21st Century Classical Greek -- turn it over and over

If you love science fiction, you may remember the episode on Star Trek:TOS where the old man says, "All is not as they tell us, for the world is hollow and I have touched the sky." Then he screams and dies.

Well, we're not going to scream and die, we're going to find out that something I've said almost since I started this blog is true. Classicists are not linguists, they are not scientific, they are more like alchemists.

The subject is, once again, the cognate accusative, which I have slammed on my Hebrew and Greek threads, and now I have more support for slamming it from my Arabic studies that have helped me with my study of Samaritan scripture.

The concept of cognate accusative derives from Latin and it’s a perfect example of somebody inventing a label for something that doesn’t deserve special notice.

The Latin grammar of Gildersleeve, section 333.2 on page 151, gives the definition: "When the dependent word is of the same origin or of kindred meaning with the verb, it is called the Cognate Accusative, and usually has an attribute."

The citation is Plautus, Rudens, Act III, mirum atque inscitum somniavi somnium.

It is the only citation.

If Gildersleeve knew what he was doing, he would give other examples to show the similarity of features. If he was a real maven, he could also give example of uses of the accusative that look like this but are not cognate. In a science you can give examples of what does or does not meet your definition.

The Latin grammar of Greenough says differently in section 238 on page 238. The verb has to be neuter, whatever that means, used with an accusative of similar meaning, modified by an adjective. The student is supposed to understand that the adjective modifies the accusative. This is quite different from Gildersleeve. If you’re working with a science, everybody has to have the same definition. So cognate accusative isn’t really a thing in Latin.

Greenough gives four examples, but what is the kinship between the noun and verb? They are nouns you would expect to be used with those verbs: generation and live; join together in alliance; re-echo and a name; thunder and the direction of thundering; looking at somebody and a noun describing the manner. There's nothing special enough about this to deserve a label separate from ordinary verb-object grammar.

This is how classicists create what I call mirages, slapping labels on things too poorly defined to warrant special treatment. And then they slapped them onto Greek.

Three of the main Greek grammars (White, Goodwin and Smyth) discuss this topic and each one has a different definition. That is not scientific. In science, you may extend or refine a definition as new data turns up, but you don't have a scientific term if everybody defines it differently.

Classicists also slapped it onto Biblical Hebrew, which doesn't even have an accusative morphology, but BH does something specific. Genesis 27:34 has the phrase  וַיִּצְעַ֣ק צְעָקָ֔ה גְּדֹלָ֥ה meaning to cry out loudly. Not only is the noun kindred to the verb, they have the same root. 

So then I looked back at Greek and found Demosthenes, Against Aristocrates 121, καλήν γ᾽ ὕβριν ἦμεν ἂν ὑβρισμένοι. “what a fine insult by them insulting us”. It’s the same structure as in Genesis and it uses accusative.

But what I did in my grammar of Biblical Hebrew was call this duplicate adverbial, because it duplicates the verb root to create an adverbial phrase. That puts it in line with my other two duplicates, conditional and unconditional, which have to do with law so I won’t explain them here. Look on my Hebrew thread.

Now we get to Price’s All the Arabic You Should Have Learned the First Time Around and he, too, babbles about a cognate accusative BUT he admits that the noun is not always in the accusative. Which is a good thing because in my notes I'm calling it Nasb like the Arabs.

IT'S NOT A COGNATE ACCUSATIVE. IT'S A VERB AND A NOUN FROM THE SAME ROOT, OFTEN WITH THE NOUN MODIFIED, TO PRODUCE AN ADVERB OF MANNER THAT MODIFIES THE VERB. If the language has case morphology, the noun will be in an oblique case.

Now let's go back to Greek. If you ignore which case the noun is in, as with Arabic, you can find other examples that no classicist would recognize because they don’t use accusative. One is Aristophanes’ Plutus, 1044: τάλαιν᾽ ἐγὼ τῆς ὕβρεος ἧς ὑβρίζομαι. “wretched me being so horribly insulted.” This uses the dative.

Why does it work to plug the result from Arabic into Classical Greek? The history.

The Semitic languages all have the d.a. Thus it descended to them from proto-Semitic.

The NE Anatolian J1/J2 Neolithic farmer subclade extended all over Anatolia and the Caucasus, and reached Minoan civilization which influenced the Greeks.

Linear B, the script of the Ahiyyawa pre-Hellenic Greeks has the d.a. So does ancient Egyptian, and remember that the Ahiyyawa were called Pelesh (Pelishtim) by the Egyptians.

Tocharian has the d.a. So does Sanskrit. So does Ancient Persian. Linguists use the Semitic description for all these languages. So if it exists in Classical Greek, all the examples have to meet the Semitic definition.

Now circle out. We find it in Chechen, a Caucasian ergative language. We find the d.a. in Hurrian, another ergative language from middle Anatolia, next to Wilusa, and in Basque, an ergative language of NW Europe. The ancestors of the Basques left Anatolia by 8000 BCE. In these languages the structure fits the Semitic definition. It is cognitive dissonance to call the structure cognate accusative because the point of an ergative language is that it HAS NO ACCUSATIVE.

Classical languages do not have a cognitive accusative. The three I’ve studied carefully have a duplicate adverbial: a verb plus an adverbial modifier consisting of a noun from the same root with a noun modifier. The noun may at first blush look like a normal object of the verb, but that noun modifier is what turns the whole thing into a duplicate adverbial. Since Latin doesn't have its act together, we have room to say neither does it have a cognate accusative. Since it's an Indic language, somebody needs to do the research to find the d.a.s in there.

This supports what I’ve said for decades now. How people teach language derives from how they were taught. These mirages keep getting taught because nobody does the objective research that blows up the definitions. They don’t do it within their languages; they don’t do it across languages. They may say that they don’t have time to do it, due to press of business, like the old publish or perish thing. But you have to do research to publish, so the real issue is that not one of them ever questioned what they were taught, or else they were pressured not to embarrass their colleagues by showing them what’s objectively wrong with what they’ve been teaching.

It's like alchemy. Certain things were taken for granted because all the alchemists for centuries had taught it so. Each alchemist wanted to find the secret and all of them kept their method private so nobody could get ahead of them.

That's not how science works -- and as soon as alchemy turned into chemistry, everything the alchemists believed got blown up anyway. 

If people would use real linguistics to study classical languages, we could clear out a lot of crappy claims and then really learn something about the cultures that used those languages. I've done it with Biblical Hebrew and Classical Greek. Now I'm doing it with Arabic. One of you needs to do it with Latin. 

It's only taken me a dozen years to get this far. You've only got a hundred years to live. Go to it.