Human languages are full of puzzles. For example, why do languages that have different forms for singular and plural nouns usually have shorter singular forms and longer plural forms for most nouns, like flower – flowers in English? Why are you more likely to say, give me my money than give my money to me? Why do languages like Latin or Czech, which have flexible word order, also tend to have case marking? For example, the Latin word liber “book” will have different forms depending on whether you say “The book is interesting” (liber), “I read a book” (librum) or “I found it in a book” (libro).

Efficiency is the key
Some linguists have argued that the answers to these and many other questions have to do with the human preference for efficient behaviour. All animals save effort and time whenever they can, and humans are no exception. We try to save time and effort required for language production and processing. This is the topic of my recent book, “Communicative Efficiency: Language Structure and Use”, which has been recently published by Cambridge University Press.

In this book I gathered many examples that illustrate our drive towards efficiency. For example, Zipf formulated the law of abbreviation. According to this law, frequent words tend to be shorter than rare words. One mechanism that accounts for this law is that we often reduce the words that we use frequently. That is why, for example, we say “corona” in Dutch, German and other languages instead of “coronavirus” (see this blog post for more details).

We also omit obvious and well-known information whenever we can. For example, a chef in a restaurant is more likely to say to his or her workers, “Cut the meat” than “Cut the meat with a knife”, or “Boil the water” than “Boil the water on a stove”. (Of course, I do not mean an extremely posh restaurant where nothing is done the normal way.)

Three principles of efficiency
Based on these, and many other examples from different languages, I formulate three principles. The first is that you should not spend time and effort on information that is highly accessible. By accessible I mean predictable, stereotypical, provided in context, and so on. For example, when ordering drinks in a bar, you are much more likely to say “Another beer, please!”, than “I would like you to give me one more glass of beer of the same kind, and I promise to give you money after I drink it”. All this information is highly accessible and should be omitted.

The second principle is not to spend time and effort on information that does not bring any communicative benefits. This is self-evident. If you say, “I was told not to submit my papers to that predatory journal”, it is not really important also to specify who told you to avoid the journal. Very little is lost if this information is omitted.

The third core principle has to do with the order of meaningful elements in speech and writing. Words that are related in meaning or grammatically should be placed as closely as possible. These preferences percolate into grammar and become rules. This is why it is grammatical to say, “The shrewd detective found out who had committed the crime”, and not “Out the shrewd who the committed detective found crime had”. This allows us to save cognitive costs in both production and comprehension.

Survival of the fittest expressions
The evolution of language patterns is driven by efficiency. Consider a phenomenon known as differential marking. A famous example can be found in Spanish. If you want to say in Spanish, “Pepe sees the film”, you should not use any prepositions before the word “film”: Pepe ve la película “. But if you want to express “Pepe sees the actress”, you should add the preposition a before “the actress”: Pepe ve a la actriz. Similar splits are found in other human languages. In fact, this seems to be a language universal.

How can we explain this tendency? One of the hypotheses is that differential marking develops for the purposes of efficient disambiguation. If the thing we talk about is inanimate like “the film”, it is very likely to be the Patient, or the Object. No extra marking is needed. But if it is animate, it is more likely to be the Agent, or the Subject. The marker helps to avoid potential confusion.

But how do we know that this is the way differential marking evolved? Similar to archaeologists, we need to sieve through many historical sources to get a picture of the evolution. But the data are hard to find, especially for languages without a writing tradition. Thankfully, we can model language evolution in the lab. In such experiments, participants learn an artificial ‘alien’ language and then use it for some tasks, for example, describing and guessing pictures.

In fact, artificial language experiments show that humans reproduce differential marking in the lab. Learners of an alien language tend to use the marker to avoid ambiguity and override the wrong expectations about the role of words in a sentence. As a result, a more efficient linguistic system emerges.

To conclude, I believe that our ‘laziness’ has massive repercussions for language. Even subtle usage preferences that help us save time and effort can gradually become part of language structure. Efficiency is one of key factors that explain why human languages are the way they are.