Tuesday, November 6, 2012

Quality of Sources: Godfrey Higgins, pt 2

Godfrey Higgins: The Irish Druids

Having not read the entire work yet - not even an entire volume - I do not know entirely what is in store for me. I have seen the occasional bits that do give me an idea about the mental agility of this man, however. It seems as though Higgins categorically rejected the use or value of anything which he could not wrap his mind around, and it seems his mind was not willing to imagine much beyond the scopes of what he was used to. 

I do grant, his theory is innovative and weird, and thus not the kind of thing he was used to, but he seems to have accepted weird things on that level - it is when the details of how something works are out of his capability of understanding things that he fails to accept that there may be something to it.

A good example of this is his inability to grasp how Hebrew worked:
It is impossible to say what would be the first step which would be adopted by man in the formation of language, but a probable near approximation to the truth I think may be contemplated, merely by the application of a little common sense. His first words would consist of simple sounds, which, by our present system, might be represented by such words as ba bal; or by our vowels, a or ah. They would be nouns or the signs of things. They would have only one number, one case, one gender. After some time, probably by very slow degrees, the words which we describe by the other parts of speech would arise. The different cases would be formed by adding words in the most simple manner; as bit, house; le bit, to house; be-bit, by house. I can have no doubt whatever that the written language which comes the nearest to this simplicity is the Hebrew of the Synagogue. Its cases are made, like the English, by words, and its use of the rude form instead of a genitive case, called being in regimine, is a peculiarity which marks almost primeval simplicity. The words which denote the past and future times may be used one for the other, and which is to be adopted, in any case, can only be known from context. They have been attempted to be subjected to rule by what is called the vau conversa, but I think in vain. Many other marks of rudeness might be pointed out. [1, p. 739]
The latter half of this section stinks with linguistic naiveté. He never explains why the use of prepositions can be taken as a mark of primitiveness, and his using a Hebrew phrase and badly translating it, making it sound like bad English in the process is pretty much a fallacy of its own.

To be more clear: Hebrew בית (bayit), by itself, means "a house". "The house" would be rendered הבית, ha-bayit. Is this somehow less primitive than Latin entirely lacking articles? Further, when prepositions are applied, usually written as prefixes, we obtain לבית ,בבית - lebayit, bebayit (note: my transliteration here is coarse; do not use it as a guide to pronunciation; it is just here for making a point, not for teaching Hebrew. It is also influenced more by Biblical Hebrew than by Modern Hebrew, but I have influences from both, which makes my transliteration somewhat uncertain and haphazard. In Biblical Hebrew class we did follow some kind of European continental standard pronunciation, but that doesn't preserve consonant distinctions made in Biblical, and for some reason, I have adopted an Ashkenazi pronunciation on my own. What can I say, I like the sound of final -s over final -t.) These translate roughly as "to a house/to the house" or "by a house/by the house". However, in text with niqqud, a further thing is marked, a thing that usually is clear from context. The first consonant is geminated (this is marked by having a dot in it, the dagesh forte, giving us בּ and similar letter-forms), obtaining lebbayit, bebbayit for "to the house" or "by the house", so in a text with niqqud, we can be entirely sure of which nouns with prefixes are definite and which are indefinite.

Now, it might seem weird to a speaker of English that such an all-important difference as the vs. a would be left unmarked in writing, even if there is a manner of clarifying that is even mandatory in speech. Try, however, to have a word-editor delete all instances of " the "," a " and " an " from a text, and see how many of them you can guess right. (Note: you'll need to keep spaces that I included in quotation-marks or words like 'theory' and 'apex' will suffer misfortune of having parts cut out. And any native speaker knows exactly where to insert "the" or "a" in that sentence without me having to tell them!)

Many languages around the world lack articles, c.f. Russian, Finnish, Latin, Korean, Urdu, Berber, Panjabi, Evenki, Ket, Yup'ik, Dhargari, Ukrainan, Estonian, Serbian and Croatian, Hindi, Canela-Krahô, Fijian, Iban, ... [2, 3]. In a sample of 534 languages (there's about 6000 in total), 198 lacked both indefinite and definite articles. In languages that mandatorily do mark definiteness, it isn't unusual that this marking only is applied to subjects and objects, or even just to objects. In another sample of 600 languages, 198 had no articles. It is likely the latter sample was expanded a bit to include some known outliers? 

Out of a sample of 261 languages, Iggersen found a full hundred with no case marking. 

Further, Higgins posits that the Hebrew way of constructing possessive phrases is somehow primitive. For those who do not know this, Hebrew nouns have two forms that are a bit like cases, but with a somewhat different function. A noun by itself is in status absolutus, whether it be object, subject or any other such constituent of a sentence. When possessed, it is instead in status constructus, which for some nouns is a somewhat reduced form, that is, it often has shortened vowels or otherwise reduced vowels. For some nouns it is identical and for some it gets a separate suffix, e.g. feminines have torah - torat תרה - תרות, behemah - behemat בהמה בהמות, and masculine plurals get alternations like susim - suse סוסימ - סוסי. 

In the plural, the plural marker for feminines is identical for the construct and absolute states, although the vowels can be reduced a bit in the construct state. Now, my Biblical Hebrew is not good enough to spot grammatical mistakes, but good enough to work my way through a text if I need it, so do not take these things as certainly correct - take them as something in the right ballpark. Now, saying "the man's house", you take the singular constructus of house - beit (c.f. the absolute state: bayit, which already has been mentioned) and stick it before the noun man, ish - habeit haish. The fact that the house is the inflected noun in this may seem weird to speakers of English, after all, it is man's, not man house's or some equally weird construction in English. The Hebrew basically is a bit like if "the house of the man" would be reanalyzed by future generations and then rendered as thehouseof theman.  

And sure, this seems weird. Until we look at what wals.info has to say about it. English does what is known as dependent marking. This means the marked noun is the one that is the attribute of the other. Hebrew does what is known as head-marking, marking the noun that is the head of the noun phrase. These are not the only approaches available - some languages mark both, and some mark neither. In the former, you'd have something like houseof theman's. In the latter, you'd have house man or man house to express 'the man's house'. In a sample of 236 languages, 98 follow the English model, marking the noun with some kind of genitive-like case. Ancient Hebrew, however, is in a category that is not significantly smaller: 78 languages in the sample mark the head noun. A further 22 mark both, and 32 mark neither. Enigmatically enough, the sample has 6 outliers that do not adhere to any of these options, the article related to the feature at wals.info explains how those work. [7]

Well, apparently this boggles Higgins so much that he must posit that a language that is not like English is primitive. 

Biblical Hebrew actually lacked a past-present distinction in its verbs, and his confusion has to do with assuming that the distinction present in the verb was the same as the distinction made by English. It had, instead, a thing linguistics of today knows as aspect. Aspect, most famously, is a central feature in most Slavic languages in conjunction with tense. It also occurs in Georgian, in various Finno-Ugric languages, in ancient Greek, Sino-Tibetan, various Asian, African and American languages. Western European languages kind of merge some tenses with aspects in a somewhat complex manner; this too is common through languages worldwide. A wals.info sample of 1132 languages, obtains that more than 10% - 123 languages - have neither aspect or tense marking on their verbs. [4] Anyways, there are reasons why aspect and tense easily get conflated in languages over time, as there tends to be a correlation - we will often speak of actions as being completed when they also are past, hence parsing 'completed' as 'past' is a reasonable thing. However, we can speak of a completed action in the future - "once I have sold the house, I'll buy a small apartment somewhere" is an example of such, where we think of what comes after a completed future action. [5, pp. 82-83] In Biblical Hebrew, completion vs. progress at the time of which the speaker is thinking was more central than time at which the action occured compared to the present. As the wals.info sample points out, many languages do not even have either in its verb forms.

The Sino-Tibetan languages are sometimes quoted as languages rich in aspects but poor in tenses. The problem here is that Higgins let his own ignorance of language and his inability to conceive of a language not lining up with how English categorizes the world be the standard by which he measured how advanced languages were - marking different categories on verbs was too much for him to grasp. I do realize he didn't have access to Bernard Comrie's "Aspect: An Introduction to the Study of Verbal Aspect and Related Problems", as it would not be written for over a century after his death, but ... being wrong is still being wrong, even and maybe especially if a better understanding of the topic one was wrong about appears after one's demise. 

I could waste a lot of electricity writing up an explanation as to why the argument Higgins presents is misguided and the method he uses to find "the original language" is downright unfit to reach any conclusion about historical linguistics whatsoever. I think I have provided enough information now about languages in the world to show that at the very least his conclusion regarding "the original language" is based on faulty data, and a sufficiently good demonstration that his understanding of the languages he was working with was flawed, imperfect and downright dumb by modern standards.

Not only did he have too little data, what data he had he misunderstood. He also drew far wider conclusions from it than was warranted. Someone who labels themselves "a linguist" using such a source in their research is beyond belief, and this is a good indication that D.M. Murdock does not really have any kind of mastery of a field in which she claims to be a scholar.

[1] Godfrey Higgins, The Irish Druids. Available at archive.org
[2] Matthew S. Dryer, World Atlas of Language Structures, feature 37A,  http://wals.info/feature/37A
[3] Matthew S. Dryer, World Atlas of Language Structures, feature 38A, http://wals.info/feature/38A
[4] Matthew S. Dryer, World Atlas of Language Structures, feature 69A, http://wals.info/chapter/69A
[5] Bernard Comrie, Aspect: An Introduction to the Study of Verbal Aspect and Related Problems, 1981
[6] Oliver A. Iggersen, World Atlas of Language Structures, feature 49A, http://wals.info/feature/49A
[7] Johanna Nichols and Balthasar Bickel, World Atlas of Language Structures, feature 24A http://wals.info/feature/24A

Copypasta notice:
I did an actual act of copypasting from another source into this article: I did indeed plagiarize an entire letter from wikipedia, a plagiarism that would shame me greatly if I did not admit to it. Since I am not very acquainted with the Hebrew keyboard layout, the entire letter "בּ" did pass from wikipedia through my clipboard, instead of me looking up a keyboard layout map and finding how to obtain the desired letter.

No comments:

Post a Comment