Back in 2007 I quoted a paragraph about the groundbreaking ideas of Ferdinand de Saussure and said it “reawakened in me the impact—not just intellectual but emotional (it was almost like falling in love)—of learning about all this as I was in the process of deciding linguistics was what I wanted to do, several decades ago”; I get the same feeling from reading Beyond Pure Reason: Ferdinand de Saussure’s Philosophy of Language and Its Early Romantic Antecedents
To provide a simplistic but not inaccurate caricature/summary of that history: for the first half-century after his death in 1913, “Saussure” meant two books, the Mémoire sur le système primitif des voyelles dans les langues indo-européenes (Memoir on the Primitive System of Vowels in Indo-European Languages), which appeared in December 1878 (though the cover date is 1879) when he had just turned twenty-one, and the Cours de linguistique générale (Course in General Linguistics), which he did not even write (it was compiled by his students from class notes) and which came out in 1916. The first put historical linguistics on a sounder theoretical footing than it had had and famously postulated the existence of a Proto-Indo-European laryngeal which actually turned up in Hittite (sadly, two years after his death); the second jump-started the entire field of synchronic/structural linguistics and was foundational for everything that came after—Jakobson, Bakhtin, and my very own pre-Chomsky American linguistic tradition. Then the tide turned; structuralism was out, poststructuralism was in, and the Cours was considered laughably outmoded. Just around that time, however, Saussure’s own notes began being excavated, thousands of pages of them, and people realized there was a whole different side of Saussure that was much more attractive to the postmodern mentality, notably his three-year obsession (which I, an unrepentant modernist and structuralist, consider utter nonsense) with what he variously called hypogram, leitwort, Stichwort, mot-thème, logogram, antigram, paragram, cryptogram, and anagram; the last is the term that has stuck, thanks to Jean Starobinski, who published some of the papers in 1964. The people who were excited by that sort of thing accused Saussure’s students of having grievously misrepresented their teacher’s ideas when putting together the Cours.
Gasparov deals with all this sensitively and sensibly. As is my wont, I’ll quote some chunks of the text to let you see his style and approach. From p. 46:
Virtually all the works Saussure published in his lifetime were dedicated to various problems of Indo-European linguistics.Yet even though he continued publishing some miscellaneous articles on the subject through the 1890s, he had essentially ceased to be an “Indo-European linguist” since the beginning of his inquiry into the nature of language. Looking now at the whole scope of Saussure’s oeuvre, including what he himself was resigned to let perish, we can say that Saussure eventually ceased to be a “linguist” altogether.
From p. 60:
We should reconcile ourselves to the fact that the definitive or “genuine” representation of Saussure’s ideas does not exist, nor has it ever existed. A plurality of representations, freely evolving in different directions, was constitutive of Saussure’s thought and essential for the conceptual vision he pursued. It corresponded to the central principle of his approach to language, which, in the end, has made any systemic and definitive picture impossible: the principle of the unfettered freedom of language, resulting in the infinite plurality of forms that language assumes and the infinite diversity of the directions in which they may evolve.
From p. 75, an admirable illustration of why apparent exceptions to the basic principle of arbitrariness do not contradict it:
Looking at words with clear derivational patterns, one can easily be lured to the conclusion that, even if one had no knowledge of the value assigned to them by convention, one would still be able to grasp it by inference. This conclusion, however, is illusory. Both the form and the meaning of housekeeper stem from house and keeper in a way that seems perfectly logical. Imagine, however, someone trying to infer the meaning of housekeeper from its ingredients, without knowing that meaning for a fact. Does housekeeper mean a person who guards the house, or owns it, or retains it temporarily, or occupies it as a squatter—or perhaps not a person but a device that prevents the house from collapsing? The only way to know that our educated guesses are wrong is to know the meaning of housekeeper as determined by convention—that is, to treat it as arbitrary.
On the late-eighteenth-century romantic tradition that Saussure seems to have drawn on (something of which I had no inkling), he first quotes a striking passage from Novalis (p. 103):
It is a ridiculous delusion, which causes one to marvel, to think that when one is speaking, one does so for the sake of things. Nobody is aware of the proper nature of language, namely, that it is concerned with nothing but itself. . . . If only one could make people understand that language is like a mathematical formula. Both constitute a world of their own—they play with nothing but themselves, express nothing but their own wonderful nature, and it is precisely because of this that they are so expressive. . . . They become a part of nature solely as a result of their freedom, and it is solely through their free movement that the universal soul expresses itself, making them a gentle scale and foundation of things.
He then goes on to say:
That seeing language as a tool through which one reaches the phenomena of the world is “a ridiculous delusion” (here, Novalis’s lächerlicher Irrtum anticipates Saussure’s favorite ridicule); that it is in fact a free interplay, responsible to and expressing nothing but itself; and that it is precisely the total freedom of language that connects it to nature, making it a “gentle scale and foundation” of things rather than a label attached to them—these points of the Monologue contain, in embryonic form, the quintessence of Saussure’s teaching about language: its immanent “emptiness” resulting in its unbounded freedom.
On pp. 115-116 there is a superb summary of his approach to Indo-European historical linguistics that is too long to quote; I will just give the last sentence, which I like very much: “Saussure’s book [the Mémoire] played a pivotal role in the development of a more cautious and abstract attitude in comparative studies, an attitude that has become prevalent in the twentieth century, when hunting for the Ursprache and its speakers gave way to the sober realization that all that could ever be achieved was to place the relations between kindred languages into a coherent model.” I won’t get into Saussure’s analysis of mythology, or his relationship with Schlegel’s ideas, or those “anagrams” (vaguely defined alleged hidden insertions of names into poetic texts, supposedly passed down for thousands of years as a poetic technique without leaving any explicit trace in the record); I hope I’ve given at least some idea of the riches to be found here. I will say, though, that if he hadn’t been so contemptuous of “mere facts” (he reproached the Junggrammatiker or Neogrammarians with whom he studied at Leipzig as being dusty fact-grubbers who had no clue how language worked) and so obsessed with the theoretical basis of everything (he famously asked “unde exoriar?”—”Where shall I begin?”), he might have wasted less time haring off on wild-goose-chases and more time using his genius for better things, and even written another book or two. I’ll close with another nice quote (from p. 86): “The truth about language lies not in ultimate generalizations but, on the contrary, in the unceasing comparative analysis of its diverse manifestations.”
Well, actually, I can’t close with that, because I have to complain about the terrible proofreading and editing of this important scholarly book. A few samples: “did not shaken the fundamental premises” (p. 43), “casts off the conventional constrains of sequentiality” (p. 47), “veritable” (for véritable) (p. 65), “bêtisses” (for bêtises) (p. 90), “emerged in 1790s” (p. 92), “toutes notre [should be nos] distinctions, toute notre terminologie, tout [should be toutes] nos façons de parler” (p. 109), “the principle vowels” (p. 113), “to name just a few” (after two names, p. 125) [n.b.: John Cowan thinks this is OK, and presumably he’s not alone], “TO HAVE A SYSTEM AND TO HAVE NONE IS [sic] EQUALLY DEADENING FOR THE SPIRIT” (heading, p. 177). A “which” is missing in the first line of p. 174, déception is translated as “deception” on p. 189 (a classic faux ami [should be “disappointment”]), every smooth breathing in the Greek on p. 143 is from the wrong font and wrongly placed, every occurrence of the title Cours de linguistique générale (except, oddly, for one on page 42) wrongly begins with “Course,” and of course (as I wrote here), “Lotharingia” is used throughout for “Lorraine.” Come on, Columbia University Press, you can do better than this.
Thanks for the suggestion/summary!
I thought the counterexample to arbitrariness wasn’t morpheme compounding but phonosemantics?
the quintessence of Saussure’s teaching about language: its immanent “emptiness” resulting in its unbounded freedom.
It is this that led to the divorce of language from its actual use and social functions, resulting in the influx of those with a bias towards abstract theory and, ultimately, Chomsky.
Mmm, I would have put the “[sic]” after AND rather than after IS. And what’s wrong with a few meaning ‘two’?
resulting in the influx of those with a bias towards abstract theory and, ultimately, Chomsky.
“… et après le déluge, Chomsky”, eh? Now that’s chilling.
The business about compounding is a big pitfall in the translation of Chinese, because while the majority of Chinese syllables have definite meanings of their own, the meanings of some compounds have no relationship at all to the meanings of the component syllables, and but in other cases a lot of relationship, but even in the latter case a compound usually has a more restricted meaning than the combined meaning of the two words, and also there are compounds without precedent use where you have to just guess. Translating Chinese from the individual characters often leads to disaster.
My example of arbitrary combination in English is “sidewalk”, which might mean catwalk, or side trip, or byway, or crabwalk, but just means sidewalk.
…great post, very interesting. Side question on European titles: why “Saussure” instead of “de Saussure”? Is this sort of thing just completely arbitrary? We have, for example, actor Willem Dafoe, or author Daniel Defoe, and no one calls either of them “Mr. Foe”. We’ve got Beethoven instead of van Beethoven, but “the Family Von Trapp” in the Sound of Music. There’s de Kooning, there’s De La Warr, there’s all sorts of conflicting examples. Um… lost my point here… guess I’m asking the juvenile but perhaps interesting question: Why Saussure but not, say, “Toro” as shorthand for actor Benicio del Toro?
And what’s wrong with a few meaning ‘two’?
Huh, it never occurred to me that my strong sense that “a few” is necessarily more than two wasn’t universal. I’ve added a note to the post mentioning your reservations on that point.
On looking it up in M-W to see if the dictionary had anything specific to say on the issue (it didn’t), I found this enticing word:
fewtrils [origin unknown] dial Eng : things of little value : trifles
Swedish fjäril, No. dial. fivril “butterfly”.
Bathrobe: Roy Harris, non-mainstream anti-abstract-linguistics guy, agrees with you. He considers Chomsky essentially footnotes to Saussure, and devotes most of his energy to discussing problems with Saussure instead.
Emerson: Just the other day I had 流行 in an exam—easy kanji, “flow-going”, so I ballparked “flowing”, but turns out it means “fad, fashion”… In time one learns to be skeptical of “kanji meanings”. It’s like prefixes and such in IE languages: knowing the morphemes help, but at the end of the day there’s no going around word-level vocabulary acquisition.
My fav Portuguese example: “parapeito” = breaststop = parapet, window ledge. Transparent, but even then, knowing the morphemes still won’t give you the compound meaning. What’s more, most native speakers don’t even realize it’s “pára-peito” until you point it.
Yesterday I found the following Icelandic headline: „Lady Gaga hrærð á friðarverðlaunaafhendingu“ . The word friðarverðlaunaafhendingu amused me so I went looking: the dictionary had “at random” for af hendingu, so I got “peace-prize-reward-randomly”. I asked Icelanders and turns out “afhending” is compounding something like “offhandedly” (henda = “throw, catch”), which can mean not only “randomly” but also “delivery”, so it was: peaceprizerewarddelivery.
I agree with leoboiko’s first comment: ‘housekeeper’ and other compounds aren’t even apparent exceptions to the principle of arbitrariness. ‘Housekeeper’ has a range of possible meanings because (a) the semantics of English noun-noun compounds is loose, and (b) it inherits the polysemy of its components ‘house’, ‘keep’, and ‘-er’. But the form-meaning mapping of the whole is not arbitrary, only underdetermined. It’s the form-meaning mappings of primitive signs that are (mostly) arbitrary, not those of rule-governed sign combinations.
I may be opening a can of worms here, but am I the only one here who thinks that “arbitrary” and “conventional” have different meanings, and that a social fact shouldn’t be called arbitrary as if that were the only alternative to considering it “natural”? The idea that the relation between words and things is “arbitrary” has caused a great deal of mischief since the 1970s. On this see e.g. Raymond Tallis’ Not Saussure.
Why “Saussure” and not “de Saussure”?
French de is a preposition usually meaning ‘of’ or ‘from’, and it was originally used in proper names before the name of a location, especially the landed property of a noble family. (Similarly in other languages as in Da Vinci or van Beethoven).
In France the general rule is that when there is no first name or title mentioned (as in Ferdinand de Saussure, Guy de Maupassant or le duc d’Orléans), de is omitted if the following word has more than one syllable, thus Saussure, Maupassant, Tocqueville, La Fontaine or La Rochefoucauld, but De Gaulle (since Gaulle is spoken as one syllable). There is at least one exception: the Marquis de Sade is usually referred to as Sade, although 19th century sources (like Flaubert) use de Sade.
I think that the same thing occurs with German von or Dutch van, eg Beethoven but Van Dyck or Van Gogh.
When de is included in a contracted form, as in du instead of de le, the whole contracted form is used, as in Du Maurier not Le Maurier. Similarly in Italian, as in Del Toro or Della Porta, not El Toro or La Porta.
Once transplanted to an English-speaking context, such names tended to be treated as units, as in the name of Lord Delaware (originally the seigneur de La Warr), but de and la were not understood, and thus became part of the name. In the hybrid name Delahunt, only the main word was translated (from de la chasse, ‘(lord) of the hunt’).
The initial syllable in DaFoe or DeArmond (which are written as one word) is not the French or Italian preposition but a Germanic article (related to the), usually found in names of Dutch origin.
During the 19th century (and possibly earlier) men would sneak a “de” into the front of their name to gain an air of nobility, Honore de Balzac being a prime example.
Having an actual last name in Scandinavia was a sign of respectability. Otherwise you were just Johnson or Olson after your father’s first name. I’ve often wondered whether the density of -sons is the same in Scandinavia as among Scandinavian-Americans.
The father of Charles Lindbergh the aviator was born Carl Månsson but changed his named to Lindbergh. Following Scandinavian custom the aviator would have been Charles Carlson. If they had switched to American naming practices a generation later, the aviator could have been named Charles Manson.
Language Hat: I found this enticing word:
fewtrils [origin unknown] dial Eng : things of little value : trifles
Me: Swedish fjäril, No. dial. fivril “butterfly”.
Or from French futile?
“Glycyrrhiza”: how many of you immediately recognize this word and its close English cognate. I expect the number to be high, but we’ll see.
I wanted to say: Off topic.
Not completely off topic: The glycyr + rhiza compound could be lots of things.
I want to comment on the reviewer’s last paragraph criticizing the editing and proofreading of Gasparov’s book. I am the managing editor at Columbia University Press. We publish dozens upon dozens of theory-heavy books with many foreign words and phrases. The editor of this book knows Spanish, Portuguese, and some French. Some of our books contain many foreign languages (including Chinese, Japanese, and Sanskrit). Would you blame the editor if some of these words were wrongly spelled? No editor knows all the languages that often appear in scholarly books. Ultimately, what appears in the book is the responsibility of the author, not the editor. Because we are a nonprofit organization, we always depend on the author to review not only the edited manuscript but the proofs. Books always contain a few errors. No book I’ve read produced by any publisher is perfect.
JC: During the 19th century (and possibly earlier) men would sneak a “de” into the front of their name to gain an air of nobility,
The custom is much older than that: I quoted Lord Delaware, whose title was first held by Roger la Warr, 1st Baron De La Warr, starting in 1299 (Wikipedia scripsit). (I have no idea what La Warr referred to).
Adding a de to one’s name is more difficult since the collection of vital statistics has become very efficient, but it was practiced for centuries. Especially, in the 17th and 18th centuries, many nouveaux riches who bought a rural property would try to have themselves declared nobles so that they could add de X to their names (X being the name of the property). But the addition gave them away: older noble families had no “family name”, only de X.
The Du Pont de Nemours family was not among them: they were plain du Pont (‘of the bridge’, a very common name,usually written Dupont). When Pierre Samuel du Pont became president of the Constituant Assembly (during the Revolution) he (like some other delegates) added the name of his city to his own name, though not officially (Wikipedia again). (He later took his family to the US, where they still are).
glycyrrhiza : I think I have always seen the Greek name written as glycorrhiza.
I know what that is. Hint: the g got lost.
M-L,
“Once transplanted to an English-speaking context, such names tended to be treated as units, as in the name of Lord Delaware (originally the seigneur de La Warr), but de and la were not understood, and thus became part of the name. In the hybrid name Delahunt, only the main word was translated (from de la chasse, ‘(lord) of the hunt’).’
In the same way as all those French loanwards scattered across American langauges that all seem to start with “li-“.
Managing editor: Because we are a nonprofit organization, we always depend on the author to review not only the edited manuscript but the proofs.
You plead poverty to your authors? Wow. If Columbia University Press is non profit it’s presumably for tax reasons, because if any New York institution is rolling in it, it’s Columbia. The only richer US university is Harvard. Columbia owns half of the Upper-West side, as well as the Rockefeller Center site and a lot of valuable real estate downtown around Wall Street; in fact it’s the third-richest landlord in NY (after the City itself and the Roman-Catholic church).
Ultimately, what appears in the book is the responsibility of the author, not the editor.
That seems to be the current view, not only at CUP but in general. I think it’s a terrible one. An author’s business is writing, not editing; as an editor myself, I know how terrible most people are at it. It should very much be the publisher’s responsibility to put out as clean a product as possible.
No book I’ve read produced by any publisher is perfect.
Perfect, no. But books by top publishers used to be a lot better than this. When I worked as a galley slave in an editorial department in Midtown, I carried around a xerox of misprints I’d found in books that had presumably been proofread to within an inch of their lives (Oxford Classical Texts, dictionaries, and the like) to prove the point about perfection to my bosses (who tended to fly off the handle when their crack proofreading staff let anything slip by), but I had to look hard for them. What I reproduced in my last paragraph isn’t imperfection, it’s sloppiness, and CUP should (I repeat) do better. I think putting it all onto the author is a shocking abdication of responsibility.
@Jim: And Portuguese has dozens of words from Arabic starting with al-.
Some of our books contain many foreign languages (including Chinese, Japanese, and Sanskrit). Would you blame the editor if some of these words were wrongly spelled? No editor knows all the languages that often appear in scholarly books.
I think you will find that in such cases most readers will find fault (albeit not necessarily with the editor, who may be working under non-ideal conditions imposed by the publisher). If I’m going to pay $40 or more for a book from an academic press (and I do, frequently), I don’t think it’s unfair of me to expect it to have actually been edited, even the obscure parts for which consulting a third party with the necessary expertise may have been necessary.
Overall, the book sounds great; I’m glad that CUP published it and I’m planning to buy it. But you can count me as another actual customer (I am looking at another hardcover CUP book on my desk right now!) who feels that academic publishers can and should do better in terms of quality, and that it’s extremely unprofessional to abdicate this responsibility and blame the author.
If I’m going to pay $40 or more for a book from an academic press (and I do, frequently), I don’t think it’s unfair of me to expect it to have actually been edited, even the obscure parts for which consulting a third party with the necessary expertise may have been necessary.
Hear, hear!
This is all unfair. I think there must have been a mixup between Columbia University Press and California Vanity Press. Could have happened anyone who doesn’t do proofreading.
Be fair, guys: Oxford University Press and Cambridge University press set the industry standard, and they’ve decided that editing is not important. It would be presumptious for Columbia University Press to do better.
@marie-lucie: In the hybrid name Delahunt, only the main word was translated (from de la chasse, ‘(lord) of the hunt’).
I always understood Delahunt to be a variant of Delahanty, which is apparently from O Dhulchaointigh, which apparently means “plaintive satirist“. I find that last bit hard to believe.
Contra AJPC, on most standard listings Columbia is merely something like the eighth-richest university in the U.S., although no doubt there can be methodological quibbles with any such ranking. But more importantly, the wealth etc. of a given university press is not necessarily directly proportionate to that of its host institution. For example, I was told in a conversation a few years back with an editor at another prominent US university press that Princeton’s press was unusually lavishly-resourced (by inference due to some sort of funds in the university’s endowment earmarked by the donors for the press) compared to the presses of peer institutions.
Garrigus: I always understood Delahunt to be a variant of Delahanty, which is apparently from O Dhulchaointigh,
I had never run into this explanation, but it sounds very plausible. In this case, the change from hanty to hunt could have been a reformation influenced by other Dela- names of French origin: the non-English hanty would have been interpreted as meaning hunt and the whole word as a “half-translation” of Delachasse.
I don’t know this word as a French name, but it does not sound implausible at all. A quick google search reveals a few places called La Chasse in Northwestern France, and in the early history of Canada there is a Pierre de La Chasse born in France.
It is probable that la chasse here does not refer directly to the hunt as an activity but to a hunting preserve. Such places in France (eg a private wood) are identified by signs saying CHASSE GARDÉE, meaning that un garde-chasse (gamekeeper) is employed (usually part-time) to keep an eye on the property and make sure no unauthorized person is hunting or poaching. This meaning of la chasse is also found in the name de Bellechasse which was that of a noble family.
the wealth etc. of a given university press is not necessarily directly proportionate to that of its host institution.
Just because it isn’t doesn’t mean it oughtn’t to be at least adequate at a place like Columbia, and there’s no reason why the authors should bear the difference (and even if you think Columbia’s currently only 8th richest I’m pretty sure it doesn’t make any difference to my argument).
Universities are among the oldest corporations: indeed, universitas actually means ‘corporation’ in Roman law. And as Baron Thurlow didn’t quite say, why should you expect a corporation to have a conscience, when it has neither a soul to be damned nor a body to be kicked?
I personally have no problem printing foreign words without their native diacritical marks in quotations in an English-language book, as long as it’s done consistently as a matter of policy. Because otherwise you have to pay extra for a proofreader who can accurately/consistently spot “typos” involving accent marks and smooth-breathing indicators. I’m not sure which of the nits LH lists above are “editing” failures rather than “proofreading” failures, assuming he has some sort of division of labor or skill-set in mind? Finally, I’m in favor at an abstract level of someone other than the author double-checking the accuracy of everything in quotation marks by comparison to a copy the original source quoted (not least b/c it always helps to have a second set of eyes because you can’t see your own obvious screwups after you’ve gone over the draft enough times), but I don’t know how customary/feasible/cost-effective that is or realistically could be in the world of scholarly publishing.
I personally have no problem printing foreign words without their native diacritical marks in quotations in an English-language book, as long as it’s done consistently as a matter of policy.
In the first place, that would be horrible; it would be better not to have foreign words at all. In the second place, it’s completely unnecessary in the age of Unicode. And in the third place, that’s not what’s going on here. They have accents, but sometimes they have the wrong ones.
Finally, I’m in favor at an abstract level of someone other than the author double-checking the accuracy of everything in quotation marks by comparison to a copy the original source quoted …, but I don’t know how customary/feasible/cost-effective that is or realistically could be in the world of scholarly publishing.
I do it routinely for the UP I usually edit for.
Sure, Hat, but you are notoriously an above-and-beyonder when it comes to your job.
notably his three-year obsession (which I, an unrepentant modernist and structuralist, consider utter nonsense) with what he variously called hypogram, leitwort, Stichwort, mot-thème, logogram, antigram, paragram, cryptogram, and anagram
Apparently he also called it anaphony; see Joshua T. Katz, “Saussure’s anaphonie: sounds asunder,” from which Laudator Temporis Acti quotes the following eloquent passage:
Go to the link for a diagram showing the loony-tunes encoding at work (and for the full Katz citation).
From the OP:
The Bierce-Hartman-McKean-Skitt law has compelled you to drop an n of indo-européenne.
So it has! I’ll leave it to show the action of the law.
Hartman sat across from me at Google.
Hah! The eventual convergence of all threads:
Hartmann painted the pictures from which ‘Pictures at an Exhibition’ was inspired.