Language
Accents
Where did that Spanish theta come from (and where did it go)?
Jan 25th 2012, 21:44 by R.L.G. | NEW YORK
ANOTHER question from readers: Why does the "theta" sound in Spanish Spanish not prevail in Latin America? For those unfamiliar, the letters c (when followed by e or i) and z (always) are pronounced like the English "th" as in "thin", in Spain. (It's often called theta after the Greek letter θ, which is pronounced the same way.) My favorite example is civilización, which comes out thivilithathion in Spanish Spanish but not in Latin America. Nate in Phoenix, Arizona asked if this was because, as he had heard, Spanish settlers in Latin America came from the south of Spain (Andalucia), where the theta is not used?
I didn't know the answer, so wrote to John McWhorter, an expert in language contact and change. His reply:
While theta happened in Castille, some of Andalusia developed into just /s/ while other places there, and elsewhere, had an opposition between /s/ and an apicoalveolar /s'/ between /s/ and theta. Both of the latter are still around in Latin America, as opposed to just /s/, and thus the idea that Latin America has /s/ from Andalusia fails from both sides of the pond. But it's true that there's no theta in Latin America because Castillians didn't come over much.
So Nate's conjecture was almost but not quite right.
In this blog, named after the dictionary-maker Samuel Johnson, our correspondents write about the effects that the use (and sometimes abuse) of language have on politics, society and culture around the world
Advertisement
- Analects | China
- Americas view | The Americas
- Babbage | Science and technology
- Bagehot's notebook | British politics
- Banyan | Asia
- Baobab | Africa
- Blighty | Britain
- Buttonwood's notebook | Financial markets
- Charlemagne's notebook | European politics
- Clausewitz | Defence, security and diplomacy
- Democracy in America | American politics
- Eastern approaches | Ex-communist Europe
- Elysée | France’s presidential election
- Free exchange | Economics
- Game theory | Sports
- Graphic detail | Charts, maps and infographics
- Gulliver | Business travel
- Johnson | Language
- Leviathan | Public policy
- Lexington's notebook | American politics
- Newsbook | News analysis
- Prospero | Books, arts and culture
- Schumpeter | Business and management
Most commented
- The euro crisis: The Greek run
- Foreigners in China: Barbarians at the gate, again
- Brazil’s economy: A bull diminished
- Russia and NATO: Rethink the reset
- Gay marriage: North Carolina begs the question too
- Nationalism online: Backfired
- Turkey’s cultural ambitions: Of marbles and men
- Politics and private equity: Bringing the Bain
- Greece and the euro: Exodus, chapter 1
- India-China relations and the media: Blame the messenger
Over the past five days
Most recommended
- The euro crisis: The Greek run
- Greek politics: Slouching towards the drachma
- Foreigners in China: Barbarians at the gate, again
- Brazil’s economy: A bull diminished
- The euro crisis: Going, going, gone?
- Jacob Zuma: Portrait of a president
- Rival versions of capitalism: The endangered public company
- Bo Xilai v Chen Guangcheng: Who is the mightier?
- Facebook goes public: Not top of the pops
- Global Zero: Fewer nukes, more security
Over the past seven days
Advertisement
Readers' comments
The Economist welcomes your views. Please stay on topic and be respectful of other readers. Review our comments policy.
Sort:
I've studied linguistics in Spain for two years, and the theta is used in the north of Spain, but also in parts of Andalusia. In Seville and the Canary Islands, where the boats left from, they only use the /s/. Therefore, even if there were Spaniards from the north, they learned the southern dialect in the few years they were there while waiting for their boat to leave.
I've heard they got it from the Habsburgs.
And yet we all speak Castillian in Latin America...
Speaking of pronunciations, as someone who listens regularly to the audio edition, I note that the readers of the Economist routinely pronounce 'junta' with an English 'j' sound. I suppose the rationale must be that the word was adopted into English, and now is an English word with an English pronunciation.
In contrast, there is an insistence on using 'furore' instead of the legitimate (American) English 'furor'.
Besides these there are numerous French words, some of which get pronounced with a pretty straight French pronunciation, some Anglicized, and some just "abominized".
First, just to clarify some recurrent misconceptions noted amongst the commentaries, from the point of view of language history:
-Neither American or Peninsular Spanish is monolithic
-It's an oversimplification to think that any one Peninsular variety is the single source of any particular American variey- America was colonized and recolonized over several centuries. Migrants and itinerant Spaniards came from different socio-economic baclgrounds, regions and in different periods of history. AN observation has been made that a good number of terms that can be characterized as Latin American, that is nearly universal in AMerica and largely unknown in Spain, relate to sailing/naval terminology, reflecting a common denominator experience throughout the development of the region. There has also been significant migration within America from early on. Many Dominican creoles went on to other countries during the expansion of the Empire, just as one example. Furthermore there has been significant influence on Peninsular Castilian from American varities, the most notorious case probably being that of the Canaries which exported much influence and re-imported it via, notably, Cuba. My personal experience in the Canaries was how similar it sounded to Venezuelan pronunciation (similar, not the same;) There again, Venezuelan means what? Caracas (presumibly), Maracaibo, Merida, Cumana...
-There are variations within and fuzzy borders between varieties
-Languages evolve on their own- no outside source required
-Neither American nor Spanish varieties are the same as any Peninsular varieties 200 years ago. All varieties have continued to evolve since then
-Political domination of a country typically has very little effect on the native language (in reference to comments about Spanish influence on Italian). The Mozarabs continued to speak Romance during several centuries of Moorish rule, just to give one example
Having said all that, now on the topic of zeta, SPain has a long history of centralization and standardization, applied to the language as well. Here is what I recall my professor of linguistics at Valladolid, saying years ago. Basically, that at some point in time in what could be genrally called Castilian, at least three distinct phonemes were current all using the sign Z: the sound current in Peninsular CAstilian now, the phoneme used throughout America, the Canaries, and, although mostly dying out no, parts of Andalucia/ SOuthern SPain. The third sound was the one represented in current Italian standard /ts/. Just as mysterious could be why this third sound is no longer current in ANY Castilian variety! ANd anyway, while the orthographic reform straightened out the spelling of two of these phonemes, in AMerica this was never adhered to. This no doubt because it had already been settled in many areas, through assimilation of all three into something similar to English S. Whether that came from ANdalucia, where it would have been current at the time, or whether all of the mentioned areas evolved similarly, etc, is not clear.
BY the way, this has nothing to do with aspiration of the ess sound. That is no doubt what the person below heard in Argentina, and is common to one degree or another in nearly all varieties of Castilian SPanish. When that aspiration occurs before a /t/, it could be understandably reconstructed as 'theta', ie caHtellano construed as caTHellano.
Here ends my unedited rant, dixe. O bien, he dicho.
It is a consequence of the evolution of the sounds that started in the 14th century onwards. When the Spaniards conquered America the sound /θ/ didn't exist in Spanish yet, the sound was like "pizza" in Italian and was written using "z" (before "a", "o" and "u") and "ç" (before "e" and "i") graphies. The sound /θ/ came up in the 17th century only in the northern dialects of the Iberian Spanish, whereas in the Southern dialects (Andalusia, Canary Islands, the Americas) it evolved into /s/. In the same way, during that time the sound /ʃ/ (like "ship" in English) evolved into /x/ ("jamón", "jarra") in all the Spanish speaking world, thus "Mexico", "Texas" or "Quixote" turned out to be said "Méjico", "Tejas" and "Quijote", and the letter "x" was left for its current sound /ks/, like in English.
This the best comment so far, it's exactly this way how it evolved. It must be added that in Southern Spain there is a double shift, in some towns this /ts/sound evolved into the /s/ sound, merging with the sound of the letter s (as in the Americas), in that way Caceres would be "Caseres", and in other towns (I'd say it's the majority in rural areas) both z and s have shifted to the sound /θ/ (as if it had been an attempt of hyper-correction when the other shift happened), so that Caceres has become an awkward "Cathereth". Although the All-/s/ sound pronunciation is seen with tolerance, the All-/θ/ pronunciation is seen as a truly low social status mark.
In Peru, you would be asked, "Do you speak castillo (habla castillano?)" , not "do you speak Spanish (habla español?)". Seems odd, if there were few Castillans who came to south america.
Most of Spanish-speaking South America, at least everywhere I've been, Castellano replaces Español. But nowhere that I've been in South America do they call the local language cathtellano. It's a pickle.
I'm from Peru and we always call Spanish, Castilian. I was surprise it was called Espanol by Hispanic in the US.
Why in Argentina/Uruguay etc., they use "sh" sound to replace "y" or "ll" ...where did that come from?
This, I think is a relative thing. In general, it seems that 'y' is pronounced as what we might call a hard 'y' sound, and in the extreme can come out sounding almost like a 'j' or I suppose 'sh' or variations thereof.
It's not "sh" but rather like the "gg" in Italian "pioggia"; perhaps the sound was donated by the numerous Italian immigrants to the area. A man using the "sh" sound would be labeled as effeminate.
To complicate things even further, in Italian, the letter S is pronounced the same as in English in story - except in northern Italy, where the S is pronounced with varying degrees of lisp, like in Castilian Spanish. It could be due to the fact that part of Italy was occupied by the Spaniards (or, at least, had Bourbon kings) until the full reunification of Italy 150 years ago - except that the pronunciation with the lisp is used in the north of the country, which underwent a relatively short Spanish occupation and was later occupied by the Austrians, while the south (from Naples to Sicily) has a perfectly clear S, like in French or English. Perhaps the Spaniards copied the Italians? I am sure that there is an explanation for that, too.
There's no such sound as the Castilian theta in Italian, either in Northern or Southern Italy [1]. Nor is there anything like a lisp when we pronounce the alveolar sibilant (I'm from Verona, Northern Italy), unless it comes from a speech impediment.
When it comes to the /s/ phoneme, the main difference between "proper" Italian (i.e. Tuscan) and its northern dialects is that the /s/ sound become voiced when it occurs between vowels, in any case: ex. "casa" ('house') is pronounced /'kasa/ in Tuscan, /'kaza/ in nothern Italian dialects. Tuscan has the voiced alveolar sibilant too, and it also occurs between vowels, albeit it's not so common: ex. "caso" ('case') is pronounced /'kazo/, "ròsa" ('rose') /'rɔza/ (but "rósa", 'gnawed' is /'rosa/). The /s/ in the "-oso" suffix, on the contrary, is always unvoiced: ex. "ventoso" ('windy') is pronounced /venˈtoso/.
[1] To be honest, there are some vernaculars in which you can find the voiceless dental (ex. some Veneto vernaculars), but no speakers of such vernaculars pronounce it when they speak Italian.
Could a reason be that the Spanish spoken in Latin America is from the Spanish a couple of hundred years ago? I'm thinking of how Chinese went into Japanese and the pronunciation is from the time of that cross-over, or the rumour that North American English is reminiscent of the English that was spoken when the first English immigrants arrived.
There may be a point there, as Americans of Italian descent, if they speak any Italian at all, are very likely to speak the Sicilian or Neapolitan dialect, hardly ever pure Italian. It's hilarious watching American films, which contain a bit of dialogue in Italian: it's usually a strong Italian dialect, usually from the south.
@Cederkoorcom,
Something similar is said about Brazilian/European Portuguese - that Brazilians have preserved the pronunciation their Portuguese grand-grandparents brought to America in colonial times.
The main difference is that Brazilians pronunce vowels clearly, while European Portuguese is clipped, very consonantal - they seem to "swallow" the vowel sounds.
The thinking is that Brazilian speech is closer to the earlier forms of Vulgar Latin developed in the Iberian peninsula and to Latin itself, a "flowing", melodious language with the vocals clearly spoken (just like modern-day Italian).
Interestingly, the same is true for Spanish speakers too.
The many regional Latin American dialects are usually more open and clear than the present-day clipped European Spanish pronunciation, be it Castillian, Northern or Andalucian.
Listen to aomeone from Colombia - maybe the clearer-speaking Hispanic-Americans - and to a Spaniard, and you'll notice the same difference as between a Brazilian from Bahia State - maybe the clearer-speaking Brazilians - and a Portuguese.
'Pure' Italian? From whose perspective? Maybe standard Italian.
It's all so arbitrary. Nonetheless I would propose Mexicans as the clearest speaking. And slowest. Maybe I get the impression of slowness because they dont seem to aspirate as much as some others. That /s/ really seems to slow things down. Having said that, another arbitrary comment, to me the more aspirating varieties sound the most normal, fluid and generally "attractive". In fact the /s/ in certain places instead of /h/ sounds almost as affected as /θ/. And I know it's not because they (neither one) are in fact affected. Just the arbitrary associations we make with something as harmless as a phoneme.
Well, the notion of the Italian immigrant being from the south is also largely exagerated. Not only were early immigrants often form the north, even in the 60s many came from there, too. If you get to talking to Italo-Americans, you will soon find this out. And again, there is no "pure Italian". Even Standard Italian is an artificial concoction. Lingua toscan in bocca romana. Even Tuscan dialects vary from the standard and Romans have their own linguistic distinctions. So in short, anytime you here Standard Italian, it's going to have one or another regional accent at least. Because it is native to no one.
Linguistic change is fairly constant. Neither contemporary British or North American varieties of English are identical to the English of 250 years ago: however _what_ has changed in each variety is different, so that North American English tends to preserve some historical features mostly lost in England (e.g. pronouncing all Rs), while English English tends to preserve some historical features mostly lost in North America (e.g. the distinction between Mary, marry and merry).
The same is true of Brazilian vs. European Portuguese, Canadian vs. European French, Spanish vs. Portuguese, Latin vs. Greek vs. Sanskrit, or virtually any two languages with a common ancestor.
@grover cleveland,
Point taken. I certainly simplified and exaggerated (in the best journalistic tradition).
Maybe I should have said that the proponents of this theory suggest that the open, vocal quality of Brazilian Portuguese is just one historical feature preserved in Brazil and lost in Portugal since colonial times.
I once told a Spanish friend that many words used currently in Castillian were close to Portuguese words long forgotten or not used anymore. Listening to a Spaniard speak, sometimes, made me think of Camoens or Portuguese writers of the 17th or 18th centuries.
"Funny", he said. "I was about to tell you the same thing. You Brazilians use words that seem to me straight out of a Cervantes or Góngora page"! Thanks for your reply.
@Bayonet,
I guess I displayed a South American bias - that of someone who has travelled to almost all Spanish-speaking countries in South America. But not to Mexico, the big northern cousin.
But I have recently watched some Mexican films. They certainly speak very clearly (and with a pleasant lilt). I'll agree to a draw. Thanks for your reply.
The comment that "Castillians did not come over much" needs to be reviewed. The sound theta came into castillian language from basque, and a significant number of basques did go to America. Furthermore at that time these basques were vassals of the king of Castilla.
This is true. I was researching family history and found that after the War of the Spanish Succession, the Bourbon Kings of Spain encouraged settlers from NW Spain to settle in NW Mexico. My conjecture is that the Spanish theta was a "posh" accent that would be viewed as effete and upper class in Latin America and mark one as an outsider.
Unlike the relationship between US and England, Hispanic creoles have never considered Peninsular SPanish to sound anything but unrefined, the term I have so often heard is more precisely 'rustic'. The SPanish zeta and jota rather typify this and are the basis of parodical miming through the AMericas. The concept of the rustic SPaniard only crystalized during the 40's-60's when millions of poor and uneducated Spaniards emigrated to Argentina, Venezuela, Cuba, Mexico, and nealry everywhere else. To add inslut to injury, the prevailing stereotype of the SPanish is that their speech is inarticulate, laconic and most of all supremely crude.
The creoles of the AMericas continue to view their own world as one of abundance, despite bungling and mostly corrupt mismanagement of it (often conveniently tagged as the SPanish legacy). They see themselves as modern and cosmpolitan compared to the still widely held stereotype of the Franco-Era countrified, old-fashined and undernourished Spaniard. Interestingly, the Spaniard immigrant archetype departs from the colonial one, in that it typifies a person who is honest, hard-working and decent.
I lived in Andalucia for 14 months and their lisp is as strong as Castilla/La Mancha and the rest of the Spanish speaking portion of the peninsula.
mcwhorter failed a bit -- not only were most people coming over from spain from andalucia, but boats almost always stopped in the canaries on the way, which are also theta-free.
Is John McWhorter saying that there are places in Latin America that use theta? I have never heard of them. As far as I know all latin american spanish speakers make no distinction of any kind between z and s (not that I can hear anyway).
By the way, the idea that a lisping king influenced the castilian switch from a sound produced with the tip of the tongue to something more like a "th" is not completely bonkers. Fashions in language do spread from influential speakers. But of course it is impossible to prove.
But the pre-lisping-King sound could not have been /s/. There were originally two sounds:
* one corresponding to written z and ç
* one corresponding to written s
Today non-Castilian Spanish accents pronounce both the same, but Castilian speakers distinguish the two. And it's easy to show that this distinction derives from Latin and earlier forms of Romance -- it's not something the Castilians invented. Basic principles of historical linguistics tell us that earlier forms of Spanish must have made this distinction somehow in pronunciation -- in other words, they could not both have been pronounced /s/.
You are absolutely correct. There must have been two distinct sounds. McWhorter describes the pre-"th" sound as an apicoalveolar /s'/. I had to look this up - it is an "s" made with the tip rather than the blade of the tongue. That sounds plausible. I believe it has turned into a "ch" sound in modern Italian.
Actually the "ch" sound of Italian must have been an earlier form.
If we compare a particular set of words for "a hundred" we can see the development.
Latin "centum" was actually pronounced "kentum" in the days of ancient Rome. The /k/ sound still survives in Sardinian. /k/ is made near the back of the mouth, but the sounds developed further and further to the front over time.
Italian "cento" is pronounced with a /tʃ/ sound (like English "ch"), which is further forward than /k/ but further back from /s/.
Further forward still is /ts/, like in English "caTS" or German "zahlen". This was used in medieval Romance, although it doesn't survive in any modern Romance languages. So, for example, medieval French "cent" was actually pronounced something like "tsent".
/ts/ developed to /s/ in modern French "cent" and Portuguese "cem", but to /t̪θ/ and then /θ/ in Castilian "ciento". /θ/ is actually the frontest sound of all.
"As far as I know all latin american spanish speakers make no distinction of any kind between z and s..."
You're absolutely correct. I've been speaking Spanish with Latin Americans for over thirty years, and I've never heard a single one that uses ceceo. Always seseo. In other words, in Latin American Spanish, the consonants C (before e and i) and Z, and S are pronounced identically.
It could be that in certain regions of Spain, the distinction never existed in the local Romance or it could be that it evolved in certain regions, or assimilated, into just /s/.
If the current Peninsular treatment isn't something of an invention (it is, actually, as it is the result of orthographic reform)then what is the basic linguistic principle that explains why modern Castilians use Zaragoza and pronounce it accordingly, when this comes from Latin Caesaraugusta? Check out this link and use comparsisons to earlier forms and Catalan and Italian to get the picture. http://www.zaragozaonline.com/2007081783/Etimologia-de-Zaragoza
I don't fully agree with the expert. Almost 3/4 of the Spanish people who went to the Americas were from Andalusia. If 3 out 4 people have certain accent, the final form will be of the majority. The /th/ sound (actually something more soft) is present in Basque and stayed around in the central north of Spain. Thus, this sounds that later developed in Castille came from pre-Roman times. It's not present in the Arab language, nor in most of the languages. But the Arabs where 8 centuries in central and mostly south Iberian Peninsula (Andalusia). In short, if in Phoenix, AZ and in southern Argentina, the Spanish speakers don't use the /th/, it's mainly because the Arabs didn't use it.
Arabic has not one but two flavors of theta: voiced ذ and unvoiced ث
In this website http://dilap.eu/arabic-alphabet/nun.htm Everyone can hear that these two, are not like the /th/ sound of Castilian "Gracias" or English "think". However, you can find these sounds in some dialects of Andalusia.
Classical Arabic does, but most of the spoken dialects don't. Interdentals are hard for non-natives to learn and use (think of all the different ways non-theta-using Europeans say "think": "fink", "sink", "tink", etc...) I can't remember which Arabic dialects have the th-sounds intact, but Moroccan isn't one of them; neither are the main Egyptican and Levantine dialects.
So theta doesn't travel well. This may be why in a mixed population, theta would lose out over time in Spanish America.
I'd like to ask briggslaw, guest-iiowljs and R.L.G if the English (th) and the Castillian theta (ceceo) are exactly the same sound, or there could be some slight phonetic difference.
Listening to Castilians and English-speakers, I always had the impression that the English "th" was softer than the Castillian theta - the ceceo seems to have a more explosive "c" component.
The premise that because SPaniards were under Moorish rule, their language would change significantly is not proven by history. Langauges are surprising resistent to superstrative influences. This is barring actually forcing people to learn and speak a language. The Moors didn't do this, as after 800 years certain Spaniards would clearly be speaking Arabic. The Goths also ruled Spain for centuries, and there is virtually no trace of their language in Spanish, save for the -ez and -es endings (son of) of many Iberian last names.
75% andalusian immigrants? I'm afraid not; rather 75% gallegos with a very distinctive accent. A Spaniard will be called "gallego/a" in Uruguay and Argentina no matter where in Spain he or she comes from.
We should also bear in mind that the andalusian /s/ is anything but homogeneous. It has variances depending on the Region. I would also like to explore to what extent the modification of the theta sound in Latin America has occurred through the contact with the vernacular indigenous languages.
V is also pronounced like b so it is more like thibilithation
We have the same case in the Canaries, offshore Africa but Spanish territory. No theta in sight.
Implied reason for the lisp was inbreeding.
We were told in secondary school (in Oklahoma) that the Spaniards acquired the fashion from the lisps of the ruling Habsburgs.
All multi-locational Californians should be screen-named thus
So many Americans have been told that in Spain people pronounce the /th/ sound because of a King. If you really think about it, how ridiculous is that?
Wrong. The Hapsburgs were hardly foreigners in SPain. Phillip Second was as SPanish as you can get. ANyway, the whole notion of 'lisp' is a distortion. What to English speakers sounds like a lisp, ie a speech impediment, is a deliberate phonetic modulation, which by the way exists in ENglish,'soft th'. Who would say that when you pronounce "think" you are lisping. ANd the voiced or 'hard th' like "that" is also present in SPanish intervocalic D. (For whatever it's worth.)