Rules and Reasons

Background to English PronunciationBackground to English Pronunciation by Ádám Nádasdy
My rating: 5 of 5 stars

This book excited me. Hard to credit I admit, and perhaps most easily explained by the free seminars I was able to attend. It was written by a Hungarian academic who had taught my closest Hungarian friend and colleague, so the reading of it was accompanied by many long and detailed discussions in the pub.

That being said there were three other good reasons why this was such a good read for me. One, it was written for Hungarian teachers of English and so gave me good insights into both teacher training and language education in Hungary. This challenged, but ultimately affirmed the notions about language and how it should be taught that I have garnered from my own training and teaching experience. Two, it contained the most detailed information I have seen about spoken Hungarian, and that in itself was manna for a struggling student of Hungarian. And three, it added immensely to my knowledge of English speech, strangely reaffirming my beliefs both that even ordinary speech is complex and beautiful, and that poetry is the ultimate expression of language. In the rest of this review I will limit myself to a discussion of language education.

In the debate that contrasts skills with knowledge, Hungarian education is definitely in the knowledge camp. So here the trainee teacher is presented with many rules and detailed lists of exceptions. Though frequently given the advice that we can choose what we should teach, the understanding is that we should know all this ourselves. Coming initially from a more skills friendly background I was shocked that there was very little instruction of how to get the student to articulate the sounds of English. I thought better of this, remembering the Hungarians are schooled in the production of speech early in their education. Nádasdy obviously felt that this ground wasn’t worth going over again. There are plenty of ad-hoc tips on how to get Hungarians to produce English like sounds, and detailed explanations of why it is occasionally difficult for them to do so.

A more lasting worry is the presentation of language as a system of rules, and those rules as a series of binary contrasts: voiced/voiceless, soft/hard, short/long, weak/full, tense/lax, free/covered, stressed/unstressed, neutral/non-neutral. It is worth pointing out that each speaker of a language has his own realisation of that language – an idiolect. And although I am pretty much a speaker of standard English with received pronunciation, there were many of his examples which simply didn’t work for me. In the classroom too, teaching simple phonetic and intonation exercises, often results in lack of conviction on my part because the recorded examples simply don’t illustrate the teaching point they’re meant to. I’ve resorted to teaching through my own voice to avoid confusion. Corpus linguistics has shown us that language is a question of more or less rather than yes or no, teaching otherwise simply flies in the face of our own and our student’s own experience.

Likewise I object to the characterization of everyday language which doesn’t conform to the ‘rules’ as ‘exceptions’: I believe there is a reason for everything. The ‘exceptions’ really do expose the inadequacy of the ‘rules’, and not “prove” them (From the proverb which is itself misunderstood due to the change in the meaning of ‘prove’ over time). The issue of language ‘rules’ really comes to a head for me in Násdasdy’s treatment of spelling. He believes that if students manage to learn a strikingly complex set of spelling rules and lists of exceptions, it is possible for learners to predict the pronunciation of an English word from its spelling. I do not, and Nádasdy’s long account of tense and lax vowels does not convince me otherwise. If he had produced some statistics to support his arguments – e.g. the percentage of words that are actually are pronounced regularly, they might of carried some weight. He doesn’t. There is an important difference here to highlight, the difference between a rule and reason, there are reasons why English Words are spelt the way they are, and to be fair he mentions some of the most important; conservatism, foreignism and morpheme identity, but these cannot be used by learners to predict the pronunciation of the word – they don’t know how a word was pronounced in the past, or where it was borrowed from, or its morphology – this is information they don’t have. To find it out, they would have to go to a dictionary, they can usually find the pronunciation there too.

Later on, Nádasdy remarks that “advanced learners [and native speakers] are generally able to place the stress on words they have never pronounced before. Obviously they possess some knowledge which enables them to do so”. Generally, but not completely true, I still encounter new words that my intuitive pronunciation of is different to the one given in the dictionary. It may be difficult for a Hungarian to grasp, because they are given lists of rules in primary school, but native English speakers generally are not. The ‘knowledge’ Nádasdy writes of isn’t a knowledge of rules, its skill in assigning written words to models of pronunciation, these patterns I was given in primary school.

In the absence of statistics, I would hazard a guess that the more unusual a word is, the more likely it is to be spelt regularly – the converse of Nádasdy’s observation that the most common words in English are the mostly spelled irregularly. The most common words in English also have multiple pronunciations (strong and weak forms). Nádasdy describes this thoroughly without reflecting on why it should be the case. I suggest that there are two things going on here: language is not homogeneous, but an amalgam of discrete systems which have to interact for language to work. The interaction of these systems can often result in conflict – therein comes the other thing, a language has to have a system of preferences so that compromises can be made. This makes language much more of an engineering problem than a logical problem. A discussion of this sort of interaction in the context of language evolution can be found in Selfish Sounds and Linguistic Evolution.

So in any language there are lexical systems, and syntactical systems and phonological systems, and so on. For example, lexically a word consists of a particular vowel but in some utterances it is difficult to realise that vowel within the syntactical and phonological constraints that are placed on it. Sometimes it can be realised with the full vowel, sometimes with a reduced vowel. Hence the formation of strong and weak forms of the same word. Obviously, English doesn’t give a very high priority to the pronunciation of vowels, so they can be changed within a word without any loss of meaning. Regional accents are another example of this, and perhaps the low priority given to vowel production is one of the reasons for English’s international success. In Hungarian vowels are given a much higher preference, so my sloppy pronunciation of Hungarian vowels does lead to the breakdown of communication. Nádasdy is aware of the sort compromises a Hungarian speaker of English has to make in making his phonetic systems conform to English, but is often wrong about what is and what is not acceptable to other English speakers. The issue of preferences within a language is not one I’ve ever seen discussed, yet it is one that would seem to be very important to language education.

View all my reviews

  1. “Obviously, English doesn’t give a very high priority to the pronunciation of vowels, so they can be changed within a word without any loss of meaning.”

    Hmm… ‘Bad’ vs “bed”? “Man” vs “men”? 😉 “Colour” vs “collar”?

    “perhaps the low priority given to vowel production is one of the reasons for English’s international success”

    No, I think if there is any linguistic reason for that (the chief reasons are surely of different nature!) then it’s the morphological simplicity of English. Which is great for beginners … but it’s a curse for advanced learners. Learning English is really fast in the beginning and it’s really slow to progress after 10 years. On the other hand, you start slow with, say, Russian and you’re fine after 10 years.

  2. Hi Leto,

    we know each other – I used to comment on as Sophist 🙂 Your examples confirm my point. I’m thinking of spoken language not written. Listening to my students struggling with ‘a’/’e’ which is difficult for Hungarians, I realise I never misunderstand them because they mispronounce this distinction. Struggling with Hungarian, my listeners can completely fail to understand me if I make such a mistake. English’s simplified morphonolgy may play a role in this, because I think English users are forced to make more use of context to resolve the ambiguities left by our simplified morphology, which would take care of pronunciation differences too. Any examples of the problem caused to advanced learners by this?

    The first comment, thanks!

  3. Hi Sophist/Adrian,

    Ah, I certainly remember. 🙂 Even though I’m still posting on, I decided to start my own blog because unfortunately things have really deteriorated there. Please feel free to comment on my blog on Hungarian politics if you feel like to. I know for a fact I can count on your strong disagreeing with me about politics! :p

    Okay, I think “context” is the key word indeed. I’ve had a conversation with a non-native English speaker recently (about black cats 😀 ) and the word “colour” and “collar” could equally fit in a sentence she said. I was puzzled and I tried to get her pronounce the word right (regarding the differing vowel) so that I would understand what she said. I suppose you, as a native English speaker, and opposed to me, a native Hungarian speaker, would have been less confused and you would have known sooner which word she meant actually. You would have been a lot more forgiving because you had have known by that time from the broader context or you would have expected you’d know soon. I, as a native Hungarian speaker, have a shorter span of attention for context and a lower level of tolerance for losing context. Also you must keep in mind that there are a *lot, lot more* homonyms in English than in Hungarian and generally words are a lot more ambigious than in Hungarian and this also increases your tolerance for ambiguity.

    So I must revise my above statement: the English language itself does give a high priority to distinguishing vowels, just like Hungarian does. That’s in sharp contrast with say, Arabic, where vowels matter much less and nearly all information is carried by the consonants. (This fact is strongly reflected in their writing system, too.)
    However native English speakers themselves rely much more on the context, instead of vowels, to understand the other one in a conversation than native Hungarian speakers do. This is because the morphological simplicity of the English language makes them to think much more in terms of context.
    You say “Peter loves Mary” and then you may simply add stress to “Peter”, “loves” or “Mary” accordingly (you rarely make a complicated structure such as “It’s Mary whom Peter loves”) or you simply deduce what is important from the context. On the other hand Hungarian relies on morphology and all these combinations are equally possible and they slightly differ in meaning: “Péter szereti Máriát”, “Péter Máriát szereti”, “Máriát szereti Péter”, “Máriát Péter szereti”, “Szereti Máriát Péter” and “Szereti Péter Máriát”. It’s unambigious for us, without further context or additional intonation, what is emphasized in such a Hungarian sentence.
    Also, because few foreigners speak Hungarian, we’re not used at all to disregarding strong deviations in pronunciation either, unlike English speakers who are very much used to a very broad spectrum of accents. The difference between the regional accents in Western Hungary and the furthest point of Transylvania (a lot more than 1000 km!) is absolutely negligable compared to the difference of the regional accents in Bristol and Liverpool (about 300 km).

    • Well, we obviously disagree much less about language than you think we do about politics!

      There’s only one thing I would like to add, because I think it highlights my idea that there is a system of preferences in a language, rather than the idea that Hungarian is generally less ambiguous than English. There seems to be one way that Hungarian is more ambiguous than English. I often find myself lost in an English conversation with Hungarians because they haven’t sufficiently defined who or what they are talking about for English speakers.

      You’ll have to forgive my knowledge of Hungarian from here! It seems that in general Hungarian is much more relaxed about referencing the subject of a sentence. In Hungarian you don’t have to use an explicit subject in a sentence, the subject and even the object can be part of the morphology of the verb as in ‘Szeretlek.’ = ‘I love you’, or ‘Esik.’ = ‘It’s raining’. And when Hungarian does use pronouns to refer to the subject it does so more ambiguously than English: the lack of he/she distinction still throws me; polite forms use the third person pronoun rather than the second. There is the use of the third person plural as an indefinite subject “csengetnek’ = ”they’ are ringing’ rather than ‘the bell is ringing’, and constructions that omit person altogether. ‘Hová tetszik menni’ = ‘Where would you like to go’, Hungarians are very adept at resolving the ambiguities introduced by these features of Hungarian, but when they import then into English the result is similar to my poor Hungarian pronunciation!

      • Oh, I cannot see ambiguities here at all!
        Let’s see your examples one by one:

        “Szeretlek” (or “látlak”, etc.) : This is a special form of verb conjugation and the meaning is unambigious: the subject is first person singular and the object is second person (informal!, that is like “du” in German or “tu” in French or “thou” in Middle English).

        “Esik”: It’s third person singular AND “subject-conjugation” (“alanyi ragozás”, sorry I don’t know the English terminus technicus for this). People sometimes do say “esik az eső” (“the rain is falling”) but the subject (“eső”) is usually left out indeed. And the object is unambigious, the conjugation tells you the object is “something”
        Let’s take a better example (you cannot say “it rains it” in English either :p ) :
        “Néz” = s/he watches *something*. On the other hand “nézi” = s/he watches *that thing*. There’s no ambiguity about the subject at all.

        “csengetnek” uses the third person plural grammatically but it’s the general subject in fact (“you” or “one” in English, “man” in German, “on” in French) . The object is unambigious again: they are ringing *something*. If you want to say “one is ringing *the bell*” (which you know about) then you say “csengetik a csengőt”. Note the different conjugation. “csengetik” alone means “one rings *that thing*”.

        As far as “tetszik” is concerned, that’s simply another level of politeness when “Ön” is too formal and “maga” sounds rude. “tetszik” = you (formal form, that is third person singular!!) like *something*. In this case you like *going* or “to go”. The grammatical form makes the person unambigious.

      • But yes, when Hungarians map these (implicite?) *unambiguities* into English improperly then you, a native English speaker, will be confused. I, as a native Hungarian speaker, will be less confused at the same time because I know what mistake they make. 😀

  4. I’ve thought about it, and ‘ambiguity’ is the wrong word. What we are really discussing here is linguistic redundancy, and how Hungarian and English have different practices for ellipsis (reducing grammatical redundancy), and phonomorphology (reducing phonemic redundancy). This is bit beyond my comfort zone! In any case your last comment shows you’ve got my point, and shows why it would be useful to teach (if it could be presented effectively to language learners!)

Leave a Reply to Adrian Buck Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: