Wednesday, May 13, 2015

What the invitees to the Athen conference are thinking

The organizers for the conference have set up the following web site: One of the things you can find there are short papers by the invitees answering the questions about the future of the field that the organizers asked be addressed. I posted the original announcement here, and my own answers to the posed questions. Well, those interested can now see what the invitees are thinking, and it is a very fancy group.

Moreover, it will allow everyone to "participate" in the discussion. I would recommend reading the answers and commenting about them here. I am sure that the participants in Athens will take a look at what you are saying and try to respond. I know that I will. So, consider this an opportunity to engage in this important discussion.

I would suggest that comments be pegged to specific position papers to make it easier to follow. I'm looking forward to hearing how others react.


  1. This comment has been removed by the author.

  2. This is not so much a response to a particular statement (I have read only 4-5 of them so far), but a general concern I would like to see addressed at a venue such as this:

    A real worry I have about current theoretical syntax is what I'll call representational proliferationism. Let me explain. It seems to me that part of the founding ethos of minimalism was the idea that GB had grown into an unwieldy collection of modules and primitives (the term "baroque" is often used); and that successful as this theory was, it just had to be derivable from something simpler. And so minimalism was born, at least in part, out of a reactionary stance towards the proliferation of devices and sub-theories that came to be included in the GB canon.

    But surveying the current landscape of generative syntax, it seems to me that we have simply traded one type of proliferation (say, of sub-modules, primitives, and operations), for another type: a massive increase in the intricacy of linguistic representations. At its worst: "a projection / feature for every phenomenon!" [Before I go any further, let me make one thing very clear: my own work is by no means immune to this criticism. I have argued, for instance and among others, for the separation of person- and number-probes into separate syntactic heads, which is clearly a proliferation (of functional heads, in this case).]

    So here's my worry: Why should we believe that this is progress? Couldn't it simply be that there is an inherent tradeoff in complexity between the derivational engine and the representations it operates upon, and we have simply moved the needle all the way to one side of the scale and declared victory?

    Let me add a couple of important notes / qualifiers. First, insofar as the intricate representations in question can themselves be derived from something deeper (see, e.g., recent work by Ramchand & Svenonius), then of course this concern dissolves. I guess I'm skeptical that this can be done in general for the massive amounts of representational "heft" that has accrued since we declared war on GB. (But I'd be glad to be proven wrong.)

    Second – and this is more of a biographical note – the worry I'm expressing here is slightly different from what I've talked about when I've argued against 'uninterpretable' features. There, I argued that the move from encoding obligatoriness in the operation (read: transformation) itself, to encoding it in terms of a representational filter, is the wrong move on empirical grounds (if you buy my argumentation, that is). Here, I'm asking something else, namely: in those cases where the two options seem to be on equal empirical footing, why should we believe that massive representations with simple operations are intrinsically superior to simpler representations with a more complicated repertoire of operations?

  3. I agree, and this is one of the two main topics of my statement for Athens. I guess the issue will be discussed in the session on "Derivations vs. Representations" (although this is strictly speaking a different thing).

    1. Oh, that's great to read/hear. And I agree that it is not a perfect fit for the "derivations vs. representations" heading: while I was formulating my point in derivationalist terms, I think the concern can be restated in purely representationalist terms, as well. In those terms, it is a tradeoff between the richness of the set of well-formedness constraints on representations, vs. the intricacy/complexity of the representations which those constraints govern.

    2. I think I disagree with Omer here, or maybe just misunderstand him. The complexity of a representation is, I think, not at all at issue. Representations can be extremely complex though derived from very simple systems. It's the system that matters (and here I think derivation vs representation isn't really an issue; the system can be thought of proof theoretically or in a constraint based fashion, and I can't see much real difference). Where I think there is perhaps a point of agreement is the issue of the primitives of the representation: how many primitive features, for example, are there, how are they distinguished, what do they do at the interfaces etc. So if Omer's worry is that, I'm completely with him.

    3. @Davidadger: yes, I've been using "complexity of representations" as a (perhaps misleading) shorthand for the complexity of the set of representational possibilities. In more concrete terms: how many (possible) functional projections are there, how many (possible) syntactic features are there, etc.

    4. So I think it's worthwhile distinguishing between the content of the representations and their structure - the latter being derivative of the simplicity of the system. In terms of content, this is an issue, I agree. I'm a fan of the idea that actually there are very few features, but they are reused as functions over different categories, so we have a suite of a limited number of properties interacting to give complexity. So Daniel Harbour's approach to phi-features, for example, or the Krifka stuff that connects mass/aspect, or mine and Peter's notion of second order features, and I think there are ways of using the interfaces to prune possible organisations of features during development so UG doesn't need to specify the sequence of categories (or at least not all of it). I agree that this is probably one of the important next big challenges.

  4. [PART 1/3]

    I fully agree with Omer’s concern. Here’s another, which may — but need not — be the result of the choice of invited speakers, namely a discernible lack of young linguists who should actually be around for a solid part of the 21st century whose road ahead for generative syntax is arguably going to be discussed. (This had been noted several times before and I’m not going to launch an attack on the line-up, which is great as it is, also for the simple reason that it is quite difficult to get a sizeable group of researchers together for such an event.)

    My concern is the blatant absence of biolinguistics on the research agenda for generative syntax covering the rest of the century — both within (central research area, theory and description, internal interfaces) and outside (external interfaces and other areas beyond theoretical linguistics).

    I would like to remark on this as editor of the free, self-funded, fully open-access journal Biolinguistics ( Note that two invited speakers are members of the Biolinguistics advisory board and four are members of the Biolinguistics editorial board (five even, if one considers the original line-up from which one more speaker seems to have now been replaced), and one of the organizers was involved in the Biolinguistics task team from its inception until the beginning of this year. That’s pretty much 30% of the providers of this gathering.

    For starters, I went through the 20 posted statements and searched for ‘bio’. (You may skip the following copy-pastes and go straight to the punchline at the bottom.)

    The most neutral — perhaps because it only mentions bio- but not -linguistic in the same word — is this statement:

    “As we all know at this workshop, the most famous of these questions concerns the interaction of innate knowledge with linguistic experience (Plato’s Problem, Chomsky 1986), with its twin pillars in acquisition research and in the cross-linguistic investigation of formal universals — to which one might add the more abstract and controversial Strong Minimalist Thesis (Chomsky 2001) and compelling questions concerning language use and biological mechanism.” (Pesetsky)

    This is nice and non-committal as well as pretty much standard, and if it had been part of a larger theme addressed by more invited speakers, I’d be happy. But it wasn’t (I only count at most six). Instead, there are three statements that I find much more worrying in the context of thinking about “The Road Ahead” for “Generative Syntax in the Twenty-First Century”.

  5. [PART 2/3]

    For starters:

    “[C]urrent discussions about biological foundations of linguistics and issues of evolution of language strike me more as part wishful thinking, part promissory notes. It would be good to make some progress on that front, but in order to do that we need a more serious conversation with evolutionary biologists and geneticists. We have to make up our minds as to what the features are that we think of as crucial in evolution and bring those to the table in our discussion with people from outer fields.” (Polinsky)

    My understanding is that this gathering aims to sketch the road AHEAD for generative syntax in the 21st century. So, if “we need a more serious conversation with evolutionary biologists and geneticists”, then 85 years would not be enough to do so? Beyond the fact that such “serious conversation[s]” have already started? Even if we take the organizers’ suggestion to look into the next 10 or 20 years — surely, this should be possible! Just think back 20 years, and what wasn’t then.

    The sheer absence of any serious engagement with biolinguistics comes out more clearly here, where the term is put in quotes, just to make sure we understand:

    “Research (sometimes falling under the term “biolinguistics”) that takes the evolutionary adequacy goal as primary, and tries to create the simplest possible theory. The weakness of this type of work is its often tenuous connection to actual complex language data. (It’s remarkable how quickly a descriptively adequate theory needs more than merge.)” (Legate)

    But perhaps I’m doing a dis-service to Legate. So when she says that “[r]esearch (sometimes falling under the term “biolinguistics”) that takes the evolutionary adequacy goal as primary”, she may very well be saying that there are people who “tak[e] the evolutionary adequacy goal as primary” and a sub-set of these call their research “biolinguistics”, and that they form yet another sub-set of all other researchers in biolinguistics. Thus she might actually mean that there is (or could be) strong biolinguistics, but the evolutionary adequacy-seeking self-labeled biolinguists are not part of that. Perhaps, but she still doesn’t say anything about the other biolinguists and their role for generative syntax in the 21st century.

    But one can do better:

    “I am not sympathetic to recent trends in Biolinguistics, which to my mind is guilty of extreme Overreach in attempting to connect linguistics to Biology. I think it gives the whole field a bad name. The granularity gap and the terminology gap (to put it in Poeppel’s terms) are still too great to sustain the specific kinds of proposals that are being taken seriously in this sub-group.” (Ramchand)

    The ‘granularity mismatch problem’ (which is the actual term Poeppel used) Ramchand alludes to is well known and well discussed. But sitting back and letting neuroscientists sort it out won’t help linguist(ic)s. Is this the beginning of an apologia that ‘generative syntax’ should stay the close circle it is? It certainly would fit at least one avid reader and frequent commentator on the FoL blog very nicely. We’re talking about the 21st century, which has another 85 years to come!


    So, despite 30% of the providers being involved one way or another in the journal Biolinguistics, the research area of biolinguistics does not feature prominently — in fact not at all, and to the extent it does, rather negatively — in the posted statements. (OK, some of the speakers clearly envision a biolinguistic agenda even if they don’t put a name to it — though how “generative syntax” alone is going to accomplish this, is beyond me; at the same time, I wonder how many of the invited speakers would actually answer to the question what they’re ultimately interested in if they only had up to two words: “generative syntax”.)

    In other words:

    Biolinguistics is not considered a worthy research aspect, certainly not a corner stone or even just a possible/plausible/desired direction, of the road ahead for generative syntax over the next, say, 85 years.

    To some extent, this attitude is mirrored by the declining interest in Biolinguistics, the journal — despite having download/viewing numbers regularly exceeding 2,000 and getting way over 10,000 for some contributions (as of 31 December 2014, the five top-viewed articles had been viewed between 6,315 and 17,842 times). And as a fully open-access journal, one doesn’t even have to be part of the over 3,000 registered users, anyone can access anything! But that is an issue that needs to be addressed properly another time.

    Don’t get me wrong, I’m still excited about this event — if only because Athens is always well worth a visit. So, in closing, I hope to see many of you there!

    1. @Kleanthes: While I cannot speak for Ramchand, Legate, or Polinsky, I think I share some of their feelings on the issue. Let me make a couple of points (which they, of course, should not be held responsible for).

      I think the discussion is clouded by an unfortunate confusion between biolinguistics as a research interest or research agenda or set of foundational questions on the one hand, and a relevant portion of the work being done right now under the umbrella of biolinguistics on the other. I don't think anyone of the gatherers in Athens has any qualms about the former; one can practice linguistics without practicing biolinguistics in the narrow sense, but that doesn't mean the questions aren't interesting and valid questions to ask.

      But I, at least, think that a lot of those (again, very valid) questions are simply not ripe to be answered yet. From my vantage point, a lot of current work under the biolinguistic umbrella asks "why does natural language behave in manner X" about Xs that are far, far from certain. (And maybe even some Xs that have already proven to be false.) There is a pervading narrative – maybe confined to the subtext, maybe occasionally making its way to the text itself – that we have pretty much finished figuring out how language is, and it is time to wonder why it is like this. On a charitable reading, this is wishful thinking.

      This is not to denigrate the massive amount we have learned about what language is like; Norbert often points this out when arguing with the flat-earthists, and rightfully so. But it appears that many of these "why" questions will prove extremely sensitive to minor perturbations in the "what." And those perturbations are still going strong. (I am no psychoanalyst, but it seems to me that people didn't start working on the "why" questions because they felt we were done answering the "what" questions, but because they simply became bored(?) with those and wanted to ask a different kind of question.)

      And hey, maybe I'm the fool screaming at the city gates, "You'll never figure out a systematic account of the planetary paths! We don't know nearly enough to do that, yet!" right before someone much smarter than me does just that.

    2. @Omer, maybe I'm reading you wrong here, but it seems that your use of the term biolinguistics here is pretty indistinguishable from minimalism (in the sense of asking 'why' questions and not just general occam's razor stuff). Chomsky of course always admits that minimalism might be premature, but it seems like you are outright asserting that. Is this a fair reading?

      I, for one, find myself pre-occupied with bioling/minimalist ideas, but maybe there is some sort of counter-reformation afoot?

    3. In our editorial for the inaugural issue (, Cedric and I wrote:

      There is both a weak and a strong sense to the term ‘biolinguistics’. The weak sense of the term refers to “business as usual” for linguists, so to speak, to the extent they are seriously engaged in discovering the properties of grammar, in effect carrying out the research program Chomsky initiated in Syntactic Structures.
      The strong sense of the term ‘biolinguistics’ refers to attempts to provide explicit answers to questions that necessarily require the combination of linguistic insights and insights from related disciplines (evolutionary biology, genetics, neurology, psychology, etc.). We regard Eric Lenneberg’s book, Biological Foundations of Language, published exactly forty years ago (Lenneberg 1967), as the best example of research in biolinguistics in this strong sense.
      We would like our journal to provide a forum for work in biolinguistics in both the weak and the strong sense. We would like to stress that the term ‘weak sense’ is not meant to indicate that we regard work focusing narrowly on properties of the grammar as inferior to interdisciplinary work. Indeed we think that such work is not only necessary, but has very often proven to be the basis for more interdisciplinary studies.
      (Boeckx & Grohmann 2007: 2)

      I believe the many contributions over the years, in the journal but also elsewhere, showed an interest in both senses, and even concerted efforts of linguists engaging with colleagues from “related disciplines” to address the “strong sense” of biolinguistics. So people are starting.

      My point about the absence of biolinguistics on the agenda for the gathering was first and foremost that we need to look beyond feature structure if we want to be doing more research throughout the remainder of the century. I acknowledge that the amount of these 85 years may be a bit too much to handle for most of us, physically, but the preview frame of the organizers’ suggested 10 to 20 years surely should involve such developments.

      But maybe I’m just bonkers, and “generative syntax” in the narrow sense it is — in that case, however, I’m curious as to why exactly the speakers were invited, of which at least a quarter most certainly not classify themselves first and foremost as syntacticians.

    4. @Audrey: I agree that the philosophical underpinnings of minimalism, insofar as they go beyond "good science," end up including those of biolinguistics. The reason I think this is that if the object of study, the language faculty, is part of the mind, then if you keep posing "why" questions ad infinitum you will ultimately end up in biology (maybe not exclusively, but biology is certainly at least one of the places you'll end up).

      As for what's afoot, I can hardly speak to that. When I said I'm only speaking for myself, I was not offering some sort of rhetorical hedge; I honestly don't know if others feel this way. What I know is that there are self-identifying generative syntacticians working within the (broadly-construed) Principles & Parameters tradition that think some of the technical apparatus that was proposed alongside the Minimalist Program is simply wrong (say, some nonempty subset of {the Activity Condition, the Case Filter (in the GB sense), phases, uninterpretability}; not meant as an exhaustive list, of course). Does that mean the Program is wrong? No, of course not; but the Program does not operate in a vacuum, and it needs actual Xs about which to ask "why X."

      And so, for example, I was extremely disheartened to see Chomsky building the already-known-to-be-false assumption that [Spec,TP] is the locus of finite agreement into his recent theory of labeling, as one of the explananda of that theory. That is a prime example of what I'm talking about: the form of the question is solid (if [Spec,TP] was the locus of finite agreement, why would that be), but it is being asked about something that's wrong in the first place.

      I don't know if that would qualify as a counter-reformation and, in any event, I'm speaking strictly for myself.

      @Kleanthes: I don't disagree with anything you wrote in your latest comment; I was just giving my personal take on the unease that emerges from the quotes you provided.

    5. Klea notes a reasonable distinction between what we might call Biolinguistics in the broad and narrow sense. In the narrow sense we are talking neuro and genetic bases of language. In the wide sense we are a talking about language as a cognitive organ. Now, IMO, we already have a lot to say about the latter and relatively little to say about the former. What I don't see is why being able to say something about the latter is not doing biolinguistics. In the local world in which I live, there is a smooth transition form questions that look like they belong firmly in the world of formal grammar to this that look like paradigm examples of work in psychology. There are even hints of some work relevant to neuroscience. So, Lidz's and Phillip's and Idsardi's work are good examples of how linguistically informed work can combine with techniques from other domains (psych) to fatten cognitive conclusions arrived at on linguistic grounds. It also open new questions the answers to which will heavily rely on what we know about linguistic structure investigated using our standard techniques. In this world, we are all studying the same thing (FoL) using different techniques. As FoL is a biological entity, we are de fact doing biolinguistics even when we don't know a damn thing about genes or even much about brains.

      Imagine a birdsong person being told that when they work on the structure of these songs (as many do) they are not doing biology? Or tell someone working on animal behavior that they are not studying the biological properties of animals. THey would laugh. So unless one believes that humans are not animals then we are all doing biolinguistics, at least in the wide sense. When will we do narrow boiling? Well, ask that question about other domains (vision, audition, face perception) and you will find, I believe, that they are also very far from knowing anything biological in this sense. Why? Because we don't know much about brains and how they link to genes. We can't even do c-elegans. So by the stringent criteria often adverted to, nobody is doing biology, i.e. linguistics is, once again, no worse than everything else in the cog-neuro sciences.

    6. @Klea – Thanks for this perspective on things, and I'm glad to know you'll be at the event. Just two quick, somewhat interrelated points. One, I know you hedged it, but I think your repeated focus on 85 years is telling. For me, “the 21st century” in this context is meant as a general evocation of the future; when I think about research strategies and how to communicate with other disciplines, I am not thinking in anything like that time-frame. Herein might lie some of the tension between the long-term way you are thinking about the nature of the science and the shorter-term way some of us organizers were thinking about the state of the field. And this bears on the second point: To some extent the statements that the invitees made have surely been influenced by the set of "SWOT"-type questions that we presented them with (Strengths, Weaknesses, Opportunities, Threats); a different set of questions might have brought out biolinguistic issues in a different light.

    7. Sure thing, Peter, that's why, as you pointed out, I also mentioned the 10- to 20-year timeframe I had read about. (Incidentally, in 20 years all invitees will be near/post current European retirement age, and almost half past 80 years of age.) If this is the timeframe you're thinking about, "a more serious conversation with evolutionary biologists and geneticists" should definitely be on the agenda.

      This said, though, (and as a general follow-up comment on my post) it's clear even to me that not everything that should be on some agenda needs to be resented as such in the statements. Sure thing. I also agree with Norbert's assessment above (which I hope to reflect the other 20+ providers' as well).

      In any case, I really am looking forward to this -- for many reasons. I mentioned one at the end of PART 3/3 but the sheer intellectual stimulation comes, of course, first. :-)

    8. We all place our bets on where the gold will be found (... and on whether there's more money to be made in digging for coal), but my own hunch is that we stand to gain more from bridges to psychology, cognitive neuroscience, and computer science than from direct bridges to biology and genetics. It has puzzled me why "biolinguistics" has tended to show a distaste for psychological, neural, and computational analyses. I have sometimes wondered if that is because there are already well-established traditions at those levels that have generally shown little interested in the phenomena that grammarians are fascinated by.

    9. The idea that we should all be doing more "strong biolinguistics" seems to clash with the point (made for example by Gallistel and often discussed here) that when reduction happens it often turns out to be the lower-level science that needs to be adjusted to fit with the higher-level one.

      Of course this doesn't mean that all reductions are going to go that way, but given that many people have expressed the concern that strong biolinguistics at the moment seems to be a bit of a stretch, the Gallistel idea should, I think, make us more comfortable with leaving the gene stuff until later and going about business as usual. (i.e. "we are de fact doing biolinguistics even when we don't know a damn thing about genes or even much about brains", as Norbert says.)

      What's more, even if we were going to try to push linguistics in a direction guided by "lower-level sciences" (this term is perhaps not quite right here but I think the point is clear enough), I agree with Colin's point that things like psychology and cognitive neuroscience are probably a better place to look. To go straight to biology and genetics seems like it's skipping a couple of levels. Certain things about the way linguistics interacts with these intermediate-level disciplines seem to be somewhat dysfunctional, for a whole variety of complex reasons probably, and perhaps that played some role in leading linguists to look elsewhere for connections. But I think a better bet would be to try to improve the interactions with these more proximate disciplines.

  7. Many of the statements give the impression that doing experimental syntax is by and large time better spent doing theoretical work. I think that this downplays the role that experimental syntax *could* play in advancing theory were more theoretically minded people involved in doing experimental research. Introspection is indeed a powerful tool for theoretical work, but our intuitions are notoriously poor at detecting interactions. In this sense, experimental syntax can help to identify the space of a problem.

    To highlight one example: Grand Goodall's work on d-linking and island violations. He observes that d-linking increases acceptability even outside of island contexts. The effect that we observe with d-linking and islandhood (where the violations do not sound as bad when the offending element is d-linked) is a result of d-linking's independent effect of increasing acceptability. The effect is simply easier to detect when it interacts with island violations. Hence, we may not need a theory of islands that affords d-linked elements a special status.

    I think the root of the problem is that most of the people actively engaged in experimental work are not asking the questions that theoreticians are interested in. They are not trained as syntacticians. The "straightforward" solution is to train more theoretical syntacticians in experimental methodology. However, as Anagnostopoulou points out, this is a considerable investment often undertaken by younger academics who ought to be engaged in more theoretically oriented research in the early state of their careers. In short, I agree that there is a disconnect between theoretical and experimental syntax, but I wanted to highlight that connecting the two is not necessarily fruitless.

    Another point of interest in this topic is Tim Hunter's work where he shows how we can use complexity metrics to probe theoretical issues (in particular a decomposed Merge). This is perhaps an experimental methodology more accessible to syntacticians.

    Furthermore, I would like second van Riemsdijk's comment that Williams’ Representation Theory has not received the attention that it deserves. This dovetails into a concern mentioned in a few of the statements, best summarised by Anagnostopoulou, that " interesting theoretical work is ignored if it is not fashionable or produced at the right places".

  8. I had a quick read of the full set of statements. It's a very interesting collection, by a distinguished set of authors. I think it is useful to see these gathered together in one place, making it possible to see the common threads across different commentators. Since they're mostly just 2-4pp long, this is perhaps better than the lengthier chapters that they might eventually grow into. I'd like to see similar exercises in other sub-fields.

    A couple of things strike me in reading the collection. These aren't generalizations that apply uniformly to all of them, but there are common threads.

    1. Norbert's summary seems to be about right. There's a lot of interest in the typological-comparative strategy, little concern for Plato's Problem or Darwin's Problem. The authors in general seem content with this.

    2. A frequent theme is that syntacticians are poorly understood, and that it's important for others to understand the work better. Less is offered about how to address this, or what the sales pitch would be.

    3. Contrasting with the sense that the world needs to know more about the results of syntax, it's less clear whether syntax needs to draw on insights from the world outside. (Well, more diverse languages, and some semantics and phonology would be useful. But not much more than that.)

    4. There seems to be tacit agreement with another of Norbert's common refrains, which is that the basic outlines of the phenomena under investigation are established. The kinds of discoveries that we should expect in the coming decades will be of a different nature than in the early years. Relatively familiar phenomena are likely to be found in novel combinations in new languages, but nobody seems to be predicting a world of new phenomena.

    5. There seems to be general agreement that the action is all to be found at the same Marr(-esque) level that the field has occupied for the past 50 years. (I wouldn’t agree with Thomas’s description of this as being algorithmic, but that’s for another day.)

    6. The general impression of the future is that it looks fairly similar to the past and present. Perhaps with increasing anxiety over the dangers of being a bit of an isolated field.

    To be clear, these remarks are not a veiled pitch for doing more of what I do (please no!). And we don’t want folks running around doing things differently just for the sake of doing something new, or for the sake of collecting so much data that there’s no time to build theories (please no!). But I am struck by the degree of consensus and the relative lack of worry (e.g., “we’ve learned a great deal about X, but Y is a key problem that we’re just getting nowhere on”). Even within the broad typological-comparative approach — which I am a big fan of — I’m surprised not to find more diverse views over how one should pursue research at that level.

    1. This comment has been removed by the author.

    2. This comment has been removed by the author.

  9. Colin,

    Forgive me, but I think these remarks are a bit unfair. As Peter Svenonius remarked earlier, we had a particular assignment to fulfill for these statements, and at least the way I read the assignment, it wasn't to imagine the future, despite the "road ahead" subtitle.

    The way I saw our assignment, it was to react and comment on the key presupposition of this conference, namely that (1) a few decades ago we lived in halcyon days of common purpose and shared assumptions, but (2) today we have lost that paradise, so (3) we should figure out how to regain the golden days that we have squandered. To this presupposition, I think it is appropriate to reply: wait a minute, this whole narrative is off.

    Speaking for myself, I would have rejected a request to engage in futurology, precisely for the kinds of reasons that you give. It would be a ridiculous exercise, precisely because the success of science (and the fun of it) comes from the fact that you can't predict the future. There's a well-known meme, widely attributed to Isaac Asimov but apparently apocryphal, that puts it in a nutshell: "The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka' but 'That's funny...'. At the same time, if forced to talk about the flying cars and talking robots of future linguistics, rest assured I and probably many of the others would have discussed Plato's Problem, Darwin's Problem and every other kind of problem that we've been chasing after for decades. But actually the right answer would probably be XYZ's problem which we can't even imagine right now, because XYZ, whoever that may be, hasn't thought it up yet.

    Do I think or hope that "the basic outlines of the phenomena under investigation are established", "the action is all to be found at the same Marr(-esque) level that the field has occupied for the past 50 years", or "that the future is that it looks fairly similar to the past and present". No, and I can't imagine that many of the others asked to contribute have such a view. But neither do I think that we're living in a post-apocalyptic Mad Max generative grammar dystopia. I suspect that many of the contributors felt the need to focus on what's going well for us now — simply to contradict the presuppositions of the conference and fulfill the assignments that we were given.

    So please (this is a comment on the entire thread, past, present and future): a bit more understanding of the context of the texts you are reading, and a bit less snark, folks. Don't shoot, we are your colleagues.


    1. [oops: by the time my comment was posted, yours had migrated down here]

      David: sorry, no snark was intended in this one. (There may have been some snark in one of my earlier comments on this meeting, but that certainly wasn't the intent in this case.) I found it genuinely interesting to read these pieces, and appreciated the format. I'd like to see more of that kind of thing, and kudos to the organizers for requesting these and making them available.

      My goal in pointing out the apparent points of consensus was not to implicitly mock them, or to argue for something different, but to say, "Ok folks - you seem to agree on these various points; does that represent a genuine consensus?" Meetings like this tend to focus on the points of difference, but I was curious to check whether the non-disagreements are genuine. Your comment indicates that my impression of the consensus is inaccurate. If so, I'd be very interested to know why.

      Example: Norbert is sympathetic to the view that the basic outlines of the phenomena under investigation are established". That sounds like a harsh assessment, but perhaps it's not so far wrong. It wouldn't be such a bad thing to conclude that the grammatical periodic table is largely built out, and that the questions should shift accordingly. Similarly, perhaps the Marr(esque) level that syntax has focused on really is where all the action is. (There's not a whole lot of extant evidence to contradict that assumption, in fact.)

      I am sympathetic to the Asimov remark, and it certainly describes a lot of my own experiences. But it only goes so far. It describes how we come across new problems, typically within the bounds of a fairly well-described set of assumptions. But once we have well-defined problems, then we're in search of the Eureka! moments. One of the things that I've always found appealing about our field is that there are some relatively well-defined Big Problems that we're accountable to. I'd include Plato's Problem in that group. If we've been mentioning it forever, but have made little progress, then it would count as a case where some Eurekas are needed, and maybe some discussion about why we haven't had many such moments.

      (I think I'd disagree on your final remark. Slightly shifting from the gun metaphor: sparring among colleagues is exactly what we want -- It improves our game, and hopefully we understand that we're not trying to draw blood.)

    2. @ Colin: Perhaps the "snark"-comment was directed at me.

      @ David: If so, I didn't mean it as bad as it may have come across (I certainly bear no guns but also agree with Colin's metaphor-shifting; especially one of the passages I quoted is very clear irrespective of context). Your post certainly put the "21st century" bit in perspective, which Peter had also picked up on.

      @ ALL: To repeat, I really *am* looking forward to the gathering, and while I share other people's concern about certain aspects of the meeting, I fully sympathize with the organizational difficulties, among other things.

      I think I'll leave it at that for now (more in situ).

    3. Colin and Kleanthes (who I wasn't thinking of so much with my remarks, actually) - thanks for the clarifications.