Welcome. On this page, in the next section, you'll find a brief description of some current research projects that I'm involved in in the areas of linguistics and psychology: that section also includes links to a current cv as well as to some recent (single-authored and co-authored) papers. In the 'personal interests' section that follows, there are also some links to two blogs reflecting my main unpaid interests: my family, writing and music (of all kinds). If you find anything of interest here that you'd like to contact me about please email me or—in the case of the blogs—post a comment: I look forward to hearing from you.
Vietnamese grammar
For several years now, I have been developing a project concerned with the grammar of Vietnamese. The current project is the continuation of research originally funded in Canada by awards from FCAR (Government of Quebec) and by the Social Sciences and Humanities Research Council of Canada (SSHRC internal grant). Since 2005, it has been partially funded by a discretionary award from the University of Sheffield. This year, from February-June 2009, the work was supported in part by the Arts and Humanities Research Council (AHRC, Research Leave Award). This is the first year for a long time that I have been able to devote a significant amount of time to the project (thanks to a half-year sabbatical and the AHRC leave extension grant): it has resulted in a number of journal articles and submission, as well as significant additions and (hopefully!) improvements to the grammar website.
There are several medium-term objectives for the project: to publish a grammatical description (in English) of contemporary Vietnamese, including a description of the sound pattern (phonology) of the language, and to develop an on-line, searchable tagged lexicon of Vietnamese, with English and French translations. I have also set up an online archive to collect both theoretical and descriptive work dealing with the grammatical structure of Vietnamese, as well as with areally- and typologically-related language varieties. See RH sidebar for link. In addition to this descriptive work, I am working on a theoretical monograph on Vietnamese phrase-structure, for eventual publication in the Trends in Linguistics series (Mouton de Gruyter).
Linguistic relativity
In co-authored work with Yayoi Tajima, arising from her 2008 MA thesis, I am exploring issues in cultural and linguistic relativity. Specifically, we are revisiting some influential work on cultural relativism by Richard Nisbett and his colleagues ( Nisbett, Peng, Choi, & Norenzayan 2001), which purports to show that Asians and Westerners 'think' differently in virtue of deep-seated cultural differences. Our preliminary research, comparing Chinese, Japanese and English participants in a series of visual recall tasks, shows that Chinese and English participants' data pattern together in contrast to those of Japanese participants, a grouping that we attribute to linguistic rather than cultural factors: both Chinese and English are mainly head-initial, right-branching languages, whereas Japanese is head-final/left-branching. Our results challenge the idea of a straightforward dichotomy between Asians and Westerners, and foreground the role played by grammatical structure in "Thinking for Speaking" (Slobin 2003). We presented the main results of this work to date at the forthcoming Conference on the Mind-Context Divide at the University of Iowa in April, and have a joint paper under revision. (See RH sidebar for a early draft of this paper, and a movie version of the Iowa talk.)
Distinguishing Sounds and Faces: the developmental relationship between visual and linguistic discrimination
In conjunction with Dr. Olivier Pascalis (formerly Department of Psychology in Sheffield, from January 2009 at the University of Grenoble), I have started a project which will look at correlations between retained skill in facial and auditory discrimination among (first and second) language learners.
One of the key research questions in language acquisition is to understand of the nature of the ‘gift for language’. For children, this gift is almost universal. For adults, it seems vanishingly rare: at least in predominantly monolingual countries, only very few second language learners achieve native-like performance in their non-native language, so that they ‘pass for native-speakers.’ Of course, there are many related linguistic, cognitive and social factors underlying success or failure in this domain. What is reasonably clear from previous studies, however, is that motivation and perseverance alone do not guarantee success in SLA, nor does sufficient exposure: for some people it just comes easily, for others the situation seems pretty hopeless. We know something about how to measure aptitude—see Skehan (2002), for an excellent overview—but we don't really understand what it is. For reasons that should become clear immediately, we will approach this question by investigating in the first instance the abilities of monolingual children and adults. The starting-point for the study stems from recent work on the development of infants’ face-recognition and face-discrimination skills, and specifically, the “Own Race Bias” (ORB) investigated by Kelly et al (2005). ORB refers to the fact that people are typically better at discriminating between faces from their own ethnic group. Specifically: (i) adults are 2.23 times more likely to correctly identify own-race faces as opposed to other-race faces (Meissner & Brigham, 2001); available data suggests children also demonstrate the ORB (e.g. Pedzeket al., 2003; Sangrigoli & de Schonen, 2004a); (iii) one study has found evidence of the ORB in 3-month-old infants (Sangrigoli & de Schonen, 2004b). Kelly et al (2005) [David Kelly, Dept. of Psychology, University of Sheffield] chart the development of ORB, and preferences during the first few months of life. They conclude that sensitivity to ethnic morphological differences emerges very early in life, as a result of faces seen within the visual environment: this sensitivity is the precursor of the own-race bias, which has its onset between 6 & 9 months of age. These studies on facial preference complement other studies by Pascalis et al on face discrimination. Simplifying considerably, the conclusion is that at birth there is no ORB and no facilitation for discrimination among own-race faces, but by 12-13 months, there is a clear ORB and “non-native” faces have become that much harder to discriminate.
The significant thing about these facts is how closely they parallel the developmental path observed in child language acquisition for phonological development: at birth, children can perceive phonological contrasts in any language; by 6-9 months, most children can only reliably discriminate vowel sounds that are contrastive in their own language(s); by 12-14 months, most children can only perceive native consonantal contrasts. This decline in perceptual skills clearly has an impact on SLA production (very few L2 learners are able to lose their "non-native" accent). The conjecture underlying this research is that the parallels between facial and phonological discrimination are not coincidental. Cutting to the chase –and there’s a lot to fill in here!—what we want to determine is whether there are reliable correlations between preserved sensitivity to non-native phonemic contrasts and preserved discrimination skills with respect to “other-race” faces (for monolingual, mono-cultural individuals). Note the assumption here, that the ORB and reduced phonological discrimination is a property of groups, but not necessarily of individuals: I expect that some people will do significantly better than others in these discrimination tasks, and that these individuals form a subset of those who make good second language learners). If everyone after the age of 13 months is uniformly bad at non-native discriminations, the project won’t get off the ground (though we may still have learned something). Currently, I am working to create a battery of discrimination tasks pairing face-recognition and auditory discrimination tasks, comparing reactions to native vs. non-native faces and sounds. This is a cross-linguistic study comparing English and Japanese children and adults (hence the international collaboration). If we are lucky, we'll have some partial answers to fairly large questions about how 'special' language is, cognitively speaking, see Fodor (1983), as well as understanding more about the cognitive prerequisites of the gift for language.
The Acquisition and Processing of VP-Ellipsis and VP-Anaphora
For some years now, I have worked with Ayumi Matsuo, and latterly with Leah Roberts (MPI, Nijmegen), on a set of experiments investigating the so-called Parallelism Effect (or Matching Effect) that applies in English VP-Ellipsis contexts. The phenomenon is illustrated in the following paradigm:
(a) Someone had to put out the garbage. So John did.
(b) The garbage had to be put out. So John did.
(c) John wanted someone to kiss him. So Mary did.
(d) John wanted a kiss. So Mary did.
It is generally agreed that English VPE-ellipsis following non-parallel antecedents, in the passive (b) and nominal (d) examples, is less acceptable than in the syntactically parallel (a/c) cases. What is more controversial is whether this contrast is due to grammatical or processing reasons, or possibly some principled combination of the two. Further questions include: whether immediately comparable cases involving VP-Anaphora (replacing did with did it) are immune to parallelism effects, as originally claimed in the theoretical papers that underpin this work (Hankamer & Sag 1976, Sag & Hankamer 1984); whether young pre-schhol-age children are sensitive to these contrasts; whether adult native-speakers and second language learners show similar patterns of response to this phenomenon. We have worked over the years with young children and L2 learners, using a variety of experimental paradigms: (adapted) grammaticality judgment tasks (Matsuo & Duffield 2001), Sentence Completion Judgment task (Duffield & Matsuo 2001, 2009; Duffield, Matsuo & Roberts 2003, in press) and most recently eye-tracking (Duffield, Matsuo & Roberts 2007, submitted). A summary of the L2 work was presented at the Workshop on Formal vs. Processing Explanations for Syntactic Phenomena, held at the University of York (April 2009). The presentation can be viewed here (click here for the accompanying notes, see also RH sidebar).
Language acquisition, convergence, and the Competence-Performance distinction
Language acquisition research, whether in the area of first or second language acquisition, is concerned with understanding the relationship between the kinds of input that the language learner is exposed to, and the kind of linguistic knowledge they end up with, and which they can use in language processing (comprehension and production of spoken language, reading and writing), and also, perhaps, in complex thought. Almost everyone involved in linguistic research accepts that there is a difference between knowledge of language (which generativists call "linguistic competence") and language use ("performance"): the ideological divisions and—often vitriolic—debates surround the questions of how to define and how best to investigate competence, and whether one or other—competence or performance—should be considered to be of any deep scientific interest. Many generativists take the view that performance/language use is of secondary interest at best, and that the every-day notion of a language (e.g., 'The English Language') is a poorly defined and theoretically uninteresting one. By contrast, most functionalists, cognitive linguists, psychologists and anthropologists consider that language acquisition and use can only be properly understood within a theory (or set of theories) that take into account the social, biological, and cognitive structures within which language is embedded.
As someone who has worked for/with language researchers on both sides of the formal/functional divide, I have a good appreciation of the ways our ideologies frame and constrain the questions we ask about competence and the ways that we advance our various discourses. In a recent overview article on psycholinguistics, I attempt to guide myself and the reader through what I term 'the two souls of psycholinguistic theory' in order to try to understand some of the main empirical results in a more ecumenical context. I also try to sketch a particular view of competence in two recent publications (Duffield 2003, 2004: also RH sidebar)
Having thought it over, it seems to me that the crux of the matter in the unfortunately-named 'nature-nuture' debate (at least as it applies to linguistic knowledge) is grammatical convergence: whether adult native-speakers of a particular language variety, from different socio-economic groups, varying in general measures of intelligence, working memory and educational attainment, with differences in infant and childhood development, nevertheless end up with essentially identical competence grammars. (This is not the question of whether everyone uses language equally well (by whoever's standard): no experiments are needed to determine that we're all quite different in this regard!). If we can show that native-speakers really do converge in this way, it reinforces the generativist position against the emergentist one. If however, the behavioral evidence for non-convergence is stronger, this would seem to undermine the nativist "Poverty-of-Stimulus" arguments (cf. Pullum & Scholz 2002, Reali and Christiansen 2004).
What is most noteable about this is that we really don't have any clear answers, because the work really hasn't been done: generativists simply assume convergence (without strong empirical evidence), while emergentists up to now have not been interested in investigating the sorts of phenomena that tell us about implicit competence, preferring to concentrate on production data. (Whatever it might say about differences in language performance, no amount of spontaneous production data will tell us much about convergence/non-convergence in grammatical competence). I address this a little more fully in a recent short commentary (The kids are alright...aren't they?: see also RH panel).
Aside from academic work, my main interest is my family: my wife, Ayumi Matsuo (who is also a linguist and acquisitionist), and our boys, Seán (8) and Julian (2). (Before you wonder, this was not a John Lennon-Yoko Ono tribute or deliberate allusion—I wasn't even aware of it until our neigbour Sue, who is a Beatles fan, asked us! There is a linguistic connection, though: Ian Roberts, who is like my neighbour a Lennon/MacCartney fan, named his son Julian, and I liked the name, while Seán is called Seán partly because it's one of the few Irish names that also works in Japanese. Which may well also have influenced John and Yoko...).
When we're not spending time together, I enjoy listening to and playing music (I'm a very poor re-learner of the piano, which pain tells me a lot about second language learning: for the past year, I've been working on only one piece—John Field's First Nocturne. If I'd become more fluent or competent (!) in that time, this single-mindedness might have some more positive effects: as it is, it's mostly only an annoyance to my family. I've also tried a bit of writing, some examples of which are posted on the Inishmacsaint blog. I love eating, but hate cooking, and—though it is unreasonable and deeply unfair—am firmly determined never to become a competent cook...