How to categorize languages based on how they use morphemes to form words
Languages create words using morphemes in drastically different ways. Let's explore.
In undergrad, to fulfill the elective credits needed for my linguistics degree, I could take ten credits of a non-European language. I chose Mandarin.
For two hours five days a week I would sit in my intensive beginner’s Mandarin course with an aching jaw from pronouncing all the new sounds and a headache from trying to remember which tone to use.
My clearest memory is of the professor hearing me read from the textbook. She would try to stifle a giggle. But it would always turn into a full-blown laugh. “Bù hǎo (not good),” she would say, and then offer the correct pronunciation. To my untuned ear, the correction always sounded exactly like I said it. Needless to say, speaking Mandarin was tough.
Surprisingly, though, writing Mandarin was less tough. Sure, I would have to remember all the strokes needed to write the characters (at my university, they taught us the traditional script which has far more complicated characters than the simplified script). But the words and the grammar made sense. They made more sense than any other language I’d ever learned.
Want to say wine? Just put one word which means grape and another word which means liquor. No need to memorize a whole new word.
And it turns out, many languages in the Sino-Tibetan language family that Mandarin belongs to form words in a similar way.
That’s our topic this week. We’re going to learn how some languages, like Mandarin, tend to have one morpheme per word, which is very different from languages, like Spanish, which have several morphemes per word. Let’s learn how to categorize languages based on how they create words using morphemes.
Morphological Typology
Language typology refers to categorizing languages based on their structural features. Language families is one type of typology. Another type is called morphological typology. It means exactly what it sounds like. It deals with categorizing languages based on how they form words out of morphemes. There’s two main language groups: analytic and synthetic. But no language is fully analytic, nor is any language fully synthetic. Thus, it’s best to picture morphological typology as a spectrum or continuum.
Analytic Languages
Analytic languages are also known as isolating languages. In these languages, the morphemes are isolated from one another because each morpheme equals one word.
Sound familiar? Mandarin is an example of one.
In these languages, there tend to be little-to-no inflectional morphemes. Rather, the grammatical relationship between words is shown with morphemes, which form their own words.
For example, in Mandarin, instead of having an inflectional morpheme to show that a noun is plural (like plural -s in English), the language has counting words.
You don’t say “two pens,” you say “two [counting word] pen.”
The grammar is not embedded into the nouns or verbs. So, you need additional ways of expressing the same thing.
If you’re telling a story in the past tense, you don’t need to conjugate the verbs. Rather, you can make the context explicit by using an adverbial, like “yesterday” or “last week.” There are also separate morphemes which can be added to indicate the tense.
Is English an analytic language?
There’s debate whether English is analytic or synthetic, but there’s a pretty solid argument that English is on the analytic side of the spectrum. Like we saw last week, English only has eight inflectional morphemes to show the grammatical relationship between words. Across all languages, that’s quite few.
Synthetic Languages
The other main category of morphological typology is synthetic languages. These languages include several morphemes into each word. But within the category of synthetic languages, there are two sub-categories: agglutinative and fusional.
Agglutinative Languages
Agglutinative languages are similar to analytic languages in that there is a clear separation between morphemes. Each morpheme has one job and it’s clear what that job is. The big difference between agglutinative and analytic languages is that the morphemes are all part of one word in agglutinative languages. In other words, the morphemes are glued together to form words.
Languages in the following families tend to be agglutinative: Turkic, Japonic, Dravidian, Bantu.
Let’s look at an example from Turkish. To say something that takes several words to say in English, like “of the villages,” you just need one word in Turkish.
That word, though, has clear morphemes (they’re clear if you know the language). If we look at the gloss of the word below, we see that “köy” means village, “ler” means plural, and “in” means it’s in the genitive case (possessive). You can clearly see that each morpheme has one grammatical meaning and that the boundary between morphemes is clear-cut.
köy-ler-in
village-PL-GEN
‘of the villages’
Fusional Languages
Fusional languages are similar to agglutinative languages in that a single word may contain several morphemes. The key difference is that in agglutinative languages the boundary between the morphemes is easy to spot. In fusional languages, though, this is not always the case.
Languages in the following families tend to be fusional: Indo-European, Semitic, and some Uralic languages.
In these languages, there are plenty of bound morphemes that help show the grammatical relationship between words. The boundary between the bound morpheme and the root is often difficult to distinguish. Additionally, each morpheme might be taking up several jobs.
Let’s take, for example, an irregular verb in Spanish, like tener.
yo tengo
‘I have’
tú tienes
‘you have’
él/ella/usted tiene
‘he, she, you (formal) has’
It’s hard to find the root tener is any of these conjugations. (Is it somewhere in ten? Or maybe tien?)
In a word like comí, in Spanish, the second morpheme is doing a lot of the heavy work, and you can’t disentangle the singular morpheme from the past morpheme because it’s all embedded in í.
comí
eat-FIRST PERSON-SG-PAST
‘I ate’
Discussion
There are additional categorizations of languages, like polysynthetic languages which have no free morphemes. But I hope this newsletter gave you an idea of how differently different languages form words with morphemes.
I would love to know (you can comment below or reply to the email): which languages do you speak? Based on this newsletter, do you think they might be analytic or synthetic? Why?
Keep Learning
I used these videos to refresh my memory on morphological typology:
This video gives a good overview of morphology and morphological typology.
This video inspired the drawings.
This video provided the Turkish example.