Scientists don’t get a lot of positive feedback for their work: Often it’s just two or three questions after a conference talk, by friendly colleagues who understood the talk only partly – and all this after months of work that went into this talk. And reviewers of journal papers are often downright negative – getting one’s journal-paper reviews back can be a depressing experience.
But we all need positive experiences, and when we publish a book, we hope for a lot of praise. And indeed, most book reviews are very positive, it seems to me. But as the pace of life increases with the internet, it seems that there is less and less time for books and book reviews, so I thought that published paper reviews might be a good idea as well. It’s not an established genre, but over the past few years, I published quite a few of them (e.g. on papers by Jeff Good, Peter Bakker and colleagues, Nikolaus Himmelmann, Fauconnier & Verstraete), as did some others on the same blog. Most of these were fairly positive, though of course I always had some questions and some criticism.
But more recently, I started writing a series of negative review blogposts, e.g. a review of a typological paper on the coexpression of complementizers, and a review of a paper on possessive/genitive marking.
These are not the first negative published reviews – I have some earlier negative book reviews, on a 1998 book by Frederick Newmeyer on formal vs. functional linguistics, and on a book by David Lightfoot on language change. And recently I wrote a very critical review of Mark Baker’s book “Case”. Moreover, I am planning to publish a negative review of Koeneman & Zeijlstra’s textbook “Syntax” (on my blog), and a fairly negative review of the “Oxford handbook of polysynthesis” (edited by Nick Evans, Michael Fortescue and Marianne Mithun).
What’s the point of all this? Wouldn’t it be nicer for everyone if we only praised our colleagues where praise is due, and otherwise ignored their work? What about the positive feedback that we all need? Who gains from negative reviews?
I find it painful to make negative comments, whether on students’ or on colleagues’ work (and definitely more than having others make negative comments on my work), and it makes me much happier when I see work that advances our science which allows me to praise them. So why do I do it anyway?
I think it’s because of my concern for the unity of science – there can only be one truth, and we have to find out eventually to what extent the approaches are complementary or dead ends. I only write negative reviews of books or papers adopting prominent approaches (mostly, but not exclusively of the generative tradition) which I think cannot be ignored. If an author at the fringes of the discipline claims that Basque is related to Nubian, or makes claims about borrowings from Chinese into English that I don’t agree with, then I will simply ignore these works, of course.
In my experience, many linguists ignore the work of others that work in a different framework and inhabit a different community. This is often forced by practical considerations (there is limited time, and one is busy enough with reading up on all the work in one’s own community), but when it becomes a culture, it constitutes a real threat to the unity of science. Not everyone working within HPSG or in laboratory phonology can read up on everything written in other frameworks (or attend conferences organized by the other communities). But do we have a culture or conversations across frameworks? Or is it rather the opposite, a culture of segregation of communities in linguistics?
My own primary community, language typology as organized in the Association of Linguistic Typology, seems to be an example of segregation, whatever the causes for this. Greenbergian typology, which arose as a community in North America in the 1970s and became stronger in Europe since the 1990s, was rather different from Chomskyan linguistics since the beginning – Joseph Greenberg had very different way of understanding languages than Noam Chomsky, even though both were keenly interested in universals of human languages in the 1960s. One might have hoped for some convergence, and I remember a plenary talk by Bernard Comrie at the 1987 International Congress of Linguists (in Berlin’s Palace of the Republic), where he talked about development of rapprochement between the two approaches.
But it never happened. On the contrary, generative typology and Greenbergian typology is more segregated than ever, it seems. At the recent ALT conference in Canberra, there was a single generative talk (I think), and at a recent DGfS workshop on syntactic asymmetries organized by Gereon Müller and colleagues, there was a single non-generative talk (my own). But segregation threatens the unity of science.
In a recent blogpost, theoretical physicist Sabine Hossenfelder puts it bluntly:
“Science isn’t immune to group-think. On the contrary: Scientific communities are ideal breeding ground for social reinforcement. Research is currently organized in a way that amplifies, rather than alleviates, peer pressure: Measuring scientific success by the number of citations encourages scientists to work on what their colleagues approve of. Since the same colleagues are the ones who judge what is and isn’t sound science, there is safety in numbers. And everyone who does not play along risks losing funding.
As a result, scientific communities have become echo-chambers of likeminded people who, maybe not deliberately but effectively, punish dissidents. And scientists don’t feel responsible for the evils of the system. Why would they? They just do what everyone else is also doing.”
So in order not to fall victim to the bad effects of group-think, at least some members of each community need to talk to some members of other, potentially relevant communities.
Some such attempts at dialogue are being made: Stefan Müller has been making efforts to find common ground between HPSG and other approaches to syntax (cf. the 2013 workshop on progress in linguistics; and his paper “Unifying everything” has a beautifully optimistic title. Roberta D’Alessandro and Marc van Oostendorp organized a nice workshop in Italy in 2017 where they invited both syntacticians and phonologists from various communities (including me), and we seriously tried to move a bit closer to understanding each other. These are just two anecdotal examples, and no doubt others could be given.
But when it comes to generative typology and Greenbergian typology, segregation is dominant, and as a result, mutual incomprehension. I once had a very nice personal interaction with a nanosyntactician who was genuinely interested in my semantic-map approach, and I made some attempts to understand nanosyntax, but we failed to understand each other. I keep talking to generative linguists on Facebook (and some of these conversations make it to blog, such as one with Richard Larson last year), and I keep encountering a deep kind of mutual incomprehension.
I have long wondered why “we cannot talk to each other” (the title of my 2000 review of Newmeyer’s book), and I find it frustrating that the divisions in the field of linguistics are still so poorly understood. So I keep thinking about this, and writing negative reviews is one way in which I am trying to make sense out of the differences.
(My current thinking is that the difference between the Greenbergian and the generative approach is that generative linguists think of grammatical categories and architectures as innate natural kinds and therefore carry over concepts from one language to another without much hesitation. Any sort of apparently confirming evidence can be taken as sufficient, which means that it is very easy to claim that language B has the same category as language A. This makes the approach very hard to test, whereas if one drops the natural-kind presumption and uses different concepts for comparison and description, these problems do not arise. Eventually we want to know whether these two approaches are somehow compatible after all, and if not, which one is the right one.)