Borges on the Fineness of English.

Jack Morava sent me a clip of Jorge Luis Borges talking about the English language; it was on Instagram, and looking around for a more accessible place to send people, I found Jordan M. Poss’s blog post, which has not only the clip (it turns out to be from a 1977 interview with William F Buckley Jr on “Firing Line”) but a partial transcript and Poss’s thoughts on it all. The transcript starts:

Borges: I have done most of my reading in English. I find English a far finer language than Spanish.

William F Buckley: Why?

Borges: Well, many reasons. Firstly, English is both a Germanic and a Latin language. Those two registers—for any idea you take, you have two words. Those words will not mean exactly the same. For example if I say “regal” that is not exactly the same thing as saying “kingly.” Or if I say “fraternal” that is not the same as saying “brotherly.” Or “dark” and “obscure.” Those words are different. It would make all the difference—speaking for example—the Holy Spirit, it would make all the difference in the world in a poem if I wrote about the Holy Spirit or I wrote the Holy Ghost, since “ghost” is a fine, dark Saxon word, but “spirit” is a light Latin word. Then there is another reason. The reason is that I think that, of all languages, English is the most physical of all languages.

The whole thing is transcribed, if in a rebarbative format, here. Of course it’s easy to poke holes in his linguistic analysis, but he’s not a linguist, he’s a writer talking enthusiastically about a language he loves, and I enjoy it — not to mention that it’s great to hear him speak. (I met him back in 1969, but my memory is not like Funes’s.) Thanks, Jack!

Comments

  1. A young English learner (I think East Asian (I include China)) once asked if he can systematically use phrasal vebrs instead of latinate verbs (or latinate verbs instead of phrasal verbs, I’m not sure).
    His example was “arrive” and “get to” – but the idea is clear.
    Present ESL teachers answered affirmatively.

    That was the first time when I seriously thought about the duality of English.

  2. David Eddyshaw says

    This was the basic dodge used in Ogden’s Basic English to get the core vocabulary down to 850 lexemes:

    https://en.wikipedia.org/wiki/Basic_English

    Ogden evidently understood little about syntax, and in particular did not know that phrasal verbs are one of the most notoriously difficult features of English for foreign learners. Basically, he was cheating.

    The WP includes a specimen from Ogden:

    What the World needs most is about 1,000 more dead languages—and one more alive.

    from which we can conclude that he was a Horrible Counter-Hat.

    The underlying idea is interesting, though: that in an invented auxiliary language it would be better to go for an isolating typology than Esperantoid agglutination. He just didn’t realise that this doesn’t get rid of the complexity, but just shifts it somewhere else. Like sweeping the dust under the carpet.

  3. Actually, the label “phrasal verb” was unknown to Russian learners in my time and as I remember from discussions on the same site, it is new to English speakers as well. What Russian learners in my time knew is that when you look up “get” in a dictionary, you see an infinitely long article, where “get” combined with numerous other words means all sort of things. But then you can make an observation that Russian prefixes are the same. Then the problem is not ‘get’ as such, but also presentation of material.

    There are obviously two groups of learners: some would rather use latinate verbs and avoid phrasal verbs, other would avoid latinate verbs.

  4. PlasticPaddy says

    The argument about simplicity does not (or should not) refer to complexity but to opacity. The word “depend” is opaque in English, but transparent to a Romance speaker. A more transparent “hang from” or “hang off” is not available. In cases where a transparent word is available, but the “opaque” word has acquired a meaning or association not assigned to the transparent word, the natural choice is the “opaque” word.

  5. David Marjanović says

    He just didn’t realise that this doesn’t get rid of the complexity, but just shifts it somewhere else. Like sweeping the dust under the carpet.

    “some of the complexity has moved from the noun morphology to somewhere else, like a lump in the carpet.”

  6. What’s wrong with agglutination?

    Just insert spaces between suffixes!

    (Or do you really want to discuss English phonetics and prosody?:-))

  7. Rodger C says

    Speaking of Basic English, the old Encyclopaedia Britannica article on it includes this sample sentence: “You can get all these words on the back of a bit of notepaper because there are no ‘verbs’ in Basic English.” This statement has mystified me for 63 years.

  8. Paul Frank says

    Borges was a great writer. But it was not just his knowledge of linguistics that was lacking. Borges says in this interview, “I know very little about politics, but I think we now have a government of gentlemen, not of hoodlums. I don’t think we are ripe for democracy as yet — maybe in a hundred years. But now I think we have the right government. I think that the government means well, and the government is acting, and as I said, we are governed by gentlemen, and not by the scum of the earth…” This at a time, February 1977, when those gentlemen’s minions were tossing dissidents into the Atlantic Ocean from Argentine Air Force aircraft.

  9. Give him a break, he was in his late seventies and blind and had other things on his mind than politics. I’m sure it would have been admirable if he’d cursed the junta and suffered the consequences, but we can’t all be heroes.

  10. J.W. Brewer says

    This vintage 1977 piece on Borges and his late-in-life political opinions may be illuminating: he had for decades hated the brutality and illiberalism of the Peron regime and was thus (late in life) naturally enthusiastic about whatever patriotic gentlemen had recently liberated Argentina from Peronism, with perhaps insufficient attention to their own possible defects, or indeed to their own parallels to the Peron regime in the areas of brutality and illiberalism. https://www.bostonreview.net/articles/kovacs-borges-on-the-right/

  11. J.W. Brewer says

    One of the early quasi-conlangs was Peano’s v 1.0 of Interlingua, a/k/a Latino sine flexione – which just took an inflected language already widely known to the educated classes throughout most of Europe and a few other parts of the world and took away the inflections. Obviously it must have done something else with its syntax to do the work the inflections had originally done, but I don’t know the details, or how intuitive this new non-inflected syntax was to speakers of major modern European languages who already knew the Latin lexicon.

  12. Stu Clayton says

    One of the early quasi-conlangs was Peano’s v 1.0 of Interlingua, a/k/a Latino sine flexione

    Hah ! In this 1901 French edition of the Formulario Matematico the notation immediately put me in mind of that used in the infamous Principia Mathematica.

    Sure enough, as this quote from PM shows, Russell and Whitehead took the notation from Peano.

    Now I know who to blame. The Latin is sine flexione, the notation is unbendingly rebarbative (think thick beard impregnated with hairspray).

    Overall, of course, Peano was cute as a bug. It’s not his fault that he could not anticipate causing me inconvenience.

    #
    “The notation adopted in the present work is based upon that of Peano, and the following explanations are to some extent modeled on those which he prefixes to his Formulario Mathematico [i.e., Peano 1889]. His use of dots as brackets is adopted, and so are many of his symbols” (PM 1927:4).[13]
    #

  13. David Eddyshaw says

    Peano actually presented Latino sine flexione in a bravura lecture/tract where he starts out in Proper Latin (like what grandpa spoke) and drops each morphological feature as he mentions it, finally ending in LSF. I found a copy of it once, but can’t place it at present.

  14. jack morava says

    I have long wondered what rebarbative might mean, am glad that now I know.

  15. Stu Clayton says

    I found a copy of it once, but can’t place it at present.

    Here you are for only Eur 7,46 from Amazon.

  16. David Eddyshaw says

    But that would involve paying for it (he whined.)

    I’m sure drasvi can do something to help us (ahem …)

  17. You can ask drasvi or you can ask Gutenberg (was he rebarbative?)

  18. Romanian also has many synonyms of Latinate and Slavic origin, many of which are discussed in Probleme de sinonimie, by Onufrie Vinţeler (Bucureşti: Editură Sţiinţifică şi Encliclopedică, 1983). mâncare ~ hrană ‘food’; nea ~ zăpadă ‘snow’; timp ~ vreme ‘time’; faţă ~ obraz ‘face’; popor ~ norod ‘people’; ştire ~ veste ‘news’; corp ~ trup ‘body’; etc.

    There are also inherited Latinate vs. borrowed French pairs such as călătorie ~ voiaj ‘trip’; a pregăti ~ a prepara ‘to prepare’; putere/tărie ~ forţă ‘power, force’.

  19. The plural of zăpadă (from Slavic *zapadati ‘to fall’) is zăpezi. I love Romanian.

  20. January First-of-May says

    The underlying idea is interesting, though: that in an invented auxiliary language it would be better to go for an isolating typology than Esperantoid agglutination. He just didn’t realise that this doesn’t get rid of the complexity, but just shifts it somewhere else.

    Classical Chinese is the limit of that direction: extremely isolating, but only more complex due to that.

    Admittedly, I don’t think it was actively trying to limit the number of different characters (more the opposite), but TBH I suspect if it was it wouldn’t have made it less complex (they’d probably have came up with some kind of equivalent of kennings instead).

    [Were there actually any pre-20C Classical Chinese works that focused on limiting the repertoire to just several hundred characters? I can’t think of any but if there were I probably wouldn’t have known anyway. I know there are some Japanese jokes involving multiple character readings, but that’s a different thing.]

  21. cuchuflete says

    Give him a break, he was in his late seventies and blind and had other things on his mind than politics. I’m sure it would have been admirable if he’d cursed the junta and suffered the consequences, but we can’t all be heroes.

    I was working in Buenos Aires in 1977. At the home of a staff member of my consulting client, whispered conversations alerted me to the “desaparecidos” and other horrors of the dictatorship. Everyone knew there was a military dictatorship, but not everyone was aware of the atrocities of the regime.

    My boss simply ordered me not to discuss anything political. Naïve young fellow that I was, I obeyed. Was invited to share the box of a cabinet member at the Teatro Colón for a symphony concert. We became close acquaintances, if not friends. I later learned that he, too, was unaware of most of the vile behavior of the militares.

    Judgment is too easy from decades away.

  22. Indeed.

  23. J.W. Brewer says

    Of course, that particular Argentinian regime (with periodic turnover in junta members) did not fall because of its regrettable domestic human-rights record. It fell because it picked a war with the U.K. and then lost it, which was apparently the opposite, legitimacywise, of hosting the World Cup and winning it, as it had done a few years earlier. Once the regime had fallen, the taboo about talking publicly about its negative points (and the accompanying plausible deniability as to whether a given Argentine had really known about them) then melted away.

  24. the notation is unbendingly rebarbative (think thick beard impregnated with hairspray).

    The bestselling book Beard on Bread had an unintentionally revolting title. It was followed by the bestselling and even worse Beard on Pasta.

  25. Joel: there’s закуска, which in Standard Romanian means what is called лютеница in Bulgarian and аjвар in Serbian. A few years ago I was hiking with my usual hiking buddies and there were some Romanian hikers at the same day, and to mess with them I used the word закуска within hearing range of them in the mess hall of the hut, and, judging by their reaction, they thought I was talking about лютеница, instead of breakfast.

    TLDR: закуска means breakfast in Bulgarian and in Romanian a vegetable mish-mash.

  26. Chris Stokes says

    If Mr Beard had only read the OED’s stab at the derivation of ‘rebarbative’, he might have been inspired to add a title on beards to his series of bestsellers:
    ‘Middle French rebarber to oppose, stand up to (13th cent. in Old French; < re- re- prefix + barbe barb n.1, hence probably literally [*]‘to stand beard to beard against'[*])'

  27. @DE,

    when I first came across someone selling an article from that time, it was Springer and an article immediately adjacent to Peano.
    It was Burali-Forte in Rendiconti del Circolo Matematico di Palermo. And of course it is not mere 7.46 (do you see that there are no 99s in this figure?). It was… is “39,95 € Price includes VAT (Russian Federation)”.

  28. Jonathan D says

    The word “depend” is opaque in English, but transparent to a Romance speaker. A more transparent “hang from” or “hang off” is not available.

    “Hang on” seems available for some people. I feel like it’s popular in legalese.

  29. Hang on ‘depend on’ also occurs outside legalese: “Everything hangs on my parents’ consent to let me go,” and so on.

  30. закуска means breakfast in Bulgarian and in Romanian a vegetable mish-mash
    And in Russian, it means “snack”, also specifically a snack to accompany drinks.

  31. Romanian zacuscă is also defined as “Preparat culinar care se servește ca aperitiv” ; “Preparat culinar făcut din vinete, gogoșari, ceapă [eggplant, pepper, onion] etc. ori din pește [fish] cu garnitură de legume.”

  32. jack morava says

    My beautiful linguist wife says that English is exceptional not in that it has phrasal verbs but that the system is productive and allows you to compound them on the fly.

    [? Being possibly opaque suggests their use as semantic markers requiring cultural fluency as part of the signal]

  33. David Eddyshaw says

    Were there actually any pre-20C Classical Chinese works that focused on limiting the repertoire to just several hundred characters?

    Well, there is this

    https://en.wikipedia.org/wiki/Thousand_Character_Classic

    but I don’t think that’s quite what you had in mind.

  34. Stu Clayton says

    … it was Springer and an article immediately adjacent to Peano. It was Burali-Forte in Rendiconti del Circolo Matematico di Palermo.

    Burali-Forti.

    “Adjacent” in more than one sense:

    #
    Er arbeitete von 1895 bis 1908 maßgeblich an dem von Peano herausgegebenen fünfbändigen Werk Formulaire de mathématiques mit, welches sehr bedeutend für die Weiterentwicklung der formalen Logik war. Die Ideen Peanos zur mathematischen Logik wurden von ihm in allgemeinverständlicher Form in dem Buch Logica mathematica beschrieben, welches 1894 erschien.
    #

    I quote the German Wipe article on Burali-Forti because the English one is exiguous to the point of vanishing.

    Tip: do not imitate that “…, welches …” introducing a subordinate clause. Unless you want to convey the impression of having diligently memorized rules in grade school. Tradesmen and waiters express themselves that way in writing. The paragraph quoted above was composed by a shop girl in her spare time.

  35. David Marjanović says

    Tradesmen and waiters express themselves that way in writing. The paragraph quoted above was composed by a shop girl in her spare time.

    In short, it’s an archaism preserved only by aging prescriptivists and the Swiss – in writing only, not in speaking; nobody has talked like that in centuries anywhere.

  36. J.W. Brewer says

    Somewhat differently and more modernly than the thousand characters evoked by David E., the kanji are taught in Japanese elementary schools in standardized-by-grade batches specified on a nationwide basis by the Ministry of Education, so that by the end of first grade, students are expected to know a specific list of 80, by the end of second grade, a specific (cumulative) list of 240, and by the end of sixth grade (after various intermediate steps), a specific (cumulative) list of 1,026.* I expect that there are books written for elementary-school-age children that deliberately use only kanji from the relevant list for the grade level of the intended audience, although I don’t know to what extent anyone has tried to write texts for a generic/grown-up-but-not-necessarily-all-that-well-educated readership using the same constraint. Of course Japanese offers kana as a way of filling in some gaps … See, e.g., https://en.wikipedia.org/wiki/Ky%C5%8Diku_kanji

    Apparently some official Japanese government documents are supposed to be drafted so as not to use any kanji outside a longer list (currently 2,136) that is believed to reflect the ones that any high-school graduate ought to be expected to know. I have no sense to what extent any private-sector publications or publishers may subject themselves to that same constraint.

    EDITED TO ADD: I don’t know to what extent there are common compound words (whether akin to English phrasal nouns or simply akin to more generic English compounds) where they are constructed from kanji on one of these specified lists but have idiomatic/non-compositional meanings that are not really predictable or guessable if you’ve just learned all the items of the list in isolation.

    *The details have shifted and mostly expanded over the decades – when I were a lad living in Japan back in the day the 6th-grade cumulative number was still only 881, although it apparently got bumped up to 996 in ’77, a year after my family had moved back to the States.

  37. Stu Clayton says

    it’s an archaism preserved only by aging prescriptivists and the Swiss – in writing only, not in speaking

    And by some German school teachers still, apparently. Where else would normal people get the idea they should write like that ? I think I saw an instance recently in a short Spiegel article online.

    I doubt there are many aging Swiss prescriptivists in the German education system. Subordinative “welcher/welche/welches” has passed into das kollektive geschriebene Unbewußte, too late now to eradicate it.

  38. @JWB, obviously a child can learn many more than 80 signs a year…

    Which reminds me my complaint at certian books for children. What I liked as a child is when there is a lot of new stuff on every page, and I can understand from context some things, other things remain obscure and a couple of them (either understood from context or obscure) i remember.
    But there were also boring books whose authors tried hard to introduce one new thing per a page. it was… Boring. But apparently they wanted me to remember this one new thing with the likelyhood 1 and were not at all comforted by that I learn many more random things from normal books.

  39. @jack morava, I suspect that the English system is under construction.
    That is, not only it is productive but English is in the process of developing a new system.

    Otherwsie, the closest thing Russian and Latin have (prefixing: de-pend, ar-rive*) are productive too.

    ___
    *Wiktionary apart of ġelandian, ġelendan mentions “Displaced native oncome, tocome.”
    These two themselves look inspired by Romance.

  40. inspired by Romance

    I don’t watch that kind of movie.

  41. @Stu

    this was a meme on the Russian Internet a decade ago.
    It is an excerpt from a textbook on analytical mechanics (in Russian theoretical mechanics, or in students’ jargon teorméx).

    The author says that just as axioms of arythmeic are not a part of arythmetics, so axioms of mechanics would be logic rather than meachnics and so he won’t give any formal system and will realy upon the reader’s intuition. To support this he quotes Poincaré:

    Nous voyons d’abord M. Burali-Forti définir le nombre 1 de la manière suivante :
    [the formula]
    définition éminemment propre à donner une idée du nombre 1 aux personnes qui n’en auraient jamais entendu parler.

    People found it amusing, and I wanted to see the original article by Burali-Forte.

    (some one on reddit says that Bourbaki’s definition is much “clearer and more logical” and links this article)

  42. Y, one can always comfort herself with that shés a character, not a viewer or God forbid critic. If she isn’t a critic…

  43. thousand characters

    Horrible Western writing systems (so called alphabetic writing, abjads and their likes) are the primary cause of the deplorable literacy levels in the Western world, especially on African and Indian (sub-)continents and in Arabia.

    The obvious mistake is that one’s ability to do what linguists do – to analyse phonology – is the prerequisite for learning to read. They try to teach children and women* do this, no wonder that many fail.

    *”women and children” is idiomatic in the sense “victims par excellence”. Also I heard female literacy is a particular issue…

  44. (in case it is not obvious, i’m mocking arguments for advantage of one system over another when those are not based on research… which is not an easy matter, because your success strongly depends on how you teach)

  45. and here i thought “rebarbative” was about architectural concrete…

  46. jack morava says

    reminiscences of grandma’s hardware store…

    https://www.youtube.com/watch?v=MHLJDkug-8M

  47. John Cowan says

    I don’t think that’s quite what you had in mind.

    There is also General Chinese (this link is to Chao’s monograph, not WP), which provides a system of 2085 characters that are intended to cover all of Chinese in this sense: 80% of them are mappable one-to-one to morphemes of Chinese; the others are mappable many-to-one in a way that would make them homophones in an abstract variety of Chinese that made all the phonetic distinctions of Mandarin, Yue, Wu, MC, and to some extent Min.

    Here are Chao’s explanations. (Note: the characters are mis-OCRed. I’ve left them alone, as you can see them on PDF pp. 5 and 7 of the link. The romanizations are Chao’s and correspond one-to-one to the 2085 characters. The rules for reading these are on pp. 81, 83, 85, 87, 89; fixing and HTMLizing them is too much work for me.)

    1.2 Vocabulary 一 In the (accompanying) Technical Report (Part Two), a draft syllabary of 2085 syliables will be proposed, of which about 80% have no homophones, so that kai means ‘open’, (并), men means “door (门),cut means ‘bone”( 骨 ), lip means “to stand’ ( 立 ), lit means “chestnut’ ( 村 ), lic means ‘strength’ (力 ), sam means ‘three’ (三 ), etc. These syllables then are morphemes, or words with definite meanings, or clusters of meanings related by extensions.

    About 20% of the syllables are homophones under each of which there will be more than one morpheme, usually written with different characters, as in Luu ‘proper name’, 南 ‘brine ; gien 乾 “male principle’, 虑 ‘devout’, cien 肩 ‘shoulder’, 芭 “solid’. The degree of homophony is so low that it will be possible to write text either in literary or colloquial Chinese with the same character for each syllable, without distinguishing 各 from 元, etc., as shown in the accompanying texts below. (Cf. English bear ‘to carry’ or ‘the animal’, in- as ‘inside’ or ‘not’, bow, the last even phonetically ambiguous.) The above figures may be compared with the 3877 syllables of the dictionary Kuangyün (1007 A. D.) and the 1277 syllables of Standard Mandarin.

    1.2 Orthography 一 The system of characters is based on the following principles :

    (1) Legalize modern usage, such as 间 for 且 ‘canal lock’, 站 for 机 ‘bolt’.

    (2) Treat enlarged characters as extensions of the same linguistic form, as 返 ‘to return’,enlarged from 反 faan ‘to reverse,, reversed’; 伸 ‘to stretch’ from 申 shen ‘to extend’; 源 ‘source’ from 原 ‘plateau’ and 元 qiuan ‘primary’; 仁 ‘humanity, humanitarian’ from 人 ren ‘[hu]man’. In all these cases the primary character (unless it is a rare character) is chosen to stand for the group. Where there is a difference in pronunciation in the major dialects, then the characters will also be distinguished, e.g, 宣 siuan ‘proclaim’, but 喀 xiuan “to be noisy”, 十 cuu ‘ancient’ but 车 kuu ‘bitter’. When a character has more than one pronunciation with a difference in meaning or grammatical function, both forms will be used, the semi-circled [say what?] form being regarded as a different character, e.g. 空 kung ’empty : 空 ; kuq ‘vacant, space, leisure’ : 看 konn ‘to look at’ : 看 kon ‘to watch’; 长 dhyang ‘long’ : 长 dyag ‘elder, grow’.

    (3) It is anticipated that if 80% of the syllables are unique morphemes, it will probably not be a difficult next step to write the other 20% with only one and the same character for each syllable even if not etymologically related. This amounts to a 100% use of writing Chinese by “phonetic loan”, which was one of the six traditional categories of formation of characters. The situation is that when the ancients wrote a character by sound regardless of meaning, it was a “loan character”, while if a modern school boy writes one, he is punished for writing the wrong character!

    If a list of 2085 characters are used, one character to a syllable, it happens that it comes quite close to the number of characters used in present-day Japanese newspapers, for which there is a standard list of 1850 characters [the pre-1981 tōyō kanji list, now superseded by the 2136-character jōyō kanji secondary-school list], but with this important difference: while the Japanese characters list is admittedly only a part of the language (the rest of the language being written in kana). General Chinese is designed to write Chinese in general, without limitation as to style or vocabulary.

  48. J.W. Brewer: “Of course, that particular Argentinian regime (with periodic turnover in junta members) did not fall because of its regrettable domestic human-rights record. It fell because it picked a war with the U.K. and then lost it”

    Precisely so.

    The admirals and generals of the junta failed at the one thing they were legitimately paid and employed by the Argentinian state to do: seize and hold a few islands in the southern Atlantic through the use of military force. Faced with a bloody-minded, no-nonsense professional force undistracted by seizing hapless civilians in the middle of the night, the junta and the Argentine military were comprehensively defeated and sent packing in disgrace.

    (Though I must acknowledge that at least the Argentinian air force acquitted itself with great honour and élan.)

  49. “Of course it’s easy to poke holes in his linguistic analysis, but he’s not a linguist, he’s a writer talking enthusiastically about a language he loves…”

    Yes, he’s a language stylist. I don’t much agree with the points he makes but I understand the motivation.

  50. Lars Mathiesen (he/him/his) says

    JJM’s first comment reminds me of a Swedish Republican (anti-monarchist) who said “We made (Napoleon’s) Field Marshall Bernadotte king so he could get us Finland back from the Russians. They’ve had 200 years now, can we fire them yet?”

Speak Your Mind

*