An Introduction to Communicative Efficiency

Written by: Natalia Levshina


For a long time, linguists have thought of language as a tool for thinking. Under this view, how we use language for communication is not particularly interesting because it does not tell us anything about the ‘core’, ‘inherent’ properties of language.

Nowadays, many language scientists argue that communication is an important factor that explains why languages are the way they are. I am one of those people. I believe that communicative pressures, alongside other cognitive and biological factors, are responsible for what our language is like and what it will look like in the future.

One of the communicative pressures is the preference of human beings for efficient behaviour. Like all animals, humans try to save effort and time whenever they can. Language is no exception. Speakers, signers and their addressees tend to save time and effort required for articulation and processing. In my book I argue that this ‘laziness’ has massive repercussions for our use of language and that the usage preferences can become part of language structure with time.

But how to be efficient and save your and others’ time and effort? This book gives you three basic rules:

  1. Don’t spend time and effort on information that does not bring any communicative benefits;
  2. Don’t spend time and effort on information that is highly accessible (that is, predictable, stereotypical, provided in context, and so on);
  3. Place words that are semantically and grammatically related as closely as possible.

The first principle is illustrated by the omission of the agent in passive constructions when it is irrelevant who performed the action. For instance, if you say, “I was advised to book the hotel in advance”, it is not important who gave the advice. Unimportant information is omitted.

To follow the second principle, you should omit obvious or highly predictable information. For example, under normal circumstances, it is more natural to say, “The criminal has been arrested” than “The criminal has been arrested by the police” because arresting criminals is the prerogative of the police.

Finally, the third principle is illustrated by the fact that speakers of English and other languages tend to put long phrases after short ones if they are preceded by the same verb. For example, it is more natural to say “I’m reading now your recent article on communicative efficiency, which I find really interesting” than “I’m reading your recent article on communicative efficiency, which I find really interesting, now”. The problem with the latter version is that the distance between the verb “reading” and the adverb “now” is very long. By the time we hear “now”, we may have difficulty remembering the verb it depends on.   Psycholinguists have shown that long dependency distances create extra memory costs for sentence processing.

These principles may seem obvious, but they have profound consequences for language. In my book I argue that many statistical universals, or common properties of the world’s languages, can be explained by these principles. For example, if a language has different singular and plural forms for nouns, the singular form is usually shorter than the plural form. Compare book (singular) and books (plural). It is more efficient to use the longer form like books to refer to several things or persons because we more often speak about one thing or person. At the same time, if some nouns in a language have a special marking when they represent single objects, like Welsh aderyn “bird” – adar “birds/flock of birds”, these nouns are usually more frequently mentioned in the plural meaning.

Unfortunately, most of the grammar rules and words evolved long time ago, so it is difficult to find data to trace back their origins. But sometimes we can actually observe the principles of efficient communication in action. For example, when a new thing becomes widely discussed, its name can become shorter. An illustration is the COVID-19 pandemic. The official name of the disease is “Coronavirus Disease 2019”. As the disease became increasingly frequently mentioned on media and in private conversations, it was shortened to “COVID-19”, “coronavirus”, or even “corona” in some languages. The champions in saving effort are probably Australians, who call it “rona”.

But even if we lack the data about the origin of some efficient constructions, there is a solution. We can use experiments for modelling language evolution in the lab. In such experiments, participants learn an artificial language, which has some amount of free variation between efficient and inefficient grammatical or lexical patterns. Again and again, we see that participants tend to prefer the efficient structures or words, even though they have never encountered them in real life. We can then extrapolate these results to existing structures in real languages. I provide examples from others’ work and my own experiments in the book.

My firm belief is, however, that we can only get a clear picture of language evolution if we use different sources of data: typological, corpus-based and experimental. I implement this approach in two case studies. One is about differential argument marking, as in Spanish, where animate definite objects are marked with a preposition “a”, and inanimate ones are not. Spanish object marking illustrates the cross-linguistic tendency to mark animate and/or definite (or specific) objects. The other case study focuses on causative constructions, such as make X do Y, and the famous correlation between directness of causation and formal compactness of causative constructions. The three types of evidence presented in these case studies point in one direction, demonstrating that efficiency is responsible for the general patterns that we observe in the world’s languages.

Finally, I want to warn the reader against a potential misunderstanding. In no way does this book advertise a vulgar utilitarian approach to language. It would be wrong to save costs by all means and in all situations. Sometimes it is worth spending more energy and time on communication. A prime example is artistic expression. Professional ballet dancers move in a very energy-costly way, but their efforts result in a wonderful aesthetic experience. Or take Yoda, a character from the Star Wars universe. As many readers know, Yoda uses strange word order. For example, “Hard to see, the dark side is” or “Named must your fear be before banish it you can.” As I show in my book, this word order is more difficult to process than the normal English order, due to longer dependency distances between related words. Yoda’s word order is therefore inefficient. But it also makes Yoda memorable and instantly recognizable all around the world (along with his distinct green ears, of course).

Communicative Efficiency Language Structure and Use by Natalia Levshina

Enjoyed reading this article? Share it today:

About the Author: Natalia Levshina

Natalia Levshina is a Postdoctoral researcher at the Neurobiology of Language department, Max Planck Institute for Psycholinguistics. She is the author of the best-selling book How to Do Linguistics with R (2015)...

View the Author profile >

Latest Comments

Have your say!