11

Today this is common, but how exactly did it start? I am looking for examples in various languages, and suggest:

  1. Exclude Latin (as more “ancient” or “international” than “foreign”)
  2. Exclude French after, say, Huygens (as well-known: Leibniz, Euler, Jacobi, Dirichlet, etc.)
  3. Exclude anything resulting from a visit or permanent move to the foreign country (thus Cauchy or Riemann in Italian, World War refugees in English, etc.)
  4. Exclude translations not originated by the author.
  5. Do not otherwise exclude (early) English and German!

Bend these rules if necessary. Mathematical physics or celestial mechanics qualify. This ties in with various questions at hsm, but I am hoping for a wider pool of knowledge here. So far I am only aware of isolated(?) examples by Abel, Plücker, Lie, Lorentz, without clear lineage to the present.

Francois Ziegler
  • 29,500
  • 1
    You probably want to exclude German as well, right? – Asaf Karagila Sep 14 '18 at 05:25
  • No, not at all. There must be some between Abel (1820s) and Lie (1870s), but I haven’t found them. – Francois Ziegler Sep 14 '18 at 05:28
  • 12
    The situation is quite comparable to Latin, so you may want to exclude it, but many scholars during the Islamic Golden Age published in Arabic, even if it wasn't their mother tongue. An early example is خوارزمی (Khorezmi, arabicised in al-Khwārizmī), a striking one is خیام (Omar Khâyyam), who published his scientific work in Arabic, but is also one of the most famous Persian poets. Some of the most prominent examples may not fit your "(from home)" clause, though. – PseudoNeo Sep 14 '18 at 07:00
  • 10
    A late example is John Brillhart, Patrick Morton, Über Summen von Rudin-Shapiroschen Koeffizienten, (German) Illinois J. Math. 22 (1978), no. 1, 126--148. MR0476686 (57 #16245). The authors, both American and living in the US, decided for some reason to publish a paper in German. If I remember right, the reviewer complimented them on the excellent English of the Abstract. – Gerry Myerson Sep 14 '18 at 12:39
  • 5
    OK, maybe a stretch, but Robert Rankin published a paper in (Scots) Gaelic: https://mathscinet.ams.org/mathscinet-getitem?mr=28851 This was not, at the time, an official language of the United Kingdom, and not his mother tongue either. A nice twist is that he gave his name in the paper as Rob Mac Fhraing, whereupon the unsuspecting editor sent it back to him to review. – Bort Sep 14 '18 at 13:45
  • 2
    While we are collecting recent examples: Jeff Paris, a native English speaker, published a paper in Czech (https://dml.cz/handle/10338.dmlcz/118206), though I guess it does not count as he was on an extended visit in Prague at the moment. – Emil Jeřábek Sep 14 '18 at 15:17
  • Regarding German: while Gauß was still writing all his works in Latin, after him German became the language almost exclusively used in German mathematics and it remained so until WWII. This seems to have changed soon after WWII. – ThiKu Sep 14 '18 at 20:40
  • 7
    A conjecture both deep and profound/ Is whether a circle is round./ In a paper of Erdős/ Written in Kurdish/ A counterexample is found. – Gerry Myerson Sep 15 '18 at 00:26
  • I believe Yuri Ivanovitch Manin learned one of the languages of Greenland at some point. Not sure if he ever published in it, and even if he did, would that count as "early" or not. –  May 15 '19 at 07:12

5 Answers5

17

Maurice Fréchet published some of his work in Esperanto (for example, La kanonaj formoj de la 2, 3, 4 - dimensiaj paraanalitikaj funkcioj) -- it can't get more foreign than that, I presume.

Carlo Beenakker
  • 177,695
  • 13
    Maybe it can: Peano wrote some of his work in a language which he had invented („Latino sin Flexione“) and which hardly anyone besides him understood. – ThiKu Sep 14 '18 at 06:07
  • 2
    which hardly anyone besides him understood seems a bit excessive. It's not that different from Latin. See for instance https://www.gutenberg.org/files/35803/35803-h/35803-h.htm . – Federico Poloni Sep 14 '18 at 13:41
  • 4
    Esperanto is not a foreign language. It is an international language. So if Latin is excluded, so should be Esperanto (... although I am not entirely sure about the purpose of the question). – Algernon Sep 14 '18 at 14:10
  • 2
    @Algernon Curiosity and no nefarious ulterior motive, I promise. – Francois Ziegler Sep 14 '18 at 15:15
  • 1
    @Algernon : I think Latin is different from Esperanto in the sense that it is more likely that someone publishes in Latin for pragmatic reasons (wider audience, prestige of the language) while publishing in Esperanto may be for ideological reasons (wanting to promote Esperanto). – Timothy Chow Sep 14 '18 at 17:48
  • A Polish logician, philosopher and historian of science, Edward Stamm (1886-1940) (of German extraction), also published at least 1 paper in Latino sine Flexione. He corresponded with Peano and was an enthusiast of the language. – Margaret Friedland Sep 16 '18 at 00:53
  • To me this "latino sine flexione" is far easier to understand than real Latin. But this must be due to the fact that I studied Latin in highschool, plus the fact that my native tongue (French) is flexionless. – Sylvain JULIEN Sep 16 '18 at 16:19
  • Earlier use of Esperanto: Aleksandr Dombrovski (1904, 1906), René de Saussure (1909). – Francois Ziegler Nov 22 '18 at 03:57
7

It seems that some (many?) Japanese mathematicians published their articles in german around the start of the 20th century until around the 40s. It seems that publishing in German was quite common around that time as Germany was probably the world leading country in several areas until the sad times began around 1930.

A famous example is Tadashi Nakayama whos first 15 papers are all in German: https://projecteuclid.org/download/pdf_1/euclid.nmj/1118801606

Mare
  • 25,794
  • 1
    Nakayama was a student of Takagi. Takagi also published his classic articles in german, but also he studied with Hilbert. – efs Sep 14 '18 at 14:45
  • 2
    In the 70's, it was quite common for Japanese mathematicians to write in French, even in japanese journals. – Denis Serre Sep 14 '18 at 15:37
  • 1
    Was it common in Japan at that time to learn another language (maybe even in school)? Or did those people learn German just for reaching a wieder audience in math journals or did the papers even get translated by someone due to their importance? – Mare Sep 14 '18 at 16:57
  • French is not a language of interest in this question, but it is the main language of publications by Kiyoshi Oka (1901-1978), a renowned complex analyst. He spent a few years in Paris, which may explain the choice of language. On the other hand, his works on iterations of holomorphic functions from 1930s, undoubtedly influenced by (slightly earlier) French reserach, are in Japanese. – Margaret Friedland Sep 16 '18 at 00:44
  • Oh right. Nakayama (1935-), Shoda (1928-), and earlier in both English and German: Takagi (1902-, at first from Göttingen), and in the same journal, Fujisawa (1888-), Kitao (1887). Anything earlier? – Francois Ziegler Sep 17 '18 at 18:09
  • @Mare From/to Japanese I don’t know, but in the late 19th century Sugaku did publish translations from French and German into English: e.g. Dirichlet, Kummer. – Francois Ziegler Sep 17 '18 at 18:20
7

The obvious examples would include pretty much anyone who was not born speaking any of the "Congress" languages. They (or should I say "we", being Polish) would publish their mathematics in a more widely spoken/read language in order to increase the chance of its being read and understood. Let me however give some less obvious answers involving two Polish students of Sophus Lie (still having to do with dissemination of the ideas, perhaps tailoring to the intended audience). Kazimierz Zorawski (1866-1953; PhD 1891) published over 70 works in his lifetime, roughly half of them in German and half in Polish (understandable). However, there are also 3 works in Czech (1914-1915). He was a member of the Czech Academy (since 1910), but otherwise had no Czech connection. Another student of Lie, Lucjan Emil Boettcher (1872-1937; PhD 1898), published mainly in Polish (with some early and late exceptions in German and French), but his most cited paper is "Glavnyshiye zakony skhodimosti iteratsiy i ikh prilozheniya k' analizu" ("Main laws of convergence of iterations and their applications to analysis"). He published it in 1903-1904 (in parts) in Russian, in "Bulletin de la Societe Physico-Mathematique de Kasan".

Margaret Friedland
  • 3,659
  • Thank you. Do you have a sense of when each “Congress” language started to be used in this way? From other answers I get the impression that it was maybe 1870s for German from Scandinavia (Bäcklund, Lie), slightly later for English and German from Japan, and that writing in one another’s “Congress” language (other than French) remained exceptional until much later. – Francois Ziegler Sep 17 '18 at 16:21
  • @FrancoisZiegler: I did not study the subject of international mathematical cooperation in depth, but I agree with your impression. I would date institutional use of Congress languages to the Heidelberg Congress in 1904-- there were 4 plenary talks, one each in German, English, French and Italian. See the book by O. Lehto, Mathematics without Borders, https://www.mathunion.org/fileadmin/ICM/Lehto/Lehto_Mathematics_Without_Borders/Lehto_Mathematics_Without_Borders.pdf – Margaret Friedland Sep 18 '18 at 00:42
  • Ah sorry... By “used in this way” I meant “used in order to increase audience” (e.g. from Poland), not “used at ICM”. – Francois Ziegler Sep 18 '18 at 19:45
  • My misunderstanding. The only early 19-th century Polish mathematician publishing in a foreign language (in this case, French) ``from home" was Kajetan Garbinski (1796-1847), a geometer, a professor of the University of Warsaw, who spent some time studying in Paris. But I do not have any titles at hand. Jozef Maria Hoene-Wronski (1776-1853) writing solely in French does not count, as he lived in France since 1800. Although Wronski apparently specifically declared that he chose to write in French rather than Polish in order to increase his audience. – Margaret Friedland Sep 19 '18 at 00:34
  • Thank you! From East, I also notice Archiv für wissenschaftliche Kunde von Russland with e.g. two papers of Lobachevsky (1855, 1858) which may however be translations. Lobachevsky also published in French, the Bolyais in Latin and once German (1851). – Francois Ziegler Sep 20 '18 at 06:44
3

Many Scandinavian mathematicians published in German. I don't know any very early examples, but some prominent ones in the 19th and 20th century. From Finland you have Mellin (1883-) or Nevanlinna (1921-). Mittag-Leffler published mostly in French, but also some articles in German (1875-), Italian (1899-), and English (1900-).

Francois Ziegler
  • 29,500
Kusma
  • 316
  • 1
    One 20th century example is Jakob Nielsen. He published some papers in his native Danish, some in German, and some in English. – Lee Mosher Sep 14 '18 at 15:01
  • One must say that Nielsen actually was a German citizen for the first 30 years of his life. (His hometown had been occupied in the German-Danish war 1864 and only came back to Denmark in 1920.) He did his PhD in Kiel with Max Dehn. – ThiKu Sep 14 '18 at 16:50
  • Sophus Lie often published in German; he was Norwegian. – Ben McKay Sep 14 '18 at 19:21
  • Well, he actually was a professor at Leipzig from 1886 on. But it is true that he published in German before that date while still being in Norway. – ThiKu Sep 14 '18 at 20:27
  • @ThiKu Can you tell if Scandinavians publishing in a foreign language became common after Lie, or only decades later? Because beside him, the few 19th century examples so far (Abel 1820s, Plücker 1848, 1865, add Clausius 1851, 1856, Beltrami 1869) feel increasingly like exceptions rather than a rule in the making. – Francois Ziegler Sep 15 '18 at 18:58
  • 1
    Looking at some of the usual suspects it seems to me that publishing in French or German was not uncommon. Lindelöf (the father) had papers in French e.g. in 1861 and 1871 (although Helsinki belonged to Russia at the time), also his son published in French, Mittag-Leffler wrote much in both French and German, Bäcklund seems to have written all his work from 1874 on in German, von Koch (somewhat later) seems to have written in French. – ThiKu Sep 16 '18 at 16:41
  • Given the probably limited number of research mathematicians at the time, this suggests that at least after 1870 publishing in other languages was more the rule than the exception. For the time before 1870 I guess the number of research math publications might have been limited anyway. – ThiKu Sep 16 '18 at 17:02
  • Carl Friedrich Degen from Kopenhagen, who was considered the leading Scandinavian mathematician at Abel‘s time, seems still to have written in Latin. Christiania (Oslo) has only one math professor at the time, Søren Rasmussen, of which I couldn‘t locate publications. (Perhaps the name is to common for googling.) The first math professor in Sweden was Carl Hill (hired 1830 in Lund), who wrote Swedish and Latin. Also the Uppsala school seems to have published in Swedish, at least this is the case for Malmstens paper from 1865. – ThiKu Sep 16 '18 at 17:04
  • Gårding‘s book is a good reference for Scandinavian mathematics, by the way. – ThiKu Sep 16 '18 at 17:11
  • @ThiKu Thank you! I just came across another: 3 articles in German by Ole Jacob Broch in 1844–1850. – Francois Ziegler Sep 19 '18 at 13:27
3

A surprisingly recent example:

Don Zagier:
Een ongelijkheid tegengesteld aan die van Cauchy
Proc. Koninkl. Ned. Akad. v. Wetensch. (Indag. Math.) 80 (1977) 349-351

There are also several papers in French in Zagier's list.

Noam D. Elkies
  • 77,218
  • 2
    curious for the early date; Zagier would later become professor at Utrecht University. – Carlo Beenakker Sep 16 '18 at 17:58
  • 1
    Also recent: http://www.numdam.org/item/ASNSP_1965_3_19_1_113_0 Manin, Juri I. Moduli fuchsiani Annali della Scuola Normale Superiore di Pisa - Classe di Scienze, Série 3 : Volume 19 (1965) no. 1 , p. 113-126 – Keerthi Madapusi Sep 18 '18 at 02:00