Am I alone in being tired of the orthodox etymologies of these controversial words (which of course make appearances in standard English)?

I should be dearly pleased to know what fellow linguists / logomaniacs make of my thoughts on the subject.

http://www.europaic.com/Etymology%20of%20L.%20femina%20and%20L.%20fellare.htm

I have to confess up front that I am put off by the content and tone of your webpage on the subject.  Example:

How probable is it that a primitive language, Proto-Indo-European ("PIE"), with barely a hundred roots (or morphemes), would have had three such similar yet supposedly distinct roots?

PIE was not a “primitive” language; there has not been such a thing in the recorded or extrapolated history of human language.  All languages we know of have large (and flexible) enough vocabularies to handle all the needs of their daily life and their culture.  The fact that we can only reconstruct a small number of roots obviously does not imply that the language only had that number of words.

As for your problem with the standard etymology, you seem to have misunderstood it.  Nobody I know (of any stature as an etymologist) connects femina with the idea of performing fellatio.  It is from the *dhe(i)- root, which means it’s connected to, e.g., Greek thēlē ‘nipple’ and thēlus ‘female,’ which seems completely unexceptionable.  Fellatio happens to be (probably) from the same root; that does not imply any semantic connection any more than a nice person is ignorant because of the etymology of the word nice.

Here are some other words that have been associated with the PIE root *dhē(i)- ‘to suck’ (nominal forms: dhē-lā ‘breast’, dhi-lo- ‘dug, teat’, dhē-lu-, dhǝi-l- ‘suckling’, dhei-nā ‘pregnant’, dhē-nā ‘fruit crop’, dhedhn- ‘(sour) milk’ (link): Sanskrit dhāyas ‘nourishing’, dhaya ‘sucking, drinking’, dhātrī ‘wet-nurse, midwife, mother’, Greek γαλαθηνός ‘milk-sucking’, θηλυς ‘female, feminine’ (used in reference to both humans and animals), θηλή ‘teat, nipple’, Gothic daddjan ‘to suckle’ (translating Greek θηλαζειν ‘to suckle’, Latin felix ‘fruitful; happy’, Old English delu, Old High German tila ‘teat’, Old Irish dīth ‘he sucked’. The ending -men- is a medio-passive verbal suffix. I don’t find the semantic connection between suck, giving suck, and nourishing or nurturing to be that great a stretch. The other word for oral sex in Latin is irrumatio which is usually interpreted as having the same root as the word ruma ‘breast’ (cf. the folk etymology of the placename Roma because of the wolf that suckled Romulus and Remus).

Languagehat, I do not think we are in disagreement. When I refer to PIE as ‘primitive’, I mean in relation to its more modern derivates eg Latin, Greek etc. As linguists I am sure we all know that primitive means ‘first or earliest’.

Orthodox PIE etymology gives dhe-, dhei- and dheig- as distinct roots. I am making the argument that they are not distinct and indeed that dheig- is incorrectly formed - rather that root is deh- which is a semantic and formal specialism (by metathesis) of dhe-.

I agree with languagehat and jheem that there are many words within this group which relate to nutrition, albeit a relatively small subset, and most of the words you both cite are also in the article. I do not ignore those words at all - instead I argue that they have emerged from the semantic stratum of be nourishing which is itself a specialism of be productive.

The basic point is that orthodox etymology includes L. fellare in this group and, further, supposes that relation somehow informs the semantics of L. femina. This is notwithstanding the fact that, even if L. fellare is connected, the Latin element FE- has dominant semantic content in to produce, be fertile. The sole justification for L. femina as one who suckles is the (imo) dubious or at best tangential reference to L. fellare. It is that semantic content which I attack as at best improbable, if not simply incorrect.

Do you accept L. femina is a neologism? What do you think the neologist had in mind? How would you translate the Latin element FE-? And what is the basic semantic value of PIE. dhe-, dhei-, and dheig-?

Edward Aftung - 12 April 2008 02:13 AM

As linguists I am sure we all know that primitive means ‘first or earliest’.

Which PIE is also not.  It’s just as far back as we can go with our SWAGs.  Primitve also carries with it the connotation of ‘crudely formed, unsophisticated’ and to ignore this is to fall prey to the etymological fallacy.  Whore and charity are derived from the same PIE root, ka-, but this doesn’t mean that there is any relation between the meaning of these words in the minds of any but the most pedantic of native English speakers;

What Faldage said; furthermore, I do not agree that “orthodox etymology… supposes that relation somehow informs the semantics of L. femina.” May I ask for some citations?  And do you teach linguistics?

Even if the conventional etymology of L. fellare be accepted, the conventional etymology and semantics do not begin to attempt to explain the central semantic change from “suck” to “suckle” - and the latter sense is already represented in Latin by a different set of words. In fact, the orthodox etymology does not have the courage of its own convictions: by its logic, L. femina derives its semantic content from a word which means whether literally, by semantic specialization or by tabu-deformation, “to perform fellatio”.

So what if there is a second (or third or fourth) set of words that have roughly the same meaning? This happens all the time. Vocabularies are not tightly constructed and highly efficient logical sets. They are highly redundant, with lots of overlapping and inconsistent semantic value. The fact that there is another Latin verb meaning to suck is irrelevant to the etymology of fellare.

And there is Languagehat’s point that you are stating that the standard etymology of femina is that it derives from fellare. No one claims this. They have a common root.

In the first place it should be noted that L. femina is an isolate - that is to say it is a specifically Latin word which is not related to the normal Indo-European ("IE") root gwena.

Femina is not an isolate. It comes from the PIE root dhei-. Again the fact that there is another PIE root that has produced many modern words for woman is irrelevant. There is nothing that says that all words meaning woman must come from gwena. That femina is part of a group of words relating to fertility, child-bearing, lineage, and child-rearing makes perfect logical sense.

You’re on better, but still very weak, ground when you consider fellare.

In classic Latin, L. fellare was considered an impolite or tabu word (eg as used by Martial and Catullus with the specific sense of “perform fellatio") and by no means a normal or everyday part of vocabulary. This is because L. fellare had (at least from the first century BC and ever since) a distinctly sexual meaning or connotation. Further, the well known linguistic phenomenon of tabu-deformation may have occurred here, in which case the word could mean something entirely different from “suck”.

Impolite words can be rare in recorded literature and still be very common and part of everyday speech. Fuck is a classic example in English. The fact that we have few examples of a sexually charged word in the surviving record doesn’t mean that it wasn’t common. Vulgar words, by the nature of the works that are selected for preservation, are going to be underrepresented in the corpus.

And rather than euphemism (i.e., tabu-deformation), I would suggest this was a case of a pejorative sense driving out the non-pejorative ones. Once a word acquires a sexually charged sense, the use of the non-sexual senses becomes problematic.

Also, a note on your sources. I wouldn’t cite Wikipedia or Etymonline if you wish to be taken seriously. While these sites are fine ones in their own way, I’d rely on more scholarly sources for an article of this nature. It’s not that these two sources are wrong, but they’re less credible. (If an article of this nature cited wordorigins.org, I’d question it.) Go to a good university library and search for any articles on the Latin etymology of fellare and femina and some good Linguistics texts instead of relying on Wikipedia for concepts and definitions of jargon. And find some good Latin dictionaries rather than relying on Etymoline.

As to the question of how old Latin femina is: it probably goes back to the Old Italic period, but it could go back gurther. The suffix -men- is no longer productive in the verbal paradigm in Latin (as it still is in Classical Greek and Sanskrit)). It does survive in a set of words in Latin. (There is another word for ‘woman’ in Latin.) It is mulier, and its etymology is obscure. The PIE root *dhē(i) also shows up in Latin in the words for ‘son’ and ‘daughter’: filius and filia. It is thought that Latin filius is cognate with Umbrian filiu (in Iguvine Tablets 1a.14., tref : sif : feliuf ‘three suckling pigs’).

Edward Aftung - 11 April 2008 11:16 AM

Am I alone in being tired of the orthodox etymologies of these controversial words (which of course make appearances in standard English)?

No, you are not alone! Welcome to this forum, Edward :)

The trouble is being “tired” of something isn’t a very good intellectual argument. Something is either true or it isn’t. Whether one is tired of it is beside the point.

I am interested in the argument of jheem that the -min- in L. femina is a medio-passive suffix. But is this not primarily a verbal,
not a nounal, suffix? (eg Gr. paraleipómena). I note equivalents in Indic languages like -mana and -mdna. Nor does it seem to be particularly common in Latin nouns.

On the analysis of jheem, I take it he constructs the word as PIE. dhe(i)-mena, -mana (that seems to be the implicit result of his argument). This I find very interesting though such an archaic construction would surely be represented in other languages? At least jheem proposes semantic content for -min-, though different from my suggestion, and in that we both go further than the present superficial thinking. It also follows from this argument that jheem is proposing semantic content for L. femina as “she who is (to be) sucked”. This simply ignores the more basic common semantic value of the related Latin vocabulary L. fecundus, fenus, fenum, felix, felios* etc.

I am surprised at the passionate defense of PIE - a language spoken 5000 years ago by expansionistic farmers, and one which has (by necessity) absorbed so many substrata and superstrata from its conquered peoples and other influences. We should be having this discussion in PIE, rather than in and about its more developed descendants, were that language so abundantly rich and subtle. Also many of the threads defend the orthodox connexion between L. femina and L. fellare while at the same time stating that the orthodox etymology makes no such connexion. What is denied is defended. All the etymologies I have seen say that L. femina means “one who suckles” and refers to L. fellare to justify that semantic content. http://www.etymonline.com, just one of many (and easily linked to), simply recites the normal, superficial orthodoxy. For those who deny that, point to a respectable source which does not say so.

I note that the isolation of L. femina, explained by me as an ancient neologism, is denied on the grounds of its connexion with PIE. dhe(i)-. But I aver that relation - I am saying that the form of L. femina is itself a neologism, unlike the basic IE. word for woman *gwen or *gwena. The root is not new - the word is. A study of Latin indicates that it is absolutely full of neologisms, most of which are compounds or extensions of more basic elements. By way of example, the Latin word domus is from a respectable IE. root. L. domina, however, is specific to Latin - at some point in Latin the word domina did not exist, then it was constructed as linguistic demand required, and we can see it has a specific construction (as a feminine adjective) and semantic content mistress of a house. Then the new word lives on as It. donna; Fr. dame etc. PIE. dom* is not new - Latin domina was, at some point, a fresh development of that.

In fact the massive number of local composite neologisms in most IE. languages is one of the main reasons why they differ so much from one another, even those closely related like Latin and Greek, or modern English and German (eg television v fernsehapparat).

If one were to invent the word *agr-eqwo field-horse from two well attested PIE. roots, agreqwo would still be a neologism would it not?

Because L. femina is an isolate, relative to contemporaneous and local IE. languages, and supplements the myterious synonym L. mulier, my attempt is to understand the process of neologism and to consider the more likely motivation and thinking behind it. Why and how did it happen, and what does it mean. The existence of the diminutive L. femella girl, no splash in the pan but still here in modern languages, also indicates that the original speakers of Latin did not give the word semantic content in suckling. Some 19th century etymologist came up with that idea all by himself, when the Latins had no such conception, and it has been repeated without critical examination, including recently in this forum, ever since.

None of the pro-fellare group (if I may use such a term) have explained the short e and -ll- of fellare. This is a serious orthographic difficulty - though I have argued not one at all if understood as connected with the wholly separate and extant L. fell-. Classic Latin, as the polished and much revised product of erudite grammarians, does not generally contain such “spelling mistakes”. The explanation normally offered, such as it is, is the intermediate non-extant form L. *felare. That seems to me a convolution to justify a semantic analysis which itself ignores more obvious, or probable, explanations.

My thanks to Pavlos for the encouragement. Ευχαριστώ πολύ φίλη μου.

Edward Aftung - 15 April 2008 03:36 AM

I am surprised at the passionate defense of PIE - a language spoken 5000 years ago by expansionistic farmers, and one which has (by necessity) absorbed so many substrata and superstrata from its conquered peoples and other influences.

The defenses posted so far in this thread are pointing out that PIE, like all natural languages, was fully formed; and that it, like all natural languages, did not spring forth from nothing, but rather developed from some ancestor.  These are not “passionate defenses”.  They are statements of the obvious.

Edward Aftung - 15 April 2008 03:36 AM

We should be having this discussion in PIE, rather than in and about its more developed descendants, were that language so abundantly rich and subtle.

What a bizarre claim!  It relies on the demonstrably false assumption that once a language is “abundantly rich and subtle” it no longer changes.  I am really trying to avoid ad hominem discussion, but this strongly suggests that you need to read a good introduction to historical linguistics.

Edward Aftung - 15 April 2008 03:36 AM

Also many of the threads defend the orthodox connexion between L. femina and L. fellare while at the same time stating that the orthodox etymology makes no such connexion. What is denied is defended.

This is a gross mischaractization of what the others have actually written. 

Edward Aftung - 15 April 2008 03:36 AM

All the etymologies I have seen say that L. femina means…

Etymologies do not tell us what words mean.  They tell us from whence words derive.  This is not entirely unconnected with their meanings, but it is not the same thing.

Edward Aftung - 15 April 2008 03:36 AM

Because L. femina is an isolate…

No, it isn’t, as the word “isolate” is used in historical linguistics.  You are assigning it an idiosyncratic meaning, apparently without realizing that it normally means something else.  If you hope to construct a persuasive argument, you would do well to find some other way to express this thought.

Thanks, Richard.  I have lost the energy to argue with these people, since it does no good.

To Richard, you object to my use of the term isolate. Enlighten me - in relation to IE. *gwena, what is L. femina. I suspect we can agree is entirely unrelated in etymology. So the Latins had a word of unique formation from a different PIE root (and indeed no word of which I am aware, not even in proper names, from the more common root, despite its attestation in Greek, Germanic, Balto-Slavic, Armenian, all of the Satem group etc). What term would you use for such a rarity - I am willing to be educated. Is there a single word for such a phenomenon? (If I had to invent a term, “idiolog(os)” would work for me.) Is not curious that L. femina and L. mulier are both interesting and difficult words. Or perhaps they are merely found to be so by me, in my ignorance. I should really just accept the explanations others have offered.

A linguistic map of gwena-femina distribution at say 800 BC would be all gwena with a little patch of femina around Latium. That would look pretty isolated to me (and of course dhe(i)- is ubiquitous but that is not the question). How did the Latins lose gwena and obtain femina? That is the question I ask, and attempted to answer. Was it a pure accident (strictly two accidents, of omission and commission). If a complete accident, then of course the word is likely to remain a complete mystery. But people have consciously invented words throughout history. In any particular case, we can attempt to think ourselves into that process, in an attempt to understand.

In the absence of contemporaneous historic verification, I do not see how to reach original semantic content without etymology as at least an excellent guide. In any event contemporaneous explanations often differ and have a strong element of folk-etymology (cf L. femur, femin-, femen-). Comparatives linguistics may give some answers in some cases - but how would they help in this case? If comparative linguistics do assist, where do they take us.

All semantic investigation at such a remote distance is guess-work. I do not assert truth, merely considered theory, through a fresh consideration of the pertinent data.

At some point post-Chaucer, some annoying chap probably said ‘gosh, Chaucer was so silly to spell abominable “abhominable”.’ And the folk-etymology of Chaucer (ab hominem) was corrected to L. ab ominem, abominari. But that would never have happened without critical thought working from fact. Spellings change, semantic shifts occur, orthodoxies (and orthographies) are proved wrong and displaced.

Having said that, ‘adhominable’ could be a useful word in the internet age - “worthy of an ad hominem”, as you say I so nearly brought you to.

Edward Aftung - 15 April 2008 07:45 AM

To Richard, you object to my use of the term isolate. Enlighten me - in relation to IE.

This ground was already covered by Dave:

Femina is not an isolate. It comes from the PIE root dhei-. Again the fact that there is another PIE root that has produced many modern words for woman is irrelevant. There is nothing that says that all words meaning woman must come from gwena. That femina is part of a group of words relating to fertility, child-bearing, lineage, and child-rearing makes perfect logical sense.

While the -men suffix is no longer productive in Latin, there are a goodly set of nominalized adjectives that have it: e.g., acumen ‘sharp point’ and acuo ‘to sharpen, whet’, flumen ‘river’ and fluo ‘to flow’, gramen ‘grass’ (< *grasmen) and Greek γρασω ‘to gnaw’, English craze, lumen ‘light’ (< *loucmen) and luceo ‘to shine, be bright’, Sanskrit rukman ‘shining’, Gothic lauhmuni ‘flame, lightening’. As for dhel- connection with suckle, suck, and femininity, I gave Greek θηλυς ‘female’ and θηλη ‘teat, nipple’ above; both exhibit a long e. A. Meillet in his etymological dictionary of Latin writes in his entry for fēlō: “Les inscriptions, presque toutes vulgaire, ont plutot la graphie fēlō, fēlātor; les manuscrits ont plutot la graphie avec gemination expressive de la liquide fellō, que semble la forme du dialecte abruzze fellatę, ML 3237.”.