Latin dvandva -- update

[David Rodez]

A non-update on my search for a Latin dvandva:

I found this quote in the O.L.D. (s.v. cicur [2]): "Bigenera dicuntur
animalia ex diverso genere nata, ut leopardalis ex leone et panthera;
cicur ex apro et scrofa domestica." (Paul., Fest. p.33M.)

'Animals born of different races are called hybrids, such as the leopard
from the lion and the panther; the hog [?] from the boar and the
domestic sow.'

Nothing too exciting. "Leopardalis" seems to be used as a noun despite
its suffixed adjective-like construction. Then again, given Monsieur
Parsons' recent spiel on nouns and adjectives, I wonder whether Paulus
Diaconus even thought of the words "leopardalis" and "leopardus" (which
is incidentally also attested) all that differently.

Perhaps other authors could furnish more examples of "bigenera". That's
where the research is headed next.

Rodericius

[Johannes Patruus]

I doubt that many of these have equivalents in Latin -
http://www.wordnik.com/lists/animal-hybrids

> Rodericius

Patruvunculus

[Evertjan.]

This list includes many nonbigenera, like "labradoodle",
being "bispecies" [just made up],
where both parents are "Canis lupus familiaris",
[earlier named "Canis domesticus"].

A bigener in the sense used Sextus Pompeius Festus
is a descendent from two different genera [sorts],
a rare event, not of two different species [race].

A typical biological property of a bigener is that they are
nonreproductive between two of them [or so nearly so], so that a new
gener cannot develop. They must be practically sterile, otherwise there
would be no sense in calling them bigener.

Interspecies breeding is common. Intergeneric breeding, while sometimes
also common, does mostly not lead to any or viable offspring.

--
Evertjan.
The Netherlands.
(Please change the x'es to dots in my emailaddress)

[John Crinnion]

David, am I being slow on the uptake, but what is this word
'dvandva'? I entered it into the alt.language.latin search engine,
and only one entry distinct from this thread came out - it looks
fascinating, but I cannot understand it either. Please enlighten!!!

[Johannes Patruus]

The short answer -
http://everything2.com/user/Muke/writeups/dvandva

The medium answer -
http://en.wikipedia.org/wiki/Dvandva

The horrendously long answer -
https://docs.google.com/viewer?a=v&q=cache:XaASaYKyW1oJ:www.euppublishing.com/doi/abs/10.3366/E1750124508000044+&hl=en&gl=uk&pid=bl&srcid=ADGEESj9KzI98ciNyxdQeb0QkNXZk7FWPwWZJUhcCutZccdjlTOMIba3g9t_8gmyqR3Ctc4rNBMrPE_om9mWiUq4eHOD6oNiLfZKqSu2ly-6Abc65HPrOSE3TS6OR9taDA3pSaqu9R7A&sig=AHIEtbR2rmUCvyK2M-p7Nah7YgpVzAax1w&pli=1

Patruus

[Ed Cryer]

Ecce modus facilis quo responsum horride longum fiat aliquantulo brevius.
http://tinyurl.com/c4ma5u7

Ed

[David Rodez]

I also noticed I had recorded "hircocervus" in my notes. It is
apparently a calque of the Gk. tragelaphos. Unattested in L&S. It makes
its English debut in 1398.

Rod.

[Johannes Patruus]

Hircocervus is medieval Latin but tragelaphus is classical -
http://archimedes.fas.harvard.edu/cgi-bin/dict?word=tragelaphus&name=ls

> Rod.

Patruus

[David Rodez]

Bauer's paper is highly illustrative. Thank you, Patruus. Of particular
interest to me was the distinction between dvandva compounds (see Fig.
1) and karmadharaya compounds (see Fig. 2). According to Bauer, the
compound I am looking for, compounds of the sort singer-songwriter or
philosopher king, are not dvandvas in the sense used by Sanskrit
grammarians. This type Bauer calls "appositional compounds". "They pick
out the intersection of two sets and name two aspects of a single
individual, not two individuals." Like dvandvas they are coordinated
(i.e., the members are syntactically coordinate) but unlike dvandvas
they are semantically headed.

Rod.

[David Rodez]

Ha, would you look at that!

Rod.

[Ed Cryer]

What if there are three in the thing, as in "sŭŏvĕtaurīlĭa"?

Ed

[David Rodez]

I don't know about suovetaurilia, but *suovetaurilis would surely fall
under Bauer, 2.12. That said, mihi nescio videtur that suovetaurilia is
headed because the animal morphemes clearly modify the -alia, which
stands in for 'feast' or 'event'.

I'll admit I don't fully understand Bauer's terms. For instance, if the
noun *suovetaurus existed and signified an animal that in some way
combined the features of a sus, ovis, and taurus, would the term be an
appositional compound or would it be a dvandva of the additive sort (see
2.7)?

Is the essence of the dvandva, per Bauer, the fact that the members do
not semantically fuse in any way when they unite to form the composite?

These are tedious and uninspiring questions really, but they do reveal
something about the pliability and creativity of human language.

Rod.

[David Rodez]

A compound not unlike suovetaurilia is strufertarii, attested only in
the plural apparently. A strufertarius delivered two kinds of sacred
cakes, a strues and a fertum.

I think the answer, if there is one, will lie in some old ritual term
like these. Perhaps Latin originally had many more types of compounds
like Sanskrit and they gradually became obsolete as the genitive-case
terminations evolved.

Rod.

[Ed Cryer]

We had a long thread on something in this line a couple of years ago. We
ended up scouring Latin dictionaries, and then got into text editing and
searching.

English; philosopher king, Batman, Cat woman, catfish etc.
Greek; Minotaur and many others.
Latin; only Greek imports. The nearest autochthonous one we got was
"sŭŏvĕtaurīlĭa" (found even in Livy), but that seemed to miss the
criteria as, although a compound, it didn't reference a living creature.

Ed

[David Rodez]

How can I forget? You and Patruus helped me locate the raw text of L&S
which has since been so useful to me. It's my hope I find you in Elysium
some day to thank you personally.

I happened to stumble upon the dissertation of F. Skutsch, _De nominum
latinorum compositione quaestiones selectae_. He first lists three
compounds he believes are truly coordinative: suovetaur-ilis,
scytalosagittipelli-ger, and strufert-arii. Then he confirms with these
eloquent words what I had been thinking:

"Nemo sibi persuadebit hominem Romanum unquam talia dixisse, qualia
secundum indicas horum compositorum leges expectamus: *suovitaurum sive
*scytalosagittipelles. Elucet igitur nisi apposito sive suffixo sive
altero compositionis membro, quale est illud -ger, fieri non potuisse,
ut Romani compositum copulativum effingeret.

"Et in hac re summa inter indica et latina composita differentia inest:
nempe latinorum unitas semper ita tantum valet, si ad tertiam quartamve
notionem referuntur, quare non *strufertum dicitur, sed strufertarius is
qui strue et ferto sacrificat. [...] Nullum igitur in lingua latina
invenitur vocabulum quod Indorum dvandvicis satis apte respondeat."

My poor and very loose translation:

No one can be convinced that a Roman ever said such things as we might
expect according to the Sanskrit rules of these [the aforementioned]
compounds: *suovitaurum, *scytalosagittipelles. It is clear therefore
that without an apposited, suffixed, or other element of composition,
such as -ger, it was not possible for the Roman geist to form a
copulative compound.

Moreover, in this we find the main difference between Sanskrit and Latin
compounds, namely: [compositorum] latinorum unitas semper ita tantum
valet, si ad tertiam quartamve notionem referuntur.* Hence, the ritual
offerer of strues and ferta is called not a *strufertum but a
strufertarius. [...] Therefore there is not found in Latin a word that
aptly corresponds to the Sanskrit dvandvas.


*An awkward way of saying that whatever collective is indicated by the
term must be referred to a third or fourth morpheme that is part of the
compound, such as "strues fertumque" are referred to a doer indicated by
the suffix -arius in strufertarius.

The dissertation could be found contained in a larger work by Skutsch at
the Internet Archive:

http://www.archive.org/details/kleineschriften00skutuoft

Rod.

[David Rodez]

On 4/10/2012 6:35 PM, David Rodez wrote:

> Moreover, in this we find the main difference between Sanskrit and Latin
> compounds, namely: [compositorum] latinorum unitas semper ita tantum
> valet, si ad tertiam quartamve notionem referuntur.* Hence, the ritual
> offerer of strues and ferta is called not a *strufertum but a
> strufertarius. [...] Therefore there is not found in Latin a word that
> aptly corresponds to the Sanskrit dvandvas.
>
>
> *An awkward way of saying that whatever collective is indicated by the
> term must be referred to a third or fourth morpheme that is part of the
> compound, such as "strues fertumque" are referred to a doer indicated by
> the suffix -arius in strufertarius.
>

>

That line--[compositorum] latinorum unitas semper ita tantum valet, si
ad tertiam quartamve notionem referuntur--is actually a holy grail of
sorts, and I felt I didn't do it sufficient justice in my last post.

‘The unit or collective (unitas) yielded by Latin composites has meaning
or semantic force (not on its own but) if and only if it is referred to
(or, in lay terms, undergirded by) [some] third or fourth morpheme’.

I take it to mean this. The referent must be indicated within the term
by some element of composition like a suffix. Whatever collective is
expressed by the compound must be referred to a third or fourth morpheme
that is part of the compound (the term itself), such as “strues
fertumque” is referred to a doer indicated by the suffix -arius in
strufertarius.

As a corollary, the referent itself cannot be the compounded unit;
otherwise, an equivalence, expressed by a compound such as *strufertum,
between a referent that is some mix of a strues and fertum, and the
grammatical compounded unit strues fertumque, would be conceivable.

Rod.

[Ed Cryer]

Vi et armis (by force of arms), far more common in native Latin writers
than "vi armorum".
Grammarians call this a "hendiadys" (one through two) but maybe there's
a deeper rule in operation; the one you discuss above.

Ed

[David Rodez]

That's a very interesting point, Ed. Maybe Latin just preferred to
employ hendiadys where other languages used dvandvas. If you think about
it, there's only a cosmetic difference between a hendiadys and some of
the dvandvas in Bauer, particularly the hyponymous ones (e.g.,
hands-feet for 'limbs' or job-work for 'business'). For, although the
elements of a hendiadys are written separately, they are meant to be
interpreted as a unit just like dvandvas.

"vi et armis" (force-arms) 'belligerent force'

"arma virumque" (arms-hero) 'heroic warfare'

Rod.