Does the pressure to be right or wrong inhibit academic enquiry?

by Craig Evans

Not too long ago, I found myself caught up in a Twitter exchange after retweeting a link to what I thought to be an interesting article in National Geographic. The title of the article, repeated word for word in the tweet, posed the question: Does geography influence how a language sounds? In response, a fellow twitterer answered simply ‘no’, before then helpfully providing a link to an article that presented a comprehensive counter-argument to that represented in the first article.


Without wanting to go into the relative merits of specific arguments, the general point of contention centred around a study that concludes there is a relationship between certain articulatory sounds and high altitude. This study had been carried out by the anthropological linguist Caleb Everett, recounted favourably in National Geographic, but then heavily criticised in the second article, published in Geocurrents. In the latter, titled ‘Ejectives, High Altitudes, and Grandiose Linguistic Hypotheses’, the authors pick through Everett’s various claims in what seems to be part of a concerted effort to completely discredit his hypothesis.


Now, it is not my intention here to support or oppose the particular arguments of Everett or his detractors. As an undergraduate student, I am eager to wade into any debate on any subject, but I am also very mindful of my lack of knowledge and limited experience. What I find most interesting about the articles and the Twitter exchange is the extent to which there often seems to be no middle ground in academia: you must be either completely for or against whatever the given theory.


At least this was the strong impression made on me by my fellow twitterer’s next response to my original retweet, which was to suggest that if I was out to confirm public misconceptions then I would in the end be proven wrong.


Yikes! This seemed a bit strong. All I had done was link to what sounded like a very interesting proposition about the possible relationship between geography and the sound of language. Does saying here’s something interesting amount to a complete endorsement of everything contained therein? In my mind it doesn’t, but in the glare of this apparent accusation, I suddenly found myself feeling the need to defend my position in relation to a subject that I’d read about for the first time only moments earlier.


It can’t be a good thing, this atmosphere of absolutes where if something is not completely proven right then it must be wrong. Even the authors of the Geocurrents article betray this kind of attitude by referring dismissively to Everett’s hypothesis as ‘grandiose’. Is it really grandiose to propose that a relationship may exist between geography and the sound of language? Irrespective of whether this is borne out in the data selected, there is still plenty of mystery about the origin of a wide range of distinctly different sounding languages to merit the exploration of new theories. The problem is, to do so is to run the risk of disreputation and not being taken seriously.


Of course, there is a flip-side to all this, and I can partly empathise with my Twitter ‘accuser’ and Everett’s detractors. Academic rigour is vital for getting to the truth of why things are the way they are. Should legitimate doubts exist about a well-publicised study, then it is important that this study is challenged. In the case of the National Geographic article, Everett’s work is discussed without alternative viewpoints being represented, even though the Geocurrents article (published 3 days later) demonstrates that these obviously exist. This does not, however, provide grounds to completely dismiss Everett’s hypothesis. If anything, it is the National Geographic that should be censured for not seeking alternative expert opinions in order to provide a more balanced journalistic account of the study.


As for Everett’s hypothesis, it seems to me that he has come up with an original and thoroughly interesting line of enquiry for which he deserves credit. Such imaginative thinking shouldn’t be scoffed at, but encouraged; and if the data gathered to support such a hypothesis proves insufficient or inconclusive, then does it completely negate what still seems a very reasonable proposition?


Ultimately, the question all academics need to ask themselves is: is it better to always be cautious and right, or to sometimes strive for something original at the risk of being wrong?


  1. Very good post, Craig – and I agree – there’s too much 100% right vs. 100% wrong out there.

    I think some of what is behind it is that people feel provoked when colleagues make ‘big’ claims, that are maybe not always proven or not always easy to prove (cf. my post about Chen’s economics/linguistics link earlier this year). This is nothing new. Also in the past, people like Joseph Greenberg (suggesting that most languages of the Americas are related) have received very harsh criticism from their peers.

    It is interesting, by the way, to see how some critics of Dan Everett have jumped straight into criticising Caleb Everett (who is Dan’s son).

    I wonder if there is too much focus about criticism and much less about collaboration, seeing the good points and taking them further.

    1. Thanks Jeanette.

      I definitely think that more concession needs to be given to the potential of good ideas. At the moment it seems that as soon as a particular hypothesis has been falsified, then this implies an invalidation of the theories behind it, which definitely feels very stifling.

  2. Dear Craig,
    I am one of the authors of the response in GeoCurrents (not Geocentrics!) that you refer to in your post. Thank you for your interest in this controversial topic and in our site. I am sorry if you feel caught in the crossfire, especially as it involves a subject that you are only learning about. Hopefully, this experience serves as a good learning experience for you.
    I agree with you that “academic rigour is vital for getting to the truth of why things are the way they are”. I disagree, however, with your characterization of our response as dismissive towards Everett’s idea. Rather than “scoff” at Everett’s article or wave it aside, we took it seriously and analyzed it in great detail. It is exactly “academic rigour” that we called for and tried to apply to Everett’s research. The reason that we called Everett’s hypothesis “grandiose” is because it is a new spin of the good ol’ geographic determinism theory. This is also the reason we are questioning whether Everett’s basic idea (the ejective-altitude correlation/causation) is “reasonable” and worthy of further exploration. I am also going to disagree with your opposition between “always being cautious and right” and “the risk of being wrong”. Any scientific hypothesis runs the risk of being “wrong”—an unfalsifiable statement is not a scientific hypothesis, by definition. Everett’s theory is falsifiable—and we showed just a few ways to falsify it.
    I am also glad that the previous commentator brought up both the “Amerind wars” and the debate regarding Chen’s work on the putative correlation between grammatical tense and saving habits (you might find my response to it instructive: The two debates illustrates the two ends of the “productiveness scale”, on which Everett’s work sits somewhere in the middle. The Amerind debate was extremely productive in as much as it propelled the study of Native American languages to new heights. The same can be said about the “Generative Semantics Wars” of the 1960s and 1970s, and many other debates in linguistics. These debates contribute to our understanding of the field by raising issues that are based on the shared accumulated knowledge base. The problem with Chen’s study and, sadly, numerous others like it is that they are ignorant of the very basics of language. Instead of contributing to progress in linguistics, they hamper it.
    I wish you best of luck in your studies and hope that you retain both the courage and the humility that you expressed so well in your post.
    Asya Pereltsvaig

    1. Thank you Asya for your comment, and my apologies for typing ‘GeoCurrents’ incorrectly.

      I had not meant to cast any doubt over the ‘academic rigour’ with which you approached Everett’s research; on the contrary, it is something that seemed to be applied very methodically in your article.

      Your response to my post has certainly given me food for thought.

      That said, at the risk of sounding very naive, it does seem to me that rigidly scientific methodologies have far-reaching theoretical implications that extend beyond what is proved or disproved. I mean this in the sense that insufficient or inappropriate data may lead to the falsification of a statement, but I don’t see why this should necessarily mean the complete invalidation of a hypothesis.

      In respect to the relationship between altitude and language, although the specific ejective-altitude correlation may be disproved, this doesn’t alter the fact that it is still an area that merits further enquiry and investigation. For example, speaking broadly, there is a correlation between altitude and breathing, and of course breathing is integral to speech. Given this, it does seem that correlations between altitude and language probably do exist, albeit not necessarily in relation to the sound of language.

      Perhaps what I’m feeling is that interesting and relevant ideas such as Everett’s, one that has certainly engaged this student, deserve at least some credit, regardless of whether or not they fall short in the analysis.

      All that said, I really do appreciate you taking the time to respond to my post, and also for your kind remarks.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s