Are all languages ​​really equally complex?

Answer

According to a widely held hypothesis, simplicity in one component of a language’s grammar is offset by complexity in another component of the grammar. According to this hypothesis, for example, languages ​​with a simple word structure have a good chance of having a complex sound theory. However, the existence of such an inverse correlation has never been demonstrated and there is a chance that there will never even be a consensus about what exactly grammatical complexity is and how it can be measured.

However complexity is defined, it is in no way to be confused with quality, suitability or efficiency. All languages ​​are equally suitable for complex communication. Moreover, grammars are always complex systems. There is also no clear correlation between the structure of the grammar of a language and the society in which that language is spoken. Languages ​​with long written traditions spoken in industrialized areas (such as Dutch) do not differ fundamentally in communicative efficiency or grammatical complexity (however defined) from the languages ​​of, say, hunter-gatherers in the rainforest.

In your question, you note that the Amazon languages ​​Munduruku and Piraha do not have plural marking. I don’t have any grammars of these languages ​​handy here and can’t confirm if it’s correct. But plural marking is a good starting point for thinking about complexity and efficiency. In Dutch, the difference between singular and plural is compulsory. I can’t say I went to the bookshop, bought something there, and left it open whether it was one book or more than one.

I bought a book. (1)

I bought books. (2, 3, 4, …)

There are languages ​​where grammar allows it. Suppose Dutch was such a language, you could say I bought a book without specifying whether it concerns one or more books (just as you can say with substance names I drank vodka without having to say how much vodka). Such languages ​​are no less efficient. It is always possible to be more concrete I bought a lot of books, I bought three books,… This shows that mandatory plural marking is strictly speaking an unnecessary property.

Is mandatory (grammatical) plural marking more complex than optional? Perhaps, but it is just one of thousands of relevant dimensions in which languages ​​can differ, even within the domain of number marking. Here is an incomplete list:

• In addition to singular and plural, some languages ​​have other values ​​in their grammatical number marking, such as dual (two things), trialis (three things) and/or paucal (few things). You cannot say in such a language that you have bought books (plural), if only two or three were purchased.
• In some languages, plural marking is mandatory for certain nouns (typically persons and animals) and optional or nonexistent for other nouns (typically inanimate things). This is more complex than a system in which all species names behave in the same way.
• In some languages, the plural of a noun is always marked in the same way, while the form of the plural marker may vary in other languages. Those who learn Dutch must learn that the plural of apple is formed with an -s and the plural of pear with an -en. There are languages ​​that are simpler in this respect than Dutch (one marker for all words), but also languages ​​that are (much) more complex.

So the question when comparing languages ​​is not what can be said but rather what should be said (and how is it expressed). With a final example we can return to the grammatical differences between Piraha and Dutch. In Dutch you can say Mira is pregnant without indicating how you got that information or how reliable she is. Amazonian languages, on the other hand, often have a grammatical category of evidentiality, which makes it mandatory to mark in a verb form whether something is “hearsay” or “seen for yourself”. In this respect, the grammar of Piraha may be more complex than that of Dutch, but we do not want to deduce from this that Piraha in general is more complex.

Assuming that the complexity of languages ​​could be measured and compared in a meaningful way, then in my opinion there is no reason to assume a priori that all languages ​​are equally complex. Judging by intuition, I wouldn’t bet on Greek in such a case, and certainly not on Chinese as the most complex language. The “winner” will more likely be Caucasian, or North American, or perhaps Southern African.

This is a recent scientific publication that addresses your question:

Shosted, Ryan K. (2006) Correlating complexity: a typological approach. in Linguistic Typology 10:1-40.