which we get our English word "semen".

The context is the "Song of songs" book from the Bible. The coincidences are interesting: Song of Songs is an erotic poem, there is phonetic similarity between "shemen" and "semen", and the meaning "ointment" that could refer to sperm.

But then, this could just be a coincidence. Can anyone disprove that the latin "semen" is not derived from or related to the hebrew "shemen"?

André Staltz
  • 199
  • 1
  • 4
  • 16
    You've got it back-to-front. The person proposing an etymology should present evidence to back it up, not the person disputing it. A rough phonological similarity and some tenuous semantic similarity are not evidence. Anyway the etymology of Latin /semen/ can be traced back to proto-Indo-European so that would seem to disprove the connection (unless you want to claim it was borrowed into pIE from proto-Semitic?). – Gaston Ümlaut Aug 22 '12 at 10:27
  • 1
    Well, I'm not proposing the etymology, I'm asking whether the etymology this article proposed has good basis. – André Staltz Aug 22 '12 at 20:10

3 Answers3

22

Latin semen comes from sero, "to sow", which in turn comes from the Proto-Indo-European *sē(i)-, "to sow". (The Latin r is from reduplication and rhotacism between vowels, in that order. The reduplication happened in Proto-Indo-European.)

The Latin suffix -men is related to Greek -ma as in comma, stigma, telegram[ma], etc.. It is quite common in Latin, as in limen, gen. liminis (English sub-lime), omen, nomen. It probably existed in Proto-Indo-European already, since it also exists in Germanic and Slavic; the ancestor of semen was probably formed in Proto-Indo-European, not in Latin.

Unless we could prove that Hebrew borrowed it from Latin or Proto-Indo-European, or that Hebrew had the same morphemes, I think it is safest to assume that this is a coincidence, as it is usually the case with resemblances across different linguistic families.

Cerberus
  • 7,976
  • 33
  • 49
  • 1
    Thanks, I'm a layman in all this. Apparently it is a coincidence. – André Staltz Aug 22 '12 at 15:09
  • 2
    The suffix also occurs in English words such as "seam" (that which is sewn) and "gleam" (that which glows). – Colin Fine Aug 22 '12 at 16:12
  • 4
    In Latin, its IE origin is not so obvious because of (1) rhotacism in a (2) reduplicated stem. However, in Modern (!) Russian, the verb is "se-jat'" and the noun is "se-mja" (NOM. reduced stem), "semen-i" (GEN.SG. full stem). – Alex B. Aug 22 '12 at 19:32
  • @AlexB.: Ah, I was wondering where the r came from (too lazy to look it up). Do you have any idea why its stem was reduplicated? This does not normally happen with non-perfective Latin verb stems. I also wonder where the p and the r came from in Greek sper-... – Cerberus Aug 22 '12 at 22:07
  • 1
    @Cerberus, in PIE there was a group of verbs that got reduplicated, and Latin sero "sow" is simply a reflex of this process. Sihler mentions the following verbs in Latin of this type: sido, gigno (cf. genui), sisto (cf. sto). Note: in Infectum, reduplication and zero degree (sero < si-sH1); whereas in Perfectum there is no reduplication (sevi < seH1-vi). – Alex B. Aug 23 '12 at 02:49
  • @AlexB.: Ah, yes, gigno and sisto, of course—why didn't I think of those? So it happened in PIE, that makes sense. In those verbs the root is obvious. I suppose it was just not productive in Latin? Perfective reduplication was quite productive in archaic Latin, I believe, as in fefeci, poposci/peposci, memini, fefelli, peperi, steti (probably), etc. – Cerberus Aug 23 '12 at 13:08
  • 2
    @Colin and Alex: I have added some information from your comments into the answers, thanks. – Cerberus Aug 23 '12 at 13:17
  • 2
    Very good. Also, שמן shemen "fat, oil" is related to Arabic Arabic: شَحْم shaħm "grease", so one would indeed have to go back to at least Proto-(West)-Semetic to find any purported common root with PIE. – Mark Beadles Aug 23 '12 at 16:55
  • 1
    @MarkBeadles: I.e. to the dark unknown...but it would be interesting to compare the Proto-Semitic roots. I just don't feel qualified to do so. – Cerberus Aug 24 '12 at 01:05
  • @Cerberus: but note that the PIE root has only one consonant. The other two are part of the suffix. – Colin Fine Dec 15 '12 at 13:26
  • Latin "semen" comes directly from PIE see̯mn (сf Russian семя). – Anixx Aug 28 '13 at 07:05
  • 1
    @Mark Beadles, where are you getting the connection between shemen and shaħm? Those don't seem to share anything other than the first consonant (I don't see how the M's can be related since in the Hebrew root that's the second consonant, in the Arabic root the third). – TKR Oct 07 '13 at 02:34
  • @Cerberus: Gk. sper- is an unrelated root; it may have some cognates in Armenian (about which Chantraine is somewhat dubious), but has nothing to do with *seh1-. – TKR Oct 07 '13 at 02:37
  • @Cerberus: Is there a greek cognate ma-stem to lat. sēmen? I instantly thought of greek ἧμα 'that which is thrown' because of its fitting semantics but it belongs to ἵημι. –  Oct 07 '13 at 15:37
  • @aorists: Chantraine says *sē- "to sow" is probably only reflected in the western Indo-European languages, so not in Greek; it notes, however, that some scholars believe there may have been some later contamination, in which case ἵημι would come from *sē- via some western Indo-European reflex. – Cerberus Oct 07 '13 at 17:36
  • @TKR: Ah, it appears the etymology of speirô is quite a minefield. Chaintraine says it comes from PIE *sper-, as you say, and Hofmann that it is related to Spreu "chaff". Duden says this word is related to the likes of spray and Dutch sproeien, root *sp[h]er[ə]- "to sprinkle", and possibly to sprinkle, Dutch sprenkelen; the Woordenboek der Nederlandsche Taal says the latter group of related Germanic words is from a causative sprengen, "to cause to spring", from spring(en). – Cerberus Oct 07 '13 at 17:56
  • ... But the WNT says about the verb springen, "origin uncertain, perhaps from *sp(e)rʰ- "to hurry". Duden says Spreu is probably related to Spur "spoor" and its root *sp[h]er[ə]- "to hurry" is probably the same as *sp[h]er[ə]- "to sprinkle" above and related to springen. Latin sperno "to spurn" may be related to spoor. – Cerberus Oct 07 '13 at 17:56
10

Hebrew šɛmɛn שמן “oil, fat” is a Semitic cognate of Arabic samn سمن “fat, butter” (with Semitic s1). It is not related to šaḥm شحم “fat, grease” (with Semitic s2 and ḥ) or to the IE words mentioned above.

It is true that šɛmɛn looks superficially like Latin semen, but the vowels of the former are the result of a specifically Hebrew development (Semitic qatl > Hebrew qɛtɛl; so-called segolisation). If you posit a proto-Semitic *šamn- the similarity with the Latin word becomes considerably less.

fdb
  • 24,134
  • 1
  • 35
  • 70
  • Given that the PIE form was *see̯mn*, the similarity does not go away. – Anixx Jul 20 '16 at 13:40
  • 3
    There are two hypothetical possibilities. One is that Sem. šamn- and IE seh¹-mn have a common “Nostratic” ancestor. But in this case one ought to expect the laryngeal to have survived in Semitic. The other is that some borrowing has happened, but one would have suggest who specifically borrowed from whom. But in either case, the main difficulty is that šamn- means “fat, grease” and that seh¹-mn means “seed”. The semantics do not really match. – fdb Jul 20 '16 at 13:57
  • is the -mn suffix known in proto-Semitic or proto-Afro-Asiatic? – Anixx Jul 20 '16 at 14:12
  • No, it is an IE suffix. – fdb Jul 20 '16 at 14:15
-2

The word for "seed" in PIE was see̯mn. It indeed uses the -men- suffix (in zero-grade).

Anixx
  • 6,643
  • 1
  • 26
  • 38