What difference does language make?

I didn’t write this, but I am thrilled to be able to share it with you.

Careful communication builds trust. My popular internal training course on effective writing spends a considerable amount of time discussing how to build empathy for the audience. It’s not just a matter of staying truthful and legal; it’s about connecting with a reader in a different place and time. The premise of the course is that good writing informs, persuades or entertains, or possibly all three. But until now, I’ve not included any discussion on language as a social tool. That is, we communicate not just to impart raw information, but to build bonds within our peer groups, to identify as a member of a community, to establish an identity and social standing.

I recently ran across this excellent essay on a customer’s intranet site, written as a personal blog, entitled On being a sherpa: what difference does language make? The essay’s author uses his professional expertise in linguistics to demonstrate that careless communication can inadvertently exclude people, marking them as outsiders. Language conveys more than is said; what is left unsaid can have derogatory and hurtful implications. A careful writer will be sensitive to these.

The essay explores linguistic concepts such as categorisation and conceptual frames; prototypes and stereotypes; essentialising and othering; implicature; marked language; and ownership of terminology. I am delighted that both the author and his employer agreed to let me share this work freely, but without identifying either of them (since this is considered a personal opinion piece and not a publication of that organisation).

I’d consider this to be essential reading for anyone who wants to be more careful and inclusive in the way they communicate. For my own part, I’ve included some of the core concepts into a whole new section of my course, on careful and inclusive communication.

On being a sherpa: what difference does language make?

Sherpa

What meaning do you take from the use of the term “sherpa” in a sentence like this?:

New Delhi, Feb 26 (PTI) BRICS Sherpas and Sous-Sherpas held their first meeting under India’s chairship, during which the country presented its priorities in 2021 under the theme — “BRICS at 15”.

Outlookindia, February 27th 2021

Your mental concept of “sherpa” in this sense may include such things as “being an official representative of some important person”, “being someone who does the hard and detailed work so that the person they represent doesn’t have to”, “being someone who leads the way and guides”. That sort of thing. Wikipedia has its own entry for the term “sherpa”:

A sherpa[1] is the personal representative of a head of state or government who prepares an international summit, particularly the annual G7 and G20 summits. Between the G7 summits there are multiple sherpa conferences where possible agreements are laid out. This reduces the amount of time and resources required at the negotiations of the heads of state at the final summit. The name sherpa—without further context—refers to sherpas for the G7 summit, but the designation can be extended to different regular conferences where the participation of the head of state is required. The sherpa is generally quite influential, although they do not have the authority to make a final decision about any given agreement.”

Wikipedia, Sherpa

We have inherited a false belief, an assumption, that words have hard and fast definitions that can be found in dictionaries. But there are few words, if any, that can be defined in this way. This is because concepts are in human minds, they do not consist of “necessary and sufficient conditions” as traditional semantics would hold. Dictionary “definitions” are just attempts at describing our concepts.

Giles Fauconnier says, “Language forms do not ‘carry’ information; they latch on to rich pre-existent networks in the subjects’ brains and trigger massive sequential and parallel activations.” (Fauconnier in Horn and Ward, 2004:658). Wilson and Sperber comment, “the encoded conceptual address is merely a point of access to an ordered array of encyclopaedic assumptions from which the hearer is expected to select.” (Wilson and Sperber in Horn and Ward 2004:619).

The point is that the meaning of any term, including “sherpa”, is a matter of what concept is salient in the minds of both speakers and hearers. One aspect of our concept of “sherpa” is that it entails some kind of subservient relationship of assisting someone else, so from the article The problem with calling your career coach a sherpa, it is apparent that the idea of assistance is in focus:

Hannah Morgan has been known as the Career Sherpa since 2008. “One reason the sherpa term has become hip is because it sounds less arrogant than expert or guru. And it sounds more unique than ‘guide,'” said Morgan.

QZ.com, The problem with calling your career coach a sherpa

For Morgan, finding a term that entails the idea of assistance and help towards someone important rather than entailing superiority or seniority is important. So, although this makes it an attractive term for Morgan, it is precisely this entailment that makes this modern use of the term problematic for those who belong to the Sherpa nation.

The Sherpas are an ethnic group in Nepal, with their own language. Because of their location, they have an ability to cope with altitude and members of their nation are employed by the numerous mountaineering expeditions in Nepal. Perhaps the Sherpa best known in the West is Tenzing Norgay, or Sherpa Tenzing, who was the first person to climb Mount Everest along with New Zealander Edmund Hillary.

Meanings, prototypes and metaphors

To understand what is going on with terms like these and with how language can be used to perpetuate stereotypes and inequalities it helps to have a better understanding of how communication works and what language is. This is not as simple as it sounds and has bedevilled philosophers for generations. It would be arrogant to claim that we understand everything now, but what has been revealed in research in cognitive science over the past few decades goes a long way to explaining what is going on when we use language – why it is not just about information but about people and relationships too. In fact, we are coming to understand that the people and relationships side of things is a more fundamental aspect of language than the information side.

We have looked a little at the notion of the entailments of concepts, and briefly at the idea that words aren’t about simple definitions of things referred to but are tools that prompt us to access relevant information from our massive encyclopaedic knowledge and experience. But there are other aspects of how we think that are important to how our use of language affects our relationships.

One innate cognitive strategy of our minds is categorisation. We could not think if our minds simply saw things as raw sensory inputs. To function in the world we need to be able to make sense of the sensory data we receive and process it into meaningful entities and events.

But it’s not just a matter of categorising things, we also need to make use of conceptual frames in order to survive. The mind has to be able to determine, for example, that the patterns of light reaching the retina represent what we categorise as an instance of a bull standing a certain distance away. But we also need some very important knowledge about bulls. Our knowledge frame for “bull” might include such things as “very powerful”, “dangerous”, “unpredictable”, “territorial”, “charges at people and gores them when angered”, as well as other information like “eats grass”, “is kept for breeding”, “for being eaten as steak”, etc.

George Lakoff, one of the pioneers in the field of cognitive linguistics, explored the importance of categorisation in his book Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. (The title derives from one of the noun categories in the Australian language Dyirbal). Some scholars, notably Eleanor Rosch and George Lakoff, have been looking into “prototypicality effects” in concepts. The idea is that there are good exemplars and less central exemplars of concepts. Typically, a sparrow or blackbird may be a prototypical example of a bird, while a penguin or a cassowary less so. Our concept of “sherpa”, even if we are unaware that it originally refers to someone from a particular ethnic group, may well include prototypically the idea of an Asian person, resiliently used to work at high altitudes in the Himalayas, who guides and porters loads for others.

So, rather than our brains having clear cut, dictionary-style definitions of concepts, they work subconsciously using prototypical examples. This enables us to use those concepts for reasoning, inferences and metaphorical extensions. “Sherpa”, when used to mean “an emissary” or a “career coach”, is being used metaphorically, alluding to aspects of guiding and “doing the hard work” for someone else.

Our brains work very efficiently when we can make short cuts, whether we are thinking about bulls, Italian meals, civil servants or sherpas. Prototypicality of concepts helps us think and make inferences fast. But part of that process is that we also stereotype our concepts, making use of a pared down version of them. This is problematic when it comes to how we relate to other people; specifically to groups of other people; the process that helps us achieve efficiency of thought also leads us to stereotype and essentialise groups of people.

Essentialising

Once we have realised that language is not simply a matter of “word-to-world” correspondence and the “truth conditions” of propositions, and have understood that it is first and foremost a social activity, it makes sense that the social and relational imports of what we say are the most important. So, when we stereotype groups of people – genders, races, ethnic groups, social classes, etc., it is not just a cognitive thing but a social thing. This stereotyping of groups of people, that such and such a group are inherently “like this”, is essentialising them, reducing them to some essence.

As we are social beings, we have “in-groups” and “out-groups”. We have “us” and “them”. We see our own group as basically normal and other groups as “different”. Because of this assumption of our own normality, we tend to see others, even those we admire for some trait or other, as being defective or odd in some way. This way of viewing outsiders to our own group is known as othering.

This concern over essentialising is seen quite acutely in the names and associated mascots and logos of sports teams like the Washington Redskins (American) Football team. Similar debates are ongoing with the Exeter Chiefs rugby team. The Washington team is changing its name (temporarily The Washington Football Team) under pressure from Native American groups. In some ways the name Redskins represents approval; it focusses on bravery and combat prowess, as well as providing increased profile to Native Americans. However, the problem is that it essentialises Native Americans, suggesting that fighting and courage are all they are about, nothing much else. It is this reductive essentialising that appears to be what is hurtful. The Exeter Chiefs plan to retain the name although they have decommissioned their mascot “Big Chief” – a caricature of a Native American.

Essentialising is the issue behind the offence of the comment to someone of South Asian heritage, “Have you got a nice curry in your lunch box?” British people generally enjoy curry greatly, so on the surface it is a positive comment. But the problem is that it reduces and stereotypes people of South Asian descent, not allowing them to be seen as people in their own right with all the individual and cultural complexity that involves.

This seems to be part of what is going on with the use of the term “sherpa”. From the article on being a career-coach sherpa:

What sounds unusual or noteworthy to consultants and coaches sounds disrespectful to some Sherpas who object to their heritage being appropriated as a branding tool or title. “It’s quite insensitive. I want people, especially in the West, to realize that sherpa is not a profession. It is an ethnic group with a rich cultural heritage,” says Dawa Futi Sherpa, a college administrator who moved from Kathmandu to New York City six years ago.

Note that Dawa Futi mentions “a rich cultural heritage”. One of the problems with essentialising is that it represents an impoverished version of people and cultures.

Implied meaning

It is difficult to understand what is going on in language or how the language we use can be problematic if we stay with the outdated and deficient view of language as containing or encoding the meaning in communication. Seeing communication as a social-inferential process helps us to understand what is going on when people interpret what others say and the language they use.

The philosopher Paul Grice coined the term implicature for meaning that is implied in communication rather than stated. An example he gave was a reference for an applicant for an academic post in philosophy. What do you think of this reference for the job?:

Dear Sir, Mr X’s command of English is excellent, and his attendance at tutorials has been regular. Yours, etc.

You can see that as a reference for a job in teaching philosophy it communicates by what is not said, not by what is. So Grice said that since the applicant’s command of English and regular attendance were the only positive things that could be said, the implicature was that he was not a good candidate for the post.

The meaning communicated is not in the words but is implied. We are so used to communicating in this way that we scarcely notice what we are doing and that communication is implied as much as said. Communication works by combining utterances with contextual information. Take for example:

Shona: Has anyone seen George?

Brian: There’s a red Fiesta in the car park.

If we take Brian’s answer as being relevant to Shona’s question, and we would assume its relevance (unless we think Brian hasn’t heard her or is deliberately seeking to ignore her), then the process of deriving relevant implicatures from his answer goes something like this:

  • Implied assumption 1: George drives a red Fiesta
  • Implied assumption 2: If George works in the same office, a red Fiesta is likely to be his
  • Implied assumption 3: George is likely to be somewhere in the vicinity of his car
  • Implied conclusion 1: Brian does not know exactly where George is (otherwise he would have provided a more informative answer)
  • Implied conclusion 2: Shona should be able to find George somewhere nearby

Note Shona’s initial question was not simply a request for information as to whether anyone (anyone, anywhere, ever) had seen George, but an implied expression of a desire to speak with George.

Note also that even if Shona didn’t know beforehand that George drives a red Fiesta, she can supply that assumption. She does so because she assumes that Brian’s answer will be relevant to her question. Human cognition and human communication depend on the search for relevance. This is why anthropologist Dan Sperber and Linguist Deirdre Wilson focussed centrally on this concept in their 1986 book Relevance: Communication and Cognition.

The point is that not only is communication about the communication of implied meaning as much as stated meaning, but that it relies on the accessing of assumptions from context. Communication depends on implied assumptions, on presuppositions.

But there is more than one sort of implied meaning. Imagine the conversation between Shona and Brian went like this:

Shona: Has anyone seen George?

Brian: That slob is too much of a party animal to be in this early.

Brian gets Shona to access the assumption that George is unlikely to be in yet and derive the conclusion that no one at the office will yet have seen him. So, essentially he is saying “No, I haven’t seen him and I don’t think other people will have either.” But his indirect reply is more relevant than a direct one would have been because he simultaneously communicates the reason he doesn’t expect Shona to find George yet.

But Brian is simultaneously communicating other (higher-level) implicatures. He is communicating an implicature of disapproval of George and his lifestyle. He could have said “He” or “George” or used other terms in place of “That slob”. The reference is the same but the attitudinal stance is different and, along with “too much” “animal” and “this early” implies Brian’s opinions of George and some things about what his own lifestyle values might be. By all our stylistic choices in discourse we imply our attitudes, relationships and values.

When someone of Asian heritage is asked “Where are you from?” and they reply “Solihull”, and the question comes back “Where are you really from?” what are some of the implied assumptions present in the question? Some possible assumptions might be:

  • People who are from Solihull/Britain are white
  • People of Asian heritage are not British
  • Having a different skin colour or different name lumps you together with “others”
  • Your skin colour is your major defining characteristic

The question may be meant in a friendly engaging way and there is something positive about finding out about each others’ backgrounds. But it is important to be aware of the implied assumptions in communication. Some of these assumptions may be unwelcome if they essentialise the person, reducing their individuality or if they communicate othering.

Another example cited was an Asian woman being told “Ooh, I do like your costume!” The woman saying it was being appreciative and it is positive when people want to chat rather than maintaining an embarrassed barrier of silence. However, this comment was uncomfortable for the addressee. It relates to the higher-level implicatures derived from the term “costume”. The term carries the assumption that the addressee is somehow exotic and other.

So, it is crucial to understand that communication is not just in the words but is a social activity that proceeds by inferring meaning from the context as well as what is said, and the context includes who is saying it, when and why. It is also crucial to understand that implied meanings may be derived not by what was intended to be communicated by the speaker but by the assumptions underlying what they say.

Critical Discourse Analysis – the establishment and maintenance of beliefs

One field of discourse analysis that investigates how implied meanings establish and perpetuate power relations is Critical Discourse Analysis (CDA). Strauss and Feiz describe CDA in this way:

Critical Discourse Analysis … centers on the view that all of discourse is social practice. … Through discourse, we formulate belief systems of class, society and nation; of right and wrong; of beauty and health and gender; even of how language should be used. Critical Discourse Analysis seeks to uncover the discursive processes through which ideologies are shaped and communicated, normalized and propagated – ideologies which involve hidden dimensions of power, control, injustice and inequity, all of which typically go unseen and unnoticed because they are couched in what appear to be common-sense assumptions of social reality and “truth.”

Strauss and Feiz, 2014:313

The term “discourse” refers to any piece of communication, however short or long. Note the mention of the perpetuation of ideologies going “unseen and unnoticed”, which relates to the fact that communication and the inferences we draw are a matter of implied meaning as much as stated meaning.

We saw that the career coach Hannah Morgan said, “One reason the sherpa term has become hip is because it sounds less arrogant than expert or guru.” In this she pays attention to the (unnoticed) implied aspect of the term being “less arrogant”, and this is because our prototypical concept of a sherpa is one of serving someone else.

In addition, the prototypical concept of sherpa has a Nepali person employed to work for and fulfil the aims of a Western, Caucasian mountaineer. In this way it helps perpetuate the post-colonial assumption of white, Western, European supremacy. That is not what Hannah Morgan is thinking of, nor are those who use the term in the context of G7 or G20 summits, but by its use as a concept in discourse it perpetuates the belief that in the world as we normally see it Westerners are those in charge.

Marked and unmarked language

Another aspect of the implied nature of the communication of social meanings is the question of markedness. Marked and unmarked categories were part of linguistic analysis in the past 50 years, as linguists sought to analyse language in terms of binary features. This was done extensively in phonology and syntax but also in semantics. Dale Spender, in her book on the male orientation of the English language Man Made Language, raises this question of marked and unmarked categories citing the work of Julia Stanley (1975). Stanley notes that the well-known linguist Geoffrey Leech, in developing a set of categories for English had explicitly used “plus male and minus male to distinguish masculine from feminine”. Spender (1985:19, 20) quotes Stanley:

In the case of gender, [minus male] must be the significant feature of girl and woman, because females are defined traditionally as “non males” since males are the standard of comparison for the entire species, and women are the beings who contrast with them.

Stanley, 1975:29

This is why, Spender notes, again quoting Stanley:

“When a woman becomes a professional in one of the fields usually reserved for males,” says Stanley, “she does not move into the corresponding semantic space covered by the noun conventionally used as its label.” Instead, she must signify that the norm, the positive, does not apply and so she becomes a lady doctor, a female surgeon, a woman lawyer, or else, in less prestigious occupations, a waitress, a stewardess, a majorette.

Spender, 1985:20

Spender quotes as an example part of a transcript from a discussion of women writers at a feminist conference “Women in Literature” in Bristol, 1976:

It’s useless trying to say I’m a writer … and a good one. I nearly said “as good as a male”. And that’s what I’m talking about. By definition you can’t be a good female writer, it’s a contradiction of terms. And the more you try to establish yourself as a writer the more you have to move towards being “as good as a male”. That’s exactly what I want to get away from. What happens if you are as good as a female? It’s laughable isn’t it. It pisses me off … [mimics] “Excuse me, I want a job on your paper. I’m an excellent female writer. I have all the female virtues … in abundance. I’m silly, irrational, irresponsible …” etc., you know the rest. You just can’t capitalize on being female. That way no good lies, you have to show that you have male virtues, and then, of course, you’re trapped. Because you are not a male! You’re a substitute male.’

Some progress has been made since the 1970s in this aspect of gendered roles and language regarding professions, but there is considerable progress still to be made, and language retains its implicit biases even when society tries to move on. As with terms and expressions such as the title of Spender’s book man-made, and with mankind, manning or man the fort, workmen, postman, chairman, etc., the intention of people using these and similar expressions is not usually to mean men only but people in general, but even so these expressions assume and perpetuate implicitly the unspoken ideology that males are the unmarked category – the default – and women are the marked category.

There has been some discussion recently of the use of “guys” and “girls”. Some women find being addressed as “guys” not offensive, while being addressed as “girls” appears demeaning and patronising. For some men, addressing a mixed group as “guys” may be a move to de-sexualise and de-gender their interaction. The term “guys” can be used in either a male or a neutral sense, while the term “girls” is specifically female (and if used for men is seen as insulting, being associated with particular assumed traits of girls). That “guys” can be used in a neutral sense exemplifies the same marked and unmarked terminology in which “man” is the unmarked, default, term and “woman” is the marked term.

The pejorative (and sometimes sexualised) connotation of female terms such as “girls” is something else that Spender mentions – that male terms tend to retain original and neutral meanings while equivalent female terms are frequently pejorative or sexualised, for example, master and mistress, bachelor and spinster, sir and madam, courtier and courtesan, governor and governess, etc. With all of these terms the female term can be used in a diminished, pejorative or sexualised sense or associated with negative attributes. (Spender 1985:16-18).

The question of markedness arises also with regard to race. In British English the underlying assumption is that “white” is the unmarked category, while “non-white” is the marked category. It would be unusual for this reason to hear people talking about “black” and “non-black” people. It is also why there are frequent attempts to find a term to talk about “non-whites”, the acronym BAME being one of the more recent attempts. (In this case the ‘ME’ – ‘Minority Ethnic’ extends the marked category to include people of marked ethnicity even if their skin colour is unmarked). One problem with a term like this is that it perpetuates the unmarkedness of whiteness (British and other “domestic” origins of people that are frequently found on race and ethnicity data forms), and the markedness of non-whiteness.

This markedness and unmarkedness is behind the discussions of white privilege in which white people are frequently heard to say that they have not experienced privilege. They have not experienced or noticed it precisely because it is unseen, unmarked. Those who are able to carry on their lives as members of the default group do not have to “make up for” their appearance, to cope with the stereotyping and disapproval of others, or to ‘be twice as good as the white people’, or to watch what they wear and how they speak, how they congregate and where they go in order to avoid negative responses from society, authorities and institutions. White privilege is belonging to the unseen, unmarked group, perceived as “normal”. It is true that race and gender are not the only dimensions of markedness and unmarkedness, the same also applies to such things as sexual orientation, disabilities, class and accents, to name a few. The privilege of unmarkedness is characterised by its invisibility.

Ownership

Given that language is so inherently a social activity, it is not surprising that it is not only important what people are saying and how they are saying it, but also who is saying it.

In terms of language use we all belong to communities of practice (COPs). That is, as well as speaking languages and dialects and using a variety of speech registers, we belong to a range of social groups who will use language in particular ways. The same person may interact with other legal professionals in a judicial context, with their fellow football team supporters in a pub, with other members of a mothers and toddlers group, etc. For each of these contexts there is a community of practice – a way of speaking and interacting, involving not just vocabulary but also conventions of formality / informality, ways of negotiating identities, roles and status, strategies for persuasion, politeness, etc.

Any particular community of practice will have its own ways of referring to its own people (in-group) and to outsiders (out-group). So, for example, the term “sherpa”, as we have seen, is used extensively in communities of practice in government and business, as well as in career-coaching contexts. But there is also an important sense in which the members of the group being referred to have a right to say what they are called in public discourse. So, it makes sense that Dawa Futi Sherpa can object to the use of “sherpa” to mean “assistant / guide / porter”. There is a sense that the Sherpa nation have ownership of the term.

In one episode of the TV programme The Good Fight, one of the black characters was being brought before HR for his use of the “N word”. Another black character in the programme made the statement “we own that word.” That ownership means that the black community is licenced to use the word in ways that other communities cannot. This is because language is not primarily about conveying information; language is about the negotiation of identities, social acts, values and relationships. A member of an out-group using the N word implies different perspectives from a member of the in-group using the term. 

From that perspective it is important that the terms we use to refer to other communities and what we say about them are in line with how they would like to be referred to and how they would like to be characterised. It makes sense that Native Americans can “own” what is said about “Redskins” and how they are characterised, and for Sherpas to say how their ethnic name is used. This is natural in the context that language is not about disembodied meanings but is a social phenomenon.

Conclusion

These notes are not intended to criticise anyone nor to make anyone afraid to speak to others from different ethnic or race backgrounds. It is vital, whatever our background, to keep talking with other people. That is the basis from which we can understand and empathise with each other as people. But these notes are meant to explore how research in pragmatics, discourse and cognitive linguistics helps us understand the real nature of communication and what lies behind the words we use and the things we say.

References

Fauconnier, G. (2004). ‘Pragmatics and Cognitive Linguistics’. In Horn, L. R. and Ward, G. (Eds.). The Handbook of Pragmatics. Oxford: Blackwell.

Grice, H. P. (1975). ‘Logic and Conversation’. In P. Cole and J. L. Morgan (Eds.), Syntax and semantics, Vol. 3: Speech Acts. New York: Academic Press.

Horn, L. R. and Ward, G. (Eds.). (2004). The Handbook of Pragmatics. Oxford: Blackwell.

Lakoff, G. 1987. Women, Fire, and Dangerous Things: What Categories Reveal about the Mind. Chicago: University of Chicago Press.

Leech, G. 1968. Towards a Semantic Description of English. Indiana University Press.

Spender, D. 1985. Man Made Language. (1st edition 1980). London: Routledge and Kegan Paul.

Sperber, D., and Wilson, D. 1995. Relevance: Communication and Cognition. (2nd Edition. 1st Edition 1986). Oxford: Blackwell.

Stanley, J. 1975. “Sexist grammar”. Paper presented to South Eastern Conference of Linguistics, Atlanta, Georgia, 7 November.

Strauss and Feiz. 2014. Discourse Analysis: Putting our Worlds into Words. Abingdon: Routledge.

Wilson, D., and Sperber, D. 2012. “Relevance Theory”. In Horn, L. R. and Ward, G. (Eds.). The Handbook of Pragmatics. Oxford: Blackwell.

By David A

Imagination of a poet - communication skills of an engineer. In no particular order, I like old cinemas, Japanese gardens, playing music, giraffes, technology, films, ducks, narrow Italian streets, and trying to write stuff that's interesting to people. I've been blogging irregularly since 2005 and have had some form of website since 1998.