logo ELT Concourse teacher training
Concourse 2

Language, Thought and Culture


Everything that we have so far seen to be true of language points to the fact that it is the most significant and colossal work that the human spirit has evolved—nothing short of a finished form of expression for all communicable experience. ... Language is the most massive and inclusive art we know, a mountainous and anonymous work of unconscious generations.
Edward Sapir, 1921:104

Philosophy may seem somewhat removed from the day-to-day concerns of language teachers.  However, if you have ever been faced with learners who simply cannot manage to grasp a concept in English or you have encountered a concept in a language you are learning which runs so counter to your own way of thinking that it seems impossible to comprehend, you may not be alone in thinking that language and mental processes, cognition, if you will, are intimately connected.
That they are connected is not in much doubt, the uncertainty lies in exactly how and to what extent they are connected.
If you have come here hoping for the definitive answer to that question, you may well be disappointed.

two schools

Two schools of thought

There are two ways of looking at the connection between thought and language.

  1. Language determines thought.  The language we speak and the concepts we are able to use to think about the world and those in it is the determining factor.
  2. Thought exists independently of language.  Humans think and their ability to do so does not depend on the language they speak but is expressed separately in 'thought language', sometimes referred to as 'mentalese'.

Like most dichotomies, this one is quite possibly false.  It is, however, a place to start.  There are two outcomes for language teachers already.

  1. If hypothesis 1. is true, learning a new language means, in part at least, learning new ways to conceptualise the world and new ways to think.  The more remote the language you are learning is from the structures, lexis and cultural background of your first language(s), the more difficult and extreme will be job of grasping new ways to think about the world.
  2. If hypothesis 2. is true, then both teachers and learners share a common way of thinking about the world which is independent of the languages they speak.  All that is needed is to learn how to express universal thoughts in another language.


The direction of causality

As is often the case, there are proponents of both hypotheses about the connection between conceptual categories and the language(s) one speaks.  Proponents tend, as is the way of things, to be somewhat extreme, denying any internal and universal grammatical and lexical systems on one hand (hypothesis 2.) or averring that all thought is determined by language (hypothesis 1.).
The issue can be summarised as either:

  1. The structure of human thought determines how a language develops and is structured
  2. The structure of language determines human thought

You don't have to choose.  It may be the case that both hypotheses are correct.  On one hand there are universals of thought that determine, e.g., that all languages distinguish between noun and verb phrases and so on and that any language which evolves determines the way people think about the world and that, in turn, determines how they can think about the world.



Theorists fall into two main camps:

  1. Those who accept that language wholly or partially determines how we think: no language = no thought
  2. Those who see language as emerging from the basic structures of the human mind: language structure = inherent mental structure.

They cannot both be right.  Here are some ideas from both camps, divided into the lingualists who take a language-first view of human thought and the cognitivists (often simply termed Chomskyists) who take the second view.

  1. Language determines thought
    1. The Sapir-Whorf hypothesis.  This is something of a misnomer in that Edward Sapir and Benjamin Whorf did not collaborate on any publication and were not working together.  The hypothesis was named by a student of Sapir.
      One quotation will suffice to get a feeling for what was said:

      We dissect nature along lines laid down by our native language. The categories and types that we isolate from the world of phenomena we do not find there because they stare every observer in the face; on the contrary, the world is presented in a kaleidoscope flux of impressions which has to be organized by our minds—and this means largely by the linguistic systems of our minds. We cut nature up, organize it into concepts, and ascribe significances as we do, largely because we are parties to an agreement to organize it in this way—an agreement that holds throughout our speech community and is codified in the patterns of our language.
      Whorf and Carroll, 1956:213

    2. Vygotsky:
      Lev Vygotsky was an early advocate of language determining thought.  Here's what he said:

      Thought is not merely expressed in words; it comes into existence through them.
      Vygotsky, 1986:214
      Consciousness is reflected in a word as the sun in a drop of water. A word relates to consciousness as a living cell relates to a whole organism, as an atom relates to the universe. A word is a microcosm of human consciousness.
      Op cit.:256

    3. Other writers:
      Others, non-professional linguists for the most part, have had similar thoughts on the topic.  Here's a bit from Percy Bysshe Shelley in Prometheus Unbound:
      He gave man speech, and speech created thought, Which is the measure of the universe
      Sapir and Whorf were not, of course, the first to make a connection between language and thought, the first determining the second.  They were following a long tradition.  Herbert von Humboldt, a German Romantic philosopher stated in 1820 that:
      The diversity of languages is not a diversity of signs and sounds but a diversity of views of the world.
      but this was in an effort to assert that western languages were superior and that led to a superior social condition.
    4. Ubiquity:
      There are many, usually not professional or informed linguists for whom the connection between thought and language appears so obvious that no support for the theory is offered.  The literature, both lay and academic, it littered with assumptions such as

      Each language is a way of thinking, expressing and perceiving the world.
      (Parara, 2021)

  2. Thought determines language
    1. Chomsky was a very influential thinker who asserted (among much else) that language and language structure springs from the innate, biologically determined nature of the human mind.  In this view, the nature of language is dependent on the inherent structure of the mind.  Hence, the positing of a universal grammar which encompasses all human languages and grows out of the structure of the mind.
      In that we have deliberately used the word human to classify language.  It is interesting to speculate what a non-human language might be like.  There's an article on this site doing just that.  Get it here.
    2. Stephen Pinker is another influential thinker in the same school and he stated:

      The idea that thought is the same thing as language is an example of what can be called a conventional absurdity: a statement that goes against all common sense but that everyone believes because they dimly recall having heard it somewhere and because it is so pregnant with implications.
      Pinker, 2007:58-59

      Earlier in the same text, he has:

      there is no scientific evidence that languages dramatically shape their speakers' ways of thinking. ... The idea that language shapes thinking seemed plausible when scientists were in the dark about how thinking works or even how to study it. Now that cognitive scientists know how to think about thinking, there is less of a temptation to equate it with language just because words are more palpable than thoughts. By understanding why linguistic determinism is wrong, we will be in a better position to understand how language itself works
      Op Cit.:57

      Elsewhere (p82), Pinker puts the argument for mentalese, this way:

      Knowing a language, then, is knowing how to translate mentalese into strings of words and vice versa.

      The counter argument is that some languages do not have what many would consider really fundamental categories such as blue, leg, if and here.

The two standpoints outlined here are not merely of theoretical interest.  In particular, many of the arguments around the use of what is called gendered language in English start from an a priori assumption that the structures of a language affect how people think about male and female social roles and, moreover, that it is possible to change how people think by reforming the language they are compelled to use.
There is a guide on this site, linked below, to how languages handle gender marking.

A further consideration concerns the sociolinguistic concepts of restricted and elaborated codes.  These concepts of we owe to Bernstein (1971), who identified two varieties of language available to its speakers:

  1. Elaborated code is used most frequently in formal situations and contains:
    1. a high proportion of subordinate clauses, using, for example, conjunctions such as although, so, because, in order to and conjuncts such as accordingly, subsequently, therefore etc.
      The use of such devices is generally terms hypotaxis (from the Greek meaning to arrange below).
    2. passive verb clauses using, for example:
          She was arrested
      instead of
          The police took her away
    3. longer and more unusual adjective phrases such as the ancient, British people, a difficult but important issue etc.
    4. less common adverbs, adverb phrases and adverbials such as regrettably, to my astonishment, not wholly surprisingly, without much difficulty etc.
    5. the use of abstract nominalisation such as creation, emergency, issues, calculations and so on.
  2. Restricted code is used in less formal settings and is characterised by:
    1. The use of simpler, and often repeated coordinating rather than subordinating conjunctions such as and, then, but etc and conjuncts such as first, next, later etc.
      This is often referred to as the use of parataxis (from the Greek meaning to arrange side by side).
    2. The avoidance of passive structures and a consequent increase in the use of subject pronouns such as he, she, they etc.
    3. The use of tag questions to elicit agreement such as wouldn't they?, didn't we? etc.
    4. Simple adjectives and the avoidance of long adjective phrases so we find more occurrences of, e.g., nice, lovely, pretty, good, bad etc. rather than pleasing, beautiful, stunning, acceptable, disrespectful etc.
    5. the use of a greater number of concrete nominalisations such as road, car, house, room etc.

More controversially, Bernstein and others have proposed that the ability to use either code depending on the appropriacy of the setting and the needs of the moment is restricted to middle-class children whereas working-class children are confined in general to the use of restricted code and are therefore educationally and career-wise disadvantaged because elaborated code structures and lexis are required to encode complex relationships.  In that view, working-class children suffer from cognitive deficiency because they are unable to think about more abstract and difficult ideas.
This is a view most strongly held by those who believe that thought is dependent on language but on that hypothesis, the jury is still considering its verdict.
This assertion has been disputed by, e.g., Labov, who presented evidence that even in restricted code the encoding of complex and subtle ideas is perfectly possible, indeed, routinely achieved and no cognitive deficiency can be identified.  In this view, elaborated code use is simply a social convention and has nothing to do with restrictions in cognitive abilities.
It will not have escaped your notice the the characteristics of restricted code are often noticeable in the speech of learners of English.


Culture: a third possibility (or an added complication)

To add to the mix, many theorists claim that the environment in which a language is born, evolves and changes is determined by the cultural needs of the speakers.  This is particularly, but not solely, relevant to the study of lexis.
The theory here is that a language evolves and adapts to the exigencies of the environment and culture in which it has developed and encodes not all possible thought but those which its speakers need to survive and prosper.  Changes in location and ways of surviving will be mirrored in changes in the language.
Here are five areas where there is evidence of the truth of this proposition.



If you are born into a culture in which kinship relationships are particularly important, it will be unsurprising if your first language has an elaborate and complex way of expressing familial relationships.  For example:

Although all languages are capable in theory of expressing any thought, the constraints produced by language in terms of kinship are clear.  Most speakers of English, for example, are unconcerned whether someone's sister is older or younger than the person in question but this is clearly significant information in other cultures and reflected in the language forms.



Evidence for culturally determined names for animals comes preponderantly from languages spoken is societies which continue to have or had a strong relationship with nature and animal husbandry.

Again, the conclusion has been reached by some that the language a speech community uses develops to include concepts which are important at the time and loses those which are less significant as social changes come into effect.
Unless one is a horse breeder it is unlikely in modern English-speaking settings that there is a need to distinguish between colt to describe a young male horses, foal for a horse of either sex under one year old, a yearling, a filly, a mare, a stallion or a gelding.
Such was not the case, arguably, before the advent of transport systems unreliant on horses.


Maps in the mind

This does not refer to the way that people draw complex diagrams as a way of understanding and remembering facts.  It is to do with how we conceptualise distances, relative positions and directions of travel.  In other words, we are considering maps in the mind.
Here are some examples:

If this is true, far from mental images being expressed in language or language determining mental images, the effect of culture and the language which arose to operate within it has been to change cognitive processes directly.  All the examples above imply that before a speaker of a language can use concepts such as these he or she is forced to consider a range of different variables.

If, as some assert, thought is independent of language, it becomes difficult to reconcile how different ways of expressing simple concepts have originated and persist.



There is some agreement that humans have an innate number sense.  Evidence for this is the fact that even very small babies show surprise when two puppets are shown, another added and then only one is revealed at the end, running counter to the assumption that 2 + 1 = 3.
Most large, especially western, languages use a base 10 to calculate so, for example, 9 + 3 is determined as 10 + 2 or 12.
(Non-mathematicians may like to know that most of us count in base 10 which means simply that as soon as one gets above 9, it is necessary to shift to the next unit up and begin again from 1.  Hence, 8 + 8 is perceived as 8 + 2, making 10 and then the remainder (6) being placed in the empty column resulting in 16.  If people had 6 rather than 10 fingers we might use base 6, in which case 8 would be express as 12 (we have filled the left-hand slot when we get to 6 and placed a 1 in the left-hand slot and the remainder (2) in the right-hand slot.  So the sum 8 + 8 will be the same as imagining 12 + 12 and the result will be 24.
As it turns out, this is not just an exercise in mathematics because some Niger-Congo languages and some from Papua New Guinea do, in fact, have base-6 mathematics systems.
Those old enough to remember the time when a shilling in Britain (and elsewhere) was divided into 12 pennies will be familiar with base-12 mathematics in which, for example, a sum such as:
    Add 1 shilling and 10 pence to 2 shillings and 4 pence (in old notation, 1s.10d + 2s.4d)
means filling the first, right-hand slot when 12 is reached and placing 1 in the next slot to the left so we get:
    3 shillings (1 + 2)
    plus 14 pence (10 + 4) = 1 shilling (12 pence) and a remainder of 2 pence
    add 1 more shilling to the left-hand slot
    place the 2 pence in the right-hand slot
The answer is, therefore, 4 shillings (the original 1 + 2 with the extra 1) and the remainder (2 pence).  4 shillings and 2 pence.
That would have been written as 4/2 or 4s.2d.)

There are other possibilities and evidence for the effects of language on mathematical ability:

Counting systems would, on the face of it, provide strong evidence for the way in which linguistic constraints impact the ability to think in certain ways.
It has been averred, for example, that children educated in Britain prior to the decimalisation of the currency in 1971 were more comfortable operating in base 12 (pennies in a shilling) and base 20 (shillings in a pound) and would find adding 7 shillings and 10 pence to 3 pounds, 15 shillings and 4 pence quite a simple task beyond similar children used only to dealing with base 10 mathematics.  (The answer = 4 pounds, 3 shillings and 2 pence, or £4.3s.2d, by the way.)
Even today, British people are probably more comfortable operating in base 3 (feet and yards) and some to working with base 16 (ounces and pounds) than others from base-10 measuring systems.  This is, arguably, an effect of language on thought.


Keeping track of time

Humans have developed ways of keeping track of time for at least 40,000 years and quite likely longer.  How they have done it varies from time to time and culture to culture and this may influence how people perceive time and how they cut it up into manageable chunks.  That, of course, will have implications for things like tense structures and other ways in which verb forms express time.

people in different cultures or groups have been shown to differ in whether they think of time as stationary or moving, limited or open-ended, horizontal or vertical, oriented from left to right, right to left, front to back, back to front, east to west, and so on.
Fuhrman, et al, 2011:1306

Grasping the ways in which the language you are learning divides time and sets events and states in time past, present or future is not easy.  It often requires a different way to conceptualise events and states.
The problems that many learners encounter with English temporal concepts often arise from the ways in which aspect and relative forms are used in English.  This is especially true for speakers of languages which do not focus at all on aspect and rely entirely on context and co-text to signal perfect, progressive, iterative and other aspects.



It was argued at the outset that the hypotheses one accepts will affect how one views the process of acquiring or learning a new language.  It is either learning how to express universal ideas in a new language or it is a process of learning to think in new ways because the language one is learning deals differently with the world.
It need not be dichotomised in such simple terms, fond as theorists are of creating unsustainable either-or distinctions.

The extreme form of the Sapir-Whorf hypothesis has probably been abandoned for good but the jury is still out on whether and to what extent language exerts constraints on the ways in which the people who speak them think.  Despite Pinker's dismissal of the Sapir-Whorf hypothesis as "wrong, all wrong" (op cit:57), there is evidence, not least from the sadly diminishing number of indigenous languages around the world, that the ways one's language encapsulates real-world states and events really does enhance or diminish one's ability to do certain types of thinking.
One alternative view is that languages reflect the things that their speakers need to talk about and as new needs arise, language alters to allow speech to reflect a changed reality.

Here are some examples of some serious implications for language learning and using:

  1. Terminology
    If, for example, my language can encode a male domesticated reindeer in its third year and first mating season, but not yet ready for mating in a single word, döngür, as the Tofa language of Siberia allows me to do (Harrison, op cit:57), do I have a better way to think about reindeer of this sort than the clumsy circumlocution in English does not allow?
    On the other hand, if my language lacks a way of expressing numbers above 10 (or in extreme cases, numbers greater than two) is my ability to do mathematics severely limited by the fact?  If language affects mathematical abilities across cultures, it must be because we have learned it that way, not because language ability is innate.
  2. Concepts
    One of the most difficult systems for speakers of some languages to acquire in English is, notoriously, the article system.  Those who suggest language and thought are separate domains would aver that this is simply because the system is difficult to learn yet speakers of, e.g., German have far fewer difficulties than speakers of Russian.  Those who see clear connections between language and thought might respond that it is precisely because some languages do not require a speaker to consider the notions of specificity and indefiniteness that lie at the heart of the English article meaning system so they need to learn to think differently about the world to get it right.  That is not easy.
  3. Classification
    In a connected domain, it is difficult for English speakers to acquire the complex system of classifiers that exist in many languages simply because English speakers are not habituated to think about the physical and social characteristics of what they are counting when they say, e.g., ten people or ten tables.
    On the other hand, English partitive expressions such:
        a rasher of bacon
        a slice of bread
        a grain of sand
        a speck of dust

        a heap of work
    become more accessible to speakers whose languages classify extensively.
  4. Number and case
    If your language demands that you consider the number of a noun and its function grammatically in a clause (as subject, object, addressee, location etc.) before you select the right forms, you are obliged to consider the world very differently from someone whose language demands no such choices.
    The misuse of pronouns is an obvious consequence as is the loss of focus on word ordering which, in a language such as English, often determines which is the subject and which the object of a verb.
  5. Countability
    If, as in English, you are required to consider whether a noun is to be understood as a mass concept or a countable one before you can select the correct determiner (e.g., a few vs. a little) you will have difficulties doing so if your own language makes no such distinction.
  6. Gender
    Many languages have gender systems which often determine how other structures agree with the noun in question.  French has two and these are reflected, for example, in adjective endings and the forms of determiners such as the articles.  Some languages have as many as five or more separate genders and not all distinguish masculine from feminine from neuter.  There is some evidence that languages such as French and German that arbitrarily classify, e.g., the noun fork as feminine will apply adjectives to an example of it differently from those whose languages make it masculine or genderless (Spanish and English, for example).
    Referring to a fork as she is a natural consequence of thinking of it as feminine and, maybe, calling the moon handsome rather than beautiful may be another consequence of referring to it as masculine.
  7. Metaphor
    Some metaphorical uses in English are so common as to pass almost unnoticed but there is ample evidence that parallel notions are not universal in all languages.  English, for example, often equates:
    argument with warfare so we:
        shoot down arguments
        attack weaknesses
        defend ideas

    life with a journey so we:
        are on the road to success
        getting on with life
        making onward progress

    For more in this area, see the guide to synonymy and related concepts, linked below.

On the other hand, there is ample evidence to suggest that humans are genetically equipped to learn language (any language) and that babies do not learn by imitation or trial and error alone.  Word and phrase categories are often cited in this respect as elements of a universal grammar and the notions of noun phrase, verb phrase and so on are probably truly universal.  However, some languages (such as Moken, an Austronesian language) seem to get by happily without anything like an adjective phrase as speakers of other languages would understand the item.


In the classroom

What if all the hypotheses are right (or none of them)?  Because the jury is still out (and may remain so for some time) when trying to reach a verdict on whether:

  1. Language determines thought
  2. Language influences thought
  3. Thought determines language
  4. Language categories and forms are innate
  5. Languages share a common universal grammar

and allied questions, it would be perilous to ignore the classroom consequences.

  1. If hypothesis a. is correct, we need to do more than concept check (because checking a concept your learners do not have and can probably only grasp with a great deal of effort and exposure is possibly futile).  It seems sensible to suggest, instead, that we supply a great deal of real-life communicative practice to help our learners internalise concepts that are outside their current mental schemata.
  2. If this weaker form, hypothesis b. is correct, then the same considerations apply but concept checking becomes more valuable because an influence, however strong, can be overcome with some effort and application in a way that a truly determining factor cannot.
  3. If hypotheses c., d. and e. are correct (even partially), then we need to single out the innate categories of universal grammar and the human-wide ways of conceptualising the universe for special attention.  This may mean, for example, focusing on the universal concepts of verb, noun, preposition and adjective phrases (unless, of course, one's learners happen to be speakers of languages like Moken which does not have the last of these).
  4. If hypothesis e. is correct then Pinker is right when he states, following Chomsky:

    According to Chomsky, a visiting Martian scientist would surely conclude that aside from their mutually unintelligible vocabularies, Earthlings speak a single language
    Op cit: 232

    However, studies in comparative linguistics reveal that the existence of language universals is questionable.
    Evans & Levinson put it this way:

    Languages are much more diverse in structure than cognitive scientists generally appreciate. A widespread assumption among cognitive scientists, growing out of the generative tradition in linguistics, is that all languages are English-like, but with different sound systems and vocabularies. The true picture is very different: languages differ so fundamentally from one another at every level of description (sound, grammar, lexicon, meaning) that it is very hard to find any single structural property they share. The claims of Universal Grammar, we will argue, are either empirically false, unfalsifiable, or misleading in that they refer to tendencies rather than strict universals.
    Evans & Levinson, 2009:2



So, is it a matter of Take your pick?
To some extent, yes.  If, for example

  1. You are teaching a multilingual group of learners whose first languages are very diverse, you may well take the view that focusing on language difference will only be partially relevant part of the time.  In this case, you may be better advised to focus on elements of supposedly universal grammar which you believe will be commonalities in the group.  That will include, inter alia, a focus on word and phrase class, case and so on.
  2. You are teaching a mono-lingual group or a group of learners whose languages share characteristics (such as Indo-European languages) you could take the view that a focus on particular ways of thinking and language differences is a productive way to proceed.  This may, for example, take the form of a focus on how English conceptualises time, space and number as well as on issues such as deixis.

This is, some would say, an empirical approach that needs to be taken because there is, as yet, no consensus concerning which view(s) of the connections between language, thought and culture is/are correct.
Others might see it as unprincipled eclecticism, of course.

Related guides
semantics for more background on the nature of meaning
deixis for considerations of how space and time are codified in English
gender for the guide to how languages handle gender and how gender marking is sometimes avoided and sometimes unavoidable in English
synonymy for a guide which also considers how simile and metaphor operate in English
types of languages for some considerations of other language factors, particularly stress and word ordering, which may influence how people can learn English (or any other language)
interference and facilitation this guide considers some of the pedagogical implications of language differences
Chomsky for a guide to his major ideas: transformational generative grammar, the language acquisition device and universal grammar as well as some counter arguments
learning style and culture for more on how cultural aspects may affect responses to learning environments and procedures
language and society this is a guide to the main aspects of studies in sociolinguistics which attempt to describe and explain varieties of use
the evolution of language this is a guide to theories about how the ability to use language (not a language) evolved in humans
How to speak to an alien an article on xenolinguistics

Bernstein, B, 1971, Class, Codes and Control, Volume I, London & New York: Routledge, Taylor & Francis Group
Dowker, A and Roberts, M, 2015, Does the transparency of the counting system affect children's numerical abilities?, Frontiers in Psychology available from https://www.frontiersin.org/articles/10.3389/fpsyg.2015.00945/full
Evans, N, & Levinson, S, 2009, The Myth of Language Universals: Language diversity and its importance for cognitive science, in Behavioral and Brain Sciences, Cambridge: Cambridge University Press
Fuhrman, O, McCormick, K, Chen, E, Jiang, H, Shu, D, Mao, S & Boroditskya, L, 2011, How Linguistic and Cultural Forces Shape Conceptions of Time: English and Mandarin Time in 3D, Cognitive Science 35 1305–1328
Harrison, KD, 2007, When Languages Die: the extinction of the world's languages and the erosion of human knowledge, Oxford: Oxford University Press
Merriam-Webster Dictionary online: https://www.merriam-webster.com/dictionary/zero
Miura, IT, Okamoto, Y, Chungsoon C, Steere, M, Fayol, M, 2016, First graders' cognitive representation of number and understanding of place value: Cross-national comparisons: France, Japan, Korea, Sweden, and the United States, Journal of Educational Psychology: American Psychological Association, available at: https://psycnet.apa.org/record/1993-23565-001
Parara, P, 2021, The Substance of the Greek Language is More Than Just Words, https://www.ekathimerini.com/262132/article/ekathimerini/life/the-substance-of-the-greek-language-more-than-just-words [retrieved 10/02/2021]
Pinker, S, 2007, The Language Instinct, New York, NY: Harper Perennial Modern Classics
Sapir, E, 1921, Language: an Introduction to the Study of Speech, New York: Harcourt Brace
Vygotsky, L, 1986, Thought and Language, Cambridge, Mass.: Massachusetts Institute of Technology Press
Whorf, BL & Carroll, JB, 1956, Language, Thought, and Reality: Selected Writings of Benjamin Lee Whorf, Cambridge, Mass.: Technology Press of Massachusetts Institute of Technology