What is the difference between agglutination and omitting space? For example (from Wikipedia), the Turkish Çekoslovakyalılaştıramadıklarımızdanmışsınızcasına is supposed to mean "as if you were one of those whom we could not make resemble the Czechoslovakian people". Why does it have to look like one word?

Is it something like the English don't you dare bring-me-my-damn-hat-as-soon-as-possible-and-make-sure-it's-at-least-marginally-drier-than-last-time,-you-retard me ever again? By the way, if this one's not called (a form of) agglutination then what is it called (other than verbing)?

  • 5
    The concept of 'word' does not have a universal definition, but it's usually possible to segment utterances into words using language-specific morphological and phonological criteria. There are entire books (e.g. Dixon and Aikhenvald 2002) written on wordhood, which you may want to look at. – WavesWashSands Feb 4 '17 at 10:03
  • 4
  • 3
    In your English example, if it was like Turkish agglutination (which it isn't), the vowels in everyword would be determined by theword immediately before it. This is not observed anywhere in English except in certain jingles and song rhymes, like bibbity, bobbity, boo. – jlawler Feb 4 '17 at 15:47
  • 2
    Turkish agglutination involves vowel harmony. So do many other agglutinative languages. And vowel harmony is possible only inside a word, not between words the way analytic languages like English form compounds. – jlawler Feb 4 '17 at 19:07
  • 1
    Eleshar's example of the German 'Altkirchenslawisch' brings ups the question what makes that so single wordish in German, even though it is so easily separable? – Mitch Apr 5 '17 at 20:21

Agglutination is when many words/morphemes are joined into a single word. If you define "word" as "the pieces that we separate by spaces in writing", then sure, if you changed the rule for adding spaces, you could arbitrarily change what's a word. If that were all it meant to be a "word", then there'd be no difference whatsoever between agglutination and omitting spaces.

However! That's not the only definition of what a "word" is—it's not even the most useful one. Let's try to identify "words" in another way. Try figuring out what the concept might mean to someone who never learned to read. Are words distinguished by pauses in speech? Listen to someone speaking on Youtube, and you'll quickly realize that's wrong; most words are spoken in a row, one after another, and there's no obvious cue to where they start and end (try listening to a language you don't know and identify the words!). Not even the divisions we mark with commas (or other punctuation marks) imply pauses (contrary to what uninformed highschool teachers preach). So what does "word" even mean in the context of pure speech?

The answer will depend on the language (see Haspelmath, The indeterminacy of word segmentation and the nature of morphology and syntax). Let's use English as an example. In English, a nominal word may get an -s suffix to make a plural, or in some cases change vowels (foot→feet). However, a piece of a word won't change even in the plural. So:

  • ToothpasteToothpastes is a single word.
  • Tooth pasteTeeth pastes is two words.

Similarly with verbs:

  • She stir-fries (one words)
  • She stirs, fries (two word)

The difference is also marked by things like stress and pitch. Each English word can only have a single stressed syllable, so "TOOTH PASte" vs. "TOOTHpaste". I think "tooth paste" also has a lower pitch (I'm not a native speaker, sorry).

By now you should be realizing that the use of spaces in English isn't arbitrary. The orthographic words are designed to basically match speech-words as defined by tests such as the above ones. If you wrote I stir – fry the tooth paste (awful example, sorry) as Istirfrythetoothpaste, your new "orthographic word" would obscure the difference between that and I stir-fry the toothpaste (e.g. the stresses and pitch melody: "I STIR[low] FRY[high] the TOOTH PASte" vs. "I STIRfry[high-low] the TOOTHpaste").

Turkish has its own particular ways of defining "word". As jlawler has pointed above, one simple test (among several possible ones) is the phenomenon of vowel harmony. In Turkish, vowels change to match each other, so that either all vowels are "front" (iüeö) or all are "back" (ıuao)… within a single word. So if you put a few things together, and they change vowels in order to harmonize, they're a single word, and it's because of that that the orthography has them without spaces.

| improve this answer | |

There is all the difference in the world.

Omitting Space: arbitrary orthographic convention designed essentially by linguists to encode a particular language in writing; while this is likely to be based on some language features, it is extra-linguistic in the end.

Agglutination: way of deriving words from morphemes - it is based on the concept of word as the minimal lexical unit and the agglutinated morpheme having only one grammatical function. Morphemes that are agglutinated to a lexical root cannot stand by themselves and form one word with the lexical root (e.g. English urgent + ly is essentially an agglutination because -ly has one grammatical function, to convert an adjective to an adverb, and it cannot stand on its own, cannot be separated from the root by another full-fledged word etc.).

It is true that agglutinative morphemes can behave in a way suggesting their original status as a word, e.g. in a multi-word phrase, they are added only to the last element: if we have very large dogs, an agglutinative language with such a feature will have very large dog-s or dog very large-s or dog large very-s (depending on its word order constraints) but the -s (plural mark) just cannot stand on its own (form a sentence) etc.

I do not know any Turkish but note that your phrase as if you were one of those whom we could not make resemble the Czechoslovakian people contains actually very little few lexical elements:

1) Czechoslovak (this you cannot do away with).

2) people (this can be easily a grammatical morpheme, in Czech we also do not say English people or Spanish people but we have morphemes allowing us to express this in one word, like Englishmen or Spaniards).

3) resemble (this can be easily encoded by a grammatical morpheme similar to English -like as in doglike behaviour, catlike gait,...)

4) make (this is almost certainly a factitive derivation of which there is plethora in Turkish).

The rest are clearly words with grammatical meaning, not lexical, hence it can all be agglutinated to the lexical basis and exist as a single word.

But it is the decision of the powers that be to determine whether you write it all together without any separation, or whether you separate strong morphological units by, say, a dash or a dot, or whether you separate every single morpheme by a space. Compare the German and English attitude towards compound words:

English: Old Church Slavonic

German: Altkirchenslawisch

| improve this answer | |
  • The points you listed are not lexical elements in Turkish. (except the first one of course) – kabraxis Mar 10 '17 at 3:07
  • 2
    Thanks for verifying this - so clearly this is agglutination, not word compounding. – Eleshar Mar 10 '17 at 20:43

Because your english example is compound sentence. You just add different words to each other. But Turkish example is like this Çekoslavakya-lı-laş.... If we inspect this, -lı and -laş are suffixes not words. -lı creates a new word which gives the meaning of Inclusiveness (like Czechoslovakia-n). -laş is a suffix that creates verb from a noun which gives the meaning of to become that noun. This is agglutination. It is common to have one word sentences in Turkish, you can express lots of things with suffixes.

There are examples similar to your English example. Buzdolabı means fridge, literally Ice+closet. In here 2 words are added together no suffixes.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.