1 of 2
1
Related; inherited; cognate; and any other similar terms
Posted: 22 March 2008 10:04 AM   [ Ignore ]
Avatar
RankRankRankRank
Total Posts:  1369
Joined  2007-01-29

Please would someone explain in very simple terms that even I can understand, precisely what these linguistic terms mean.  I thought I understood them until I read the Laguvulin thread.

Profile
 
 
Posted: 22 March 2008 10:30 AM   [ Ignore ]   [ # 1 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

"Inherited” means a changed form of the same word in an earlier form of the language.  Father is inherited English vocabulary in that it is the same word as Old English fæder (i.e., King Alfred said fæder and his kid learned it from him and passed it on to his kids and so on down the generations, with the sounds gradually changing to the modern form); the Old English word is inherited Germanic vocabulary, and the Proto-Germanic word (reconstructed as *fadar) is inherited Indo-European vocabulary (i.e., the folks wandering around the steppes circa 3000 BC called their fathers *patēr or something similar, and their children passed the word down the generations as its sound gradually changed to *fadar, fæder, and father in English, pater and thence père, padre, etc. in Latin and its descendents, etc.).

“Cognate” means (in its strict or technical sense) that two words are descended/inherited from the same earlier form; English father and German Vater are cognate because they both descend from Proto-Germanic *fadar, and they’re both cognate with Latin pater and Greek πατήρ because they all descend from PIE *patēr.

I’ve been using “related” to mean “cognate,” because the latter struck me as an overly technical term, but apparently lots of people think it’s unnatural to say words borrowed from the same source aren’t related, so I have vowed to try to avoid that usage.  (This came up in the context of words for ‘mill’ that are borrowed from Latin molina; English mill and Irish muileann are “related” in that they are both borrowed from the same source, which is a very different sort of relationship from the kind I mean when I use the word.)

I hope that’s clear, but if it’s not, perhaps somebody else can do a better job of explaining it.

Profile
 
 
Posted: 22 March 2008 10:46 AM   [ Ignore ]   [ # 2 ]
Avatar
RankRankRankRank
Total Posts:  1369
Joined  2007-01-29

Thank you, lh.

Profile
 
 
Posted: 22 March 2008 11:43 AM   [ Ignore ]   [ # 3 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

Always a pleasure, Eliza!

*doffs cap*

Profile
 
 
Posted: 23 March 2008 05:06 AM   [ Ignore ]   [ # 4 ]
Avatar
RankRankRankRank
Total Posts:  1178
Joined  2007-02-14

The human family analogy for “related” would be, e.g., Joe and Moe were identical twin brothers who were given up by their biological mother and adopted by different families.  I think the normal non-linguistic interpretation of the word “related” would include referring to Joe’s and Moe’s children, as long as the genetic relationship between Joe and Moe had become common knowledge.  This is just another example of a technical definition of a word not having the same meaning as the colloquial definition.  Cf., “theory”.

Correct me if I’m wrong, lh.

Profile
 
 
Posted: 23 March 2008 12:37 PM   [ Ignore ]   [ # 5 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

Doesn’t work for me.  Borrowed words are just that, borrowed—there’s no genetic relationship at all.  This is hard to see in the case of mill and muileann because English and Irish (and Latin, from which they borrowed the word) are genetically related languages; it may be easier to see in the case of English imam and Fulani almaami.  Both languages have borrowed the word from Arabic, a language neither is related to.  Does it make sense to you to say that English imam and Fulani almaami are not related?  If not, how would you prefer to describe the distinction?

Profile
 
 
Posted: 23 March 2008 02:37 PM   [ Ignore ]   [ # 6 ]
Avatar
RankRankRankRank
Total Posts:  1178
Joined  2007-02-14

Are imam and almaami taken from the same Arabic word?

Profile
 
 
Posted: 23 March 2008 04:29 PM   [ Ignore ]   [ # 7 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

Yes, Arabic imam (Fulani borrowed it complete with definite article: al imam ‘the imam’).

Profile
 
 
Posted: 23 March 2008 08:10 PM   [ Ignore ]   [ # 8 ]
Avatar
RankRankRankRankRank
Total Posts:  2836
Joined  2007-01-31

I think that linguists would do well to develop a better technical vocabulary to describe all these situations without resorting to applying counterintuitive meanings to common words (like “related").  Very similar issues have cropped up in genetics, especially since the advent of whole genome sequencing.  If we view words as analogous to genes and languages to genomes, then “borrowing” is equivalent to what is called “horizontal” or “lateral” gene transfer: the transfer of a gene from one species to another, not a descendant of the source.  (This happens mostly in unicellular organisms, but can occur to a lesser extent in higher organisms, both naturally and artificially, e.g., the glow-in-the-dark cats).

So, the term “homolog” refers to any two genes deriving from some common ancestral DNA sequence.  As I see it, in linguistics this could apply both to pairs such as father/Vater and mill/muileann.

A subset of homologs is orthologs.  These are genes related to each other by the normal evolutionary process of speciation: for example, the horse insulin gene and the zebra insulin gene, both descended from the insulin gene of the last common ancestor of horses and zebras, and inherited by its descendants in the usual way, along with the rest of the genome.  Father/Vater would be orthologs but mill/muileann would not. In other words, orthologs would be cognates according to MW3 sense 2b.

Homologs related through lateral gene transfer (in the way that the fluorescent protein gene in the cats is related to that in the jellyfish the gene was originally isolated from) are called xenologs.  Applied linguistically, this term would cover homologs related by borrowing, such as mill/muileann: cognates according to MW3 sense 2c.  (Or, FTM, imam/almaami.)

There’s another subset, paralogs, describing homologs related through gene duplication that might be useful in describing pairs like bust/burst and passel/parcel, though these don’t bear on the recent brouhaha.

The adoption of such unambiguous and non-misleading terminology would be far too sensible to actually happen, of course.

(For the benefit of those who come across this thread in the archives months or years from now, all of this relates to issues that came up in the Lagavulin thread.)

[ Edited: 23 March 2008 08:31 PM by Dr. Techie ]
Profile
 
 
Posted: 24 March 2008 05:33 AM   [ Ignore ]   [ # 9 ]
RankRankRank
Total Posts:  236
Joined  2007-02-13
Dr. Techie - 23 March 2008 08:10 PM

I think that linguists would do well to develop a better technical vocabulary to describe all these situations without resorting to applying counterintuitive meanings to common words (like “related").

Of course if they do this, they will be criticized for using jargon.

Profile
 
 
Posted: 24 March 2008 08:21 AM   [ Ignore ]   [ # 10 ]
Avatar
RankRankRankRankRank
Total Posts:  2836
Joined  2007-01-31

True, but they’re not immune to that charge even now, and a certain amount of jargon is necessary when workers in a field need to make fine distinctions that the common language lacks the tools for (without cumbersome prolixity).

Another nice feature of the nomenclature I described is that the terms are fairly intuitive if you look at their radical meanings; i.e., if you infer the meaning from the etymology, it matches the intended sense pretty well.

homolog = “same word” or “like word”.
ortholog = “straight word” (connected by straight descent).
xenolog = “foreign word” or “guest word” or “strange word”.
paralog = “beside word” (a word that exists alongside another in the same language derived from the same etymon).

[ Edited: 24 March 2008 08:23 AM by Dr. Techie ]
Profile
 
 
Posted: 24 March 2008 06:39 PM   [ Ignore ]   [ # 11 ]
Avatar
RankRankRankRankRank
Total Posts:  3100
Joined  2007-02-26

"Mekong foozed.”

Surely imam and almaami are related AND cognate.

Profile
 
 
Posted: 25 March 2008 05:21 AM   [ Ignore ]   [ # 12 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

Surely imam and almaami are related AND cognate.

I’m giving up on the word “related” because it appears to be hopelessly confusing, but no, they’re not cognate.  “Cognate” refers to words descended from a common ancestor—descended in the natural course of language change, just to make that clear.  Hebrew shalom and Arabic salaam are cognate because they descend from the same Proto-Semitic form.  English salaam ("he made a deep salaam") is not cognate with either; it is a borrowing from Arabic, as are imam and almaami.

Profile
 
 
Posted: 25 March 2008 06:07 AM   [ Ignore ]   [ # 13 ]
Administrator
Avatar
RankRankRankRankRank
Total Posts:  4742
Joined  2007-01-03

“Cognate” refers to words descended from a common ancestor—descended in the natural course of language change, just to make that clear.

Where do you draw the line and declare that a borrowed word is old enough to be called inherited?

Take the word that started all this, mill. The Old English mylen was borrowed it from the Latin molina sometime before or during the 9th century. That word eventually became our modern mill. Is it proper to say this thousand-year-old word is inherited from Old English?

Or is the distinction one of jumping language groups, Latin/Romance to Germanic? In which case, how do you define language group? Can a West Germanic language inherit a North Germanic word? What about skirt/shirt? Are these inherited or borrowed? The English words come from the Old Norse, borrowed from the Danes when they invaded England, but “cognates” appear in other West Germanic languages, indicated a common Proto-Germanic root, or maybe that all the Germanic languages borrowed this one from the Norse.

If you go back far enough, all these distinctions become fuzzy, which was my original point about “inherited” being a catch-all, miscellaneous category meaning “so old that we don’t know where it comes from.” All inherited words were originally obtained by some other process.

Profile
 
 
Posted: 25 March 2008 11:05 AM   [ Ignore ]   [ # 14 ]
RankRankRankRankRank
Total Posts:  3498
Joined  2007-01-29

Dave, I don’t understand why you find this useful and universally accepted distinction so distasteful that you’re trying to muddy or derail it.

Where do you draw the line and declare that a borrowed word is old enough to be called inherited? Take the word that started all this, mill. The Old English mylen was borrowed it from the Latin molina sometime before or during the 9th century. That word eventually became our modern mill. Is it proper to say this thousand-year-old word is inherited from Old English?

Yes, modern English mill is descended from OE mylen, which is borrowed from Latin.  What’s so hard about that?

If you go back far enough, all these distinctions become fuzzy, which was my original point about “inherited” being a catch-all, miscellaneous category meaning “so old that we don’t know where it comes from.”

If you go back far enough, everything becomes fuzzy.  It can be hard to tell whether an ancient fossil is from an ape or a human.  That doesn’t mean there’s no difference between apes and humans, or that it’s pointless to try to fit modern mammals into one category or another.  Yes, it can sometimes be hard to tell whether a word is borrowed (way back when) or inherited.  So?  It can sometimes be hard to tell whether a person is a man or a woman; in isolation, the word attention could be either English or French; etc. etc.  So?

All inherited words were originally obtained by some other process.

This is an unprovable statement that smacks more of philosophy than science, and seems of dubious relevance to one of the basic tasks of historical linguistics, which is precisely to distinguish borrowings from inherited words.  You may as well say that since all elements were originally indistinguishable parts of the Big Bang soup, there’s no point doing chemistry.

Profile
 
 
Posted: 25 March 2008 01:07 PM   [ Ignore ]   [ # 15 ]
Avatar
RankRankRankRankRank
Total Posts:  2836
Joined  2007-01-31

FWIW, I think the problem is using “inherited” without a “from”.  Modern English mill is inherited from Old English (by LH’s own definition, above: “a changed form of the same word in an earlier form of the language"), but the Old English word was borrowed from Latin. 

So what does it mean when you say that a word is “inherited"--without a “from...”?  LH appears to be saying that this must be interpreted as “inherited as far back as can be traced, without borrowing"--under which interpretation, mill is not “inherited”.  OTOH, Dave seems to be interpreting it as “inherited from some ancestral form of the language"--by which interpretation mill certainly is “inherited”.

So, again, what does it mean to say that a word is simply “inherited”?  I’d say it means you’re missing an important prepositional phrase.

[ Edited: 25 March 2008 01:13 PM by Dr. Techie ]
Profile
 
 
   
1 of 2
1
 
‹‹ Cheese Food      She's the cat's mother ››