More adventures in Latin American Spanish

Argentina was a rather surprising experience. In Spain, where I had learned Spanish, the stereotype of the Argentine in television and films must be based on people from Buenos Aires: one hears the same invariable accent with no hint of the immense variety that one would actually encounter in Argentina. As I cycled west across the country, I found the regional accents clearly changing every 300 km or so.

Once I reached the provinces of La Rioja and Santa Fe, I was shocked to discover that the dialect here had not experienced the shift of *y (and *ʎ > *y) to /ʒ/ like Rioplatense Spanish and the Argentinian stereotype. Instead, it was *r that had shifted to /ʒ/, while *y remained /y/. My first inkling of this was when rápido ‘fast’ was increasingly heard as [ʒapiðo], but it happened to instances of word-medial *r as well and took some getting used to in fast speech. A child came up to my wife and I at a campground and asked if we had seen a man in a [ɣoʒaroxa], and only after a minute of thought did I realize he was looking for someone wearing a gorra roja ‘red hat’. Weeks later, in Chile, while I was cycling on the motorway, another tourist stopped his car to ask me if he had missed the turnoff to [βiyaʒika], i.e. Villarrica. I laughed, thinking that he was lucky to have come across a non-local who could understand his question.

I have seen it claimed in several popular sources that the dialects of western Argentina are transitional to Chilean Spanish, but I didn’t find that to be the case at all. Not only does the shift of *r to /ʒ/ stop at the Andes, but the intonation of Chilean Spanish is vastly different. The Andes serve as a mighty wall. For the first week or so in Chile, I had to concentrate very hard to understand what people were saying, and I could sympathize with the many Spanish speakers who point to Chilean Spanish as the most difficult to understand of all the Latin American varieties. Fortunately, after that first week, my difficulties vanished and the local speech came to feel entirely normal.

I’m not quite able to determine what phonetic quirks set Chilean Spanish apart, and I’m not sure that if I hear this accent qua accent again in some other part of the world, I would be able to trace it to Chile. However, the Chilean colloquial lexicon is very sui generis, and I’m sure I’ll be able to immediately identify Chileans by the presence of certain words. People are very fond of the item ueyá/ueyón, which is not only a generic word for ‘thing’ rather like Philadelphian English jawn, but apparently even works as a exclamation and more. Chileans also tend to end sentences with po’h, a reduction of pues and a particle which has an exotic, non-Spanish air about it, as if something from an East Asian language.

Curiously, while Argentines accepted my use of vosotros without batting an eye, Chileans have been much more ready to make fun of me for it. They complain that the mere existence of such a form is silly, because Spain is the only place in the world where people say that. (Clearly Chileans never get to talk to a Spanish speaker from Western Sahara or Equatorial Guinea.) Once when having dinner with several upper-class and well-educated Chileans, I found tiresome the company of a writer-who-should-know-better who kept claiming that vosotros, and not the word itself as much as the grammatical form in general with its verb marking, was an innovation that appeared in Spain after the colonization of the New World; my appeal to Latin *‑atis etc. was dismissed because, as a foreigner, I surely cannot have any understanding of the history of the Spanish language.

Hopefully, after making my way through Uruguay, Argentina and especially Chile and finding it entirely possible to communicate with the locals (with perhaps a few days of acclimatization), I can now travel in the remaining countries of Latin America without fear. Still, it is always the variety of the language in the place where you first learn it that sounds the sweetest, and I am very much looking forward to passing through Madrid next month.

Adventures in Uruguayan and Argentinian Spanish

Except for a few very brief orders made at Mexican restaurants in North America, these last few days in Uruguay and the Entre Ríos province of Argentina was the first time I had ever spoken Spanish outside of Spain. All in all, what surprised me is how easy it was to communicate on both sides, in Uruguay at least. I could imagine someone who learned some particular regional variety of UK English having some problems in the American South, for example. Even when I used more recently-coined colloquialisms common to Spain, rural Uruguayans understood me. I do find that a bit puzzling, since the Uruguayans to whom I spoke claimed to have virtually no contact with Spanish of Spain: no music or films or television, and Latin America is a large enough market to sustain its own publishing without having to import any books from Spain. In Argentina, however, I’ve been forced to start adapting to their way of talking in certain contexts.

Over the years, other foreigners who learned Spanish in Spain have told me that going to Latin America would require avoiding vosotros and the verb coger ‘take’, but I find that an exaggeration. No one I met seems to mind the use of vosotros as the second person plural, and the indicative endings are so close to the vos forms used here that nobody would be confused by the morphology. While the verb coger has become an obscenity here, no one batted an eye when I used it in its Spanish meaning ‘to take’. Speaking with ceceo provoked no jokes at our expense.

The main aspects of pronunciation which required a brief moment of adaptation was the seseo and the pronunciation of *y/λ as [ʒ]. Once I crossed the border into the Entre Ríos province of Argentina, I started to hear people dropping final /s/, a common development in varieties across Latin America. Otherwise, it feels like everyone here speaks “clearly”. The major differences found were naturally lexical ones:

  • For ‘tap, faucet’, grifo is understood, but apparently only canilla is used here.
  • For ‘tent’, carpa is used here, though tienda has generally been understood.
  • Uruguayans understand los aseos/los servicios for ‘toilet’, but they say el baño, and I’ve found that I have to use the latter in Argentina to be understood.
  • For ‘peanuts’, people here say maní instead of cacahuete, and Argentinians don’t even understand the latter (if the word is explained to them, they tend to laugh at it).
  • For the simple small-town eateries in Entre Ríos, everyone says comedor, which elsewhere means ‘dining room’. I wonder if my asking Hay un restaurante por aquí? suggested that I wanted something posher than these little communities could boast.

Four levels of politeness in 17th-century Spanish

One of the more interesting books that I’ve read lately is Christopher J. Pountain’s A History of the Spanish Language Through Texts (London: Routledge, 2001). For the so-called Golden Age of Spanish literature, Pountain especially chooses texts by standardization-minded authors who inadvertently offer many details of the popular speech of their time. The following passage from Gonzalo de Correas’s Arte de la lengua española castellana (1625) suggests a much more complex system than the one found in Peninsular Spanish today, which is down to just tu and usted (and when I moved to Spain in the early millennium, I was urged to use usted much more sparingly than foreigners – on the basis of learning materials from Latin America – usually feel they should).

Devese tanbien mucho notar la desorden, i discordante concordia, que á introduzido el uso, ora por modestia, ora por onrra, ò adulazion. Para lo qual es menester primero advertir, que se usan quatro diferenzias de hablar para quatro calidades de personas, que son: vuestra merzed, él, vos, tu… De merzed usamos llamar à las personas à quien rrespetamos, i debemos ò queremos dar onrra, como son: xuezes, cavalleros, eclesiasticos, damas, i xente de capa negra, i es lo mas despues de señoria. Él usan los maiores con el que no quieren darle merzed, ni tratarle de vos, que es mas baxo, i propio de amos à criados, i la xente vulgar i de aldea, que no tiene uso de hablar con merzed, llama de él al que quiere onrrar de los de su xaez. De vos tratamos à los criados i mozos grandes, i à los labradores, i à personas semexantes; i entre amigos adonde no ai gravedad, ni cunplimiento se tratan de vos, i ansien rrazonamientos delante de rreies i dirixidos à ellos se habla de vos con devido rrespeto i uso antiguo. De tu se trata à los muchachos i menores de la familia, i à los que se quisieren bien: i quando nos enoxamos i rreñimos con alguno le tratamos de él, i de vos por desdén. Supuesto lo dicho, en las tres diferenzias primeras de hablar de merzed, él, vos, se comete solezismo en la gramatica i concordanzias contra la orden natural de las tres personas, xeneros i numeros.

The disorder and disconcordant concord which usage has introduced, whether through modesty, respect or adulation, should also be noted. For this it is necessary, first, to state that four different ways of speech are used for four qualities of person, namely: vuestra merzed, él, vos, tu … We usually call people we respect by merzed, such as judges, gentry, clergy, ladies and black cape people, and it is the highest after señoría. Él is used by older people for someone they do not wish either to call merzed or address as vos, which is lower, and typical of masters to servants; and common and village people, who are not accustomed to using merzed in their speech, address as él people to whom they want to show respect from their class. We call servants and grown up boys vos, and labourers, and such like people; and among friends where there is no gravity nor ceremony vos is used, and so in speeches made in front of kings and addressed to them vos is used with due respect and old usage. Children, younger members of the family and loved ones are called ; and when we get angry and quarrel with someone we call them él, and vos to disparage them. Bearing in mind the foregoing, in the first three of speaking (merzed, él, vos) there are violations of grammar and agreement against the natural order of three persons, gender and number.

One wonders how much this system was really agreed upon by all, and how much it was an idealization of shifting norms across time and space. The Hungarian I learned from Zsuzsa Pontifex’s Teach Yourself Hungarian back in the 1990s seemed to present a straightforward four-level system too: te, maga, Ön, tetszik. However, foreign learners are told very quickly that maga has been on the way out for decades, and if used today is just as likely to be pejorative as it is to tend towards showing respect. In other descriptions, the tetszik address is either replaced by another form of address, or a fifth level is added to the system.

Similarly, of the four-level system I’ve often heard proposed for Romanian – tu, dumneata, dumneavoastră, domnul/doamna – the second is rarely heard in Transylvania and the last is only heard from waiters at high-class restaurants who are clearly aping the French experience.

Romanian–Albanian parallels and the location of the Proto-Albanian Urheimat

Ranko Matasović has written a freely available grammatical sketch of Albanian for students of Indo-European. This is a useful resource alongside the University of Texas Indo-Europeanists’ Albanian lessons, and it’s nice that what has always been an obscure language is now easier to make an acquaintance with. (And if you really want to learn to speak the language, not just track its historical development, there’s the helpful new edition of Routledge’s Colloquial Albanian).

Matasović lists some developments in Albanian that are readily familiar to Romanian speakers. Lexically there is, for example, the borrowing of Latin paludem ‘swamp’ with metathesis and shift to the meaning ‘forest’: Albanian pyll < *padule, Romanian pădure. Romanian and Albanian both shift the velar in velar–stop clusters to a labial, e.g. Lat. luctāre ‘to fight’ > Alb. luftoj, Ro. a lupta.

Such Romanian–Albanian lexical parallels and some shared sound changes, which have been documented for at least a century now, suggest that these people were in fairly close contact as the Roman period came to an end. Romanian nationalist hysteria aside, I think the evidence overwhelmingly points to a Romanian origin west/south of the Danube. The area around Niš in modern Serbia is sometimes suggested as the Vlach homeland, from which the Romanians went north and the Aromanians south.

That the Albanians migrated to modern Albania from somewhere else has also been proposed, mainly based on the fact that words describing the sea and seafaring are Latin loanwords. If the claim that the Proto-Albanian lexicon also doesn’t reflect a mountainous homeland is true, then it would be hard to see southern Serbia, where Albanian–Romanian contact would have supposedly occurred, as the Proto-Albanian Urheimat. That region is not at all flat. One would have to go up further north in the Balkans to avoid mountains. Should the Proto-Albanian Urheimat be located in Vojvodina or Slavonia instead, with southern Serbia as only a later point of Albanian expansion?

Proto-Albanian had undergone some drastic sound changes in a comparative Indo-European context before contact with the Romans, the sort one would expect from an isolated population, not one on the plains open to contact (and convergence) with neighbouring language families. (Slavonia has marshes, but I wonder how much of a barrier they could have presented.) Perhaps Albanian only seems so weird because a larger IE branch in Pannonia and surrounding areas, with the same innovations, was gradually reduced by the Celtic, Roman, Germanic, Slavic migrations of the last centuries BC/early centuries AD.

A calque on Romanian in Transylvanian Hungarian

Another foreigner learning Hungarian, mainly in Hungary with references based on the standard language, has drawn my attention to a grammatical feature he has heard only in Transylvania and which he finds quite jarring. In standard Hungarian, the verb kell ‘to be necessary’ is used with an infinitive (with a personal ending) and optionally a subject in the dative case, e.g. (neked) kell dolgoznod ‘you must work’. Such usage is typical in the Uralic languages.

Some speakers, however, organize things differently, following the verb of necessity with an conjunctive-imperative form and optionally specifying the subject with a nominative personal pronoun: (te) kell dolgozzál ‘you have to work’. This is supposedly found only in these parts.

And if this is a feature specific to Transylvanian Hungarian, then I suppose it is a a calque from Romanian, inasmuch as the Romanian ‘to be necessary’ verb trebuie is optionally preceded by a nominative personal pronoun and followed by a subjective clause containing a finite verb: (tu) trebuie să lucrezi ‘you have to work’. It would be interesting to check if this phrasing or a similar one is also found in Hungarian of Vojvodina, as Serbian has the same construction, just with an indicative verb after treba da instead of a subjunctive mood.

It always comes as a bit of a shock to me when people from Hungary or other foreigners point to Romanianisms in Transylvanian Hungarian, since the ethnic lines are drawn so thick here that influence of Romanian (historically low-prestige) on Hungarian is not ordinarily expected. Hungarians in Cluj who have spent their entire lives around Romanian speakers, speaking Romanian from infancy as they play with Romanian children in the street, nonetheless speak Romanian with a distinct Hungarian accent, because it must be some kind of marker of identity, and with that aloof attitude they must also be resistant to conscious borrowings into the native language. In the Szekler land, where this feature is also attested, many Hungarians have virtually no contact with Romanian, and yet they still get these influences.

’lentil’ as a Romance loan into Common Slavonic

While reading Ronald O. Richards’ The Pannonian Slavic Dialect of the Common Slavic Proto-Language, where the author reconstructs Pannonian Slavic on the basis of loans in Hungarian, I was struck by the comparison Common Slavonic *lętja ‘lentil’ > Hu. lencse, as we must be dealing with a Romance loan in Common Slavonic. The Latin word for ‘lentil’ was lens, lentis, also a feminine noun. Richards notes that this lexeme is found natively only in South Slavic, spreading to East Slavic only in the 13th century.

As the Latin word persisted in Romanian as linte, one might be inclined to posit this as a word out of the Romanian/Aromanian Urheimat (presumably southern Serbia) that was brought far enough north to pass into Hungarian. On the other hand, Richards points to the presence of *lętja in both Slovenian and Bulgarian, and as those two languages are descended from South Slavonic dialects that moved in opposite directions out of Pannonia (Slovenian from Pannonia to the southwest, Bulgarian southeast through Transylvania and Wallachia), it’s hard to see how they could have each taken the loanword from one and the same language. I suppose there are three possibilities here:

  1. South Slavic unity managed to persist across the entire Balkans for some time, so that a word picked up by the South Slavic speakers in Serbia would easily spread to both Slovenian and Bulgarian;
  2. Contact with Romance occurred in multiple places in the Balkans, so that Slovenian borrowed ‘lentil’ from early Dalmatian, and Bulgarian from Romanian/Aromanian, and the forms just happen to look like they go back to Proto-South-Slavic;
  3. Or should we forget about positing a Dalmatian/Romanian/Aromanian origin for ‘lentil’ and instead consider this one of the very early Latin loanwords into Common Slavic, presumably dating from the time of the Roman presence in Pannonia? Unfortunately, since *lętja is not attested in North Slavic like some of those early Latin loanwords (e.g. *kolęda ‘carol, first day of the year’ < calendae), it’s hard to securely date it so early.

The languages of Czernowitz and old Bucovina

From my acquaintance with the life and work of Paul Celan – not to mention passing through on several occasions and seeing traces of its imperial past – I was aware that the former Austro-Hungarian town of Czernowitz was once home to a remarkable ethnic diversity, later erased as the surrounding province of Bucovina was ceded to Romania – whereupon it gained the name Cernăuți – and then the USSR and Ukraine for which it is now known as Chernivtsi. I was happy to discover Gregor von Rezzori’s memoir Blumen im Schnee (in English translation as The Snows of Yesteryear) which sheds much light on the changing demographics of the town. In reminiscing on his childhood nanny Cassandra, hired out of some remote village in the Carpathians, Rezzori makes the following comment on the languages that he heard spoken in his childhood:

She spoke both Romanian and Ruthenian, both equally badly—which is not at all unusual in the Bukovina—intermixing the two languages and larding both with bits from a dozen other idioms. The result was that absurd lingua franca, understood only by myself and scantily by those who, like her, had to express themselves in a similarly motley verbal hodgepodge. Even though it may be questioned whether I was actually fed at Cassandra’s breast, there can be no doubt that linguistically I was nourished by her speech. The main component was a German, never learned correctly or completely, the gaps in which were filled with words and phrases from all the other tongues spoken in the Bukovina—so that each second or third word was either Ruthenian, Romanian, Polish, Russian, Armenian or Yiddish, not to forget Hungarian and Turkish. From my birth, I heard mainly this idiom, and it was as natural to me as the air that I breathed.

How much things have changed in a century. Yiddish disappeared from Czernowitz with the genocide of its Jewish population in World War II. As Rezzori observes, German was already on the wane right after Trianon. From my experiences walking the streets of the city, it’s pretty much down to just Ukrainian, Russian and Romanian now. And while the intermixing of languages was simply accepted as a fact of life back then, today in at least southern (Romanian) Bucovina, the observation that a word in Romanian is of foreign origin is often taken as an insult.

The Romanian translation of A Clockwork Orange

My local bookstore has the Romanian translation of Anthony Burgess’s A Clockwork Orange (as Portocala mecanică) on display, and I was curious to see how translators Domnica Drumea and Carmen Ciora would deal with Burgess’s Russian-based Nadsat slang, as some of the book’s Slavisms had already entered Romanian and would not sound as jarring as they do in English. Let’s consider some of the first lines of the novel:

There was me, that is Alex, and my three droogs, that is Pete, Georgie, and Dim, Dim being really dim, and we sat in the Korova Milkbar making up our rassoodocks what to do with the evening, a flip dark chill winter bastard though dry. The Korova Milkbar was a milk-plus mesto, and you may, O my brothers, have forgotten what these mestos were like, things changing so skorry these days, and everybody very quick to forget, newspapers not being read much neither.

Eram eu, adică Alex, și ai mei trei gășkari Pete, Georgie și Moho, care era chiar mohorît, și stăteam noi în Lactobarul Korova, zdrobindu‑ne răzdoacele cum să ne umplem acea seară de iarnă, puiankă de lele săl­batică, rece și neagră, dar în același timp uscată. Lac­tobarul Korova era un mesto în care se servea lapte plus încă ceva și poate că voi, Fra, ați uitat cum erau mestourile alea, lucrurile schimbîndu‑se de speriat în zilele astea, toată lumea grăbindu‑se să uite și nici mă­car ziarele nemaicitindu‑se prea mult.

The cover of Humanitas’s edition of the Romanian translation of Anthony Burgess’s novel A Clockwork Orange as Portocala MecanicaThe Romanian translation entirely avoids English droogs (< Ru. друг ‘friend’), probably because Romanian already has a word drug of Slavic provenance that means something else (‘crowbar’). Instead it uses the ordinary Romanian word găşcari ‘fellow gangmembers’ < gaşcă ‘gang’, but it spells the word with a k to make it look foreign. The same is done later to the word puiancă, lit. ‘chick (fem.)’ but essentially just a filler noun like the English bastard.

The name of the character known as Dim in the original is, I daresay, a mistranslation. I’m fairly positive that Dim was so called because he was dim-witted. The translator names him Moho < mohorât, the passive participle of the verb a mohorî ‘to become dark in colour’, and the Dicționar explicativ al limbii române gives no use of the term as ‘stupid’ or the like.

I wonder if there is also a mistranslation present in Eng. skorry (< Ru. скорый ‘fast’) to Ro. de speriat lit. ‘scary (adv.)’. Did the translator perceive the English skorry as a mere alteration of the native word ‘scary’, unaware of its Russian origin?

The Russian borrowings rasoodocks (< Ru. рассудок ‘reason, sense’) and mesto (< Ru. место ‘place’) are adopted more or less as they are into Romanian, răzdoacele and mesto respectively.

Having already found a couple of choices I object to, I should probably stop now. My general experience with Romanian translations of foreign literature and quality control has not been a pleasant one. Still, A Clockwork Orange is a particularly interesting text to look at when considering translational choices across languages.

Down the Romanian cocoaz/Albanian kokëzë etymological rabbit hole

Curious about the etymology of Romanian coacăz ‘redcurrant (bush) (Ribes rubrum)’ I turned to Dexonline, and the entry for this word sees it as a back-formation from coacăză ‘redcurrant (fruit)’.

Things get more complicated with the etymology of coacăză. According to the entry for this word, the 2009 edition of DEX compares it to Albanian kokëzë. So, presumably another one of those several dozen words in Romanian drawn from a Balkan substrate.

However, the Dicționarul etimologic român from 1958–66 writes, In general one prefers to start from Albanian kokjë ‘boabă’, which is the same as the Latin [coccum], but this hypothesis is not useful, as a form *kokëzë does not appear in Albanian. So is the 2009 DEX giving us a ghost word with *kokëzë, an Albanian form that does not really exist and was simply unquestioningly repeated by editors personally unfamiliar with the Albanian language?

Luckily we live in the time of the internet search engine and can easily check this for ourselves. A Google search for kokëzë gives 347 results and mainly brings up hits from similar Romanian–Albanian comparative wordlists. This is understandable when one considers that there is a much smaller internet corpus of Albanian to draw on than Romanian or English. However, an online Albanian dictionary gives kokëzë for English ‘bulb’. So, it seems like this is perfectly cromulent Albanian and the old Dicționarul etimologic român was wrong.

But look here, Albanian kokëzë has been used in this translation into Albanian of Pablo Neruda’s poem “Juegas todos los días con la luz del universo” (number 14 from his collection Veinte poemas de amor y una canción desesperada). Neruda’s blanca cabecita is rendered in Albanian as kokëzë e bardhë. Neruda’s Spanish diminutive cabecita ‘head’, which clearly refers to the head of the poet’s lover, is translated into Albanian with a word that apparently means ‘bulb’. What is going on here? Did the translator perhaps misunderstand Neruda’s use of cabecita, which in faraway Mexico does refer to a bulb of garlic? No, for it turns out that kokë is the normal Albanian word for ‘head’, and kokëzë is a regularly formed diminutive (see here for this and many other Albanian diminutives in -(ë)z(ë)).

Yet Albanian kokë ‘head’ is, according to Orel’s Albanian Etymological Dictionary likely borrowed from Latin coccum ‘berry’, so the same botanical meaning as coacăz is primary. A semantic shift ‘bean, bulb’ > ‘head’ is understandable, I guess.

The decline of German (and French) as languages of culture

In his book Fractured Times: Culture and Society in the Twentieth Century, the late Eric Hobsbawm writes on the decline of Mitteleuropa as a sociocultural phenomenon. This remark on the disappearance of German as a language of pan-European culture struck me, being in line with some anecdotal musings that I have had for some time:

Equally, and perhaps even more significant, is the end of German linguistic hegemony. German is no longer the lingua franca of the educated from the Baltic to Albania. It is not merely that a young Czech meeting a young Hungarian or a Slovene will most probably use English to communicate with him or her, but that none of them can any longer expect the other to know German. It is that nobody who is not a native German speaker is now likely to use Goethe and Lessing, Hölderlin and Heine as the foundation of educated culture, let alone as the way from backwardness into modernity.

European literary culture has been massively affected by the change of the European lingua franca to English and increasingly nothing but English. I have always thought it remarkable how many European intellectuals born in the 1920s and 1930s were fond of Hölderlin, participating in his 20th-century resurgence even if they were from outside German-speaking countries proper. Hungarians and Italians upheld him as a poet to know. Now, a few decades on, Hölderlin may well be of interest only to those working in Germany, Austria or Switzerland.

It is not just German. I wonder if Samuel Beckett’s novels (namely the trilogy) have fallen so drastically into obscurity compared to his plays because the former were usually read on the continent in their original French, and now people are less likely to know French or at least read it for pleasure.