Linguistic Typology as a Field of Research and as an Academic Course
Sofia University St. Kliment Ohridski
Настоящият текст разглежда езиковата типология като научноизследователска област и като учебен предмет. Установяват се основни проблеми на областта. Отчита се, че те пречат на съвременния преподавател да разгледа преподаването на типология като съществен проблем. Надгражда се върху материал, изведен от конспекти за курсове по типология, за да се обсъдят основните промени в методически план и в самата дисциплина, които водят до промяна в целите и подходите на преподаване. Нарастващата употреба на големи набори от данни и нуждата от генерирането и обработката им водят до промяна във фокуса на курса по типология, който се преориентира от изчерпателност към достатъчност на познанията и уменията. На база приведената информация се извежда профилът на бъдещия типолог.
Ключови думи: езикова типология, преподаване, дисциплина, предмет, конспект.
The present text dwells on linguistic typology as a field of research but also as an under- or post-graduate course. Major issues within the field are pinpointed: I propose that these trivialise the teaching of typology to an extent. The text builds on data obtained from typology syllabi to discuss the major methodological and disciplinary changes that lead to a subsequent change in long-term goals and approaches to teaching typology. The increasing use of big data sets and the need to generate and process these result in a shift in emphasis within a typology course, the course moving from an emphasis on comprehensiveness to an emphasis on sufficiency. The profile of the typologist of the future is outlined.
Keywords: linguistic typology, teaching, field of research, course, syllabus.
This paper has been prompted by the scarcity of resources discussing the teaching of linguistic typology, which is interpreted as reflecting certain insecurities as regards the field of study, and what it should encompass. The first section discusses possible setbacks the typology lecturer/textbook writer may encounter. These reflect problems relating to the discipline itself. The second section builds on data coming from nine syllabi to discuss some changes to typology as a field of research and as a course. The third section rounds off the discussion, and suggests upcoming changes to the profile of the typologist.
Teaching Linguistic Typology and the Three Problems when Doing So
Scholars are quick to acknowledge the fluidity of the term linguistic typology (see, for instance, Croft (2003: 1-3)), noting that this leads to confusion as to the purpose of typological inquiry. Disclaimers of this sort appear in the first pages of typology textbooks, but are often abandoned at a later stage as the need to provide streamlined, student-friendly definitions takes the upper hand. Nevertheless, insecurity remains evident in the sections where researchers discuss the motivation behind their choice of topic and/or framework. Thus, Comrie (1989: xii) justifies his take on the field by explicitly asserting his view that formal phenomena are amenable to a functional explanation, leaving little to no room for misinterpretation on the reader’s part as to whose views those are. In line with the above, most textbooks, including the recently published Moravcsik (2013), begin by informing the reader that the text they are reading is merely a part of a more complex picture.
This positive tendency of informing the reader about framework-related considerations, however, does not have that pronounced a parallel when it comes to justifying one’s methodological choices. Comrie (1989: xii), for instance, notes that comparing his approach with other typological frameworks is undesirable, but that he considers his ‘no worse.’ This wording steps around an important problem in linguistic typology: that of the comparability of concepts used in different typological traditions. In this sense, the reader is only half-introduced to the complexity of the field: they have been made aware of the existence of different paradigms, but the choice of paradigm has been made for them.
One might well argue that it is unfeasible, and indeed undesirable, for a textbook to offer different, often conflicting perspectives to the novice learner. Such an argument has both merits and demerits. A reader with sustained intrinsic motivation should be driven to look for sources to accompany their primary readings. Yet it is entirely conceivable that they may gloss over the opening remarks of the author. This, in turn, may foster misconceptions based on the seeming homogeneity of the field presented in the rest of a textbook. Either way, most introductory textbooks offer only a piece of the otherwise convenient metaphor of the typological puzzle. So how does one add to, and assemble, that puzzle in the classroom?
This question brings us back to the reason for writing this paper: the scarcity of material dealing with teaching linguistic typology. There appear to be three general explanations for why there should be so few sources dealing with the matter, which also present the typology lecturer with problems. In the first place, teaching typology is largely a non-question in academia. Judging by my discussion with colleagues, it appears that there is – at least on the surface – a certain lack of concern with why, if at all, the subject is taught as well as with how this is to be done, and what skills aspiring typologists are to develop. This reflects a fallacy of identifying the field of research with the content taught in a given subject. Neither lecturers, nor students are immune to this fallacy.
One might imagine that the field being so challenging to define should result in a consequent ambiguity when it comes to the course. However, this need not be so as a certain degree of reification (as defined by Wenger (1998: 58)) is necessarily a part of adapting any field of studies for classroom purposes. This leads us to a second problem as regards teaching typology: the lack of consistency in goal setting. This will become more evident in the next section, but for now it is important to note that there does not seem to be much agreement as to the long-term targets of typology classes (which have been variously defined in terms of acquired content, subject-specific skills, and awareness of the field among others), and hence short-term course goals are also ambiguously set.
The above, in turn, introduces us to a third obstacle the typology lecturer could face: the lack of systematicity in teaching practices. This lack of systematicity can in part be attributed to a difference in the availability of resources across departments. It should be noted, nevertheless, that with the emergence of large-scale open-access projects, data are becoming more accessible to less well-funded institutions. Admittedly, other reasons may be institution-specific. Yet, more often than not, it is the background of researchers and their teaching philosophy that inform the practices in a certain course (for this, refer to the next section).
In light of the discussion above, I propose that a self-aware, methodology-conscious stance is necessary when teaching typology. That entails (among others) an increased awareness of the changes of data sets, their processing, and – in general – of the field’s need for legitimisation at present. Formulated in this way, teaching typology is likely to lay greater emphasis on skills, thus lifting it (to some extent) from the breadth of factual knowledge. These ruminations mean that looking for, or trying to establish, a uniform standard for teaching typology would be unproductive if done in terms of content. This is why, instead of doing that, I will focus on extracting possible generalisations on viable practices instead. These will be concerned with the purposes of a course on linguistic typology and its design.
2. Examining Typology as an Academic Course
This section comprises a discussion loosely based on nine syllabi in linguistic typology for courses conducted between 1997 and 2015, and dwells on some changes to typology as a field of research and as an academic course. All syllabi are publicly available, and were compiled by researchers working in the field. The common denominators have been the programme within whose framework they were set and their duration: all were one-semester courses conducted at master level. Below, I spell out some trends.
In the first place, the majority of the syllabi examined here describe courses embedded in theoretical linguistics curricula. There are some exceptions. Maya Pencheva (p.c.) has shared that her course was targeted at students in a master’s programme dealing primarily with linguistic issues. There were no prerequisites set for students entering the course, so no prior knowledge of the internal branches was assumed. Childs’ (2009) situation is similar in that the course formed part of an applied linguistics programme.
A further trend is for formative assessment to be downplayed in earlier syllabi, and for it to assume greater importance in later ones. For example, in our earliest syllabus – Comrie (2009), which describes a course conducted in 1997 – formative assessment contributes 30% to the overall grade. In contrast, the latest syllabus (Toosarvandani 2015) allots 60% to formative assessment. This reflects a general move towards skill-oriented syllabi, and a consequent shift of focus away from conceptualising a typology course as an exploration of linguistic phenomena in different languages. The motivation behind this shift will be discussed later in this section.
It is difficult to ascertain the types of skills in question as syllabi are not explicit about this. They often formulate their targets in terms of abilities such as recognition and generalisation (Goldberg 2014: 1), or in terms of familiarity with phenomena and desired output (Childs 2009). However, there does not appear to be any major preoccupation with how the targets are to be reached. Most syllabi gloss over this aspect of work with students, and it may be argued that it need not be included in a syllabus as it is technically part of the approach to teaching, and not of the discipline proper. The flip side of the argument, nevertheless, may be that skills are not discussed because they are not considered in detail. There is, of course, no way to claim this solely on the basis of syllabi, but if the supposition holds true then skills are reduced to subject-specific techniques. The acquisition of such techniques, thus, may tacitly be considered a tool for meeting the targets set before students. This would be in line with the reasoning in the previous paragraph, where focus was observed to be laid on the exploration of linguistic phenomena in more recent syllabi. Does that mean, then, that a focus on skills is discernible in later syllabi?
Indeed, a process of rethinking the role of skills in the typology syllabus does appear to be taking place. Goldberg (2014: 1), who was cited as setting targets typical of older syllabi, also mentions that students are to ‘strengthen their ability to write clearly and precisely about linguistic data, using (…) conventions (…) in linguistic typology.’ This blend of subject-specific techniques and transferrable skills is typical of other more recent syllabi. Thus, in relation to certain conventions in language description, Pearson (2012: 2) notes that ‘learning this style is a major focus of the first part of the course.’ These data, along with those presented further above, indicate that a reversal of the roles of exploring linguistic phenomena and of skills is taking place. In other words, linguistic phenomena become a tool for teaching skills (and there need to be both subject-specific techniques and transferrable skills, or else a course risks devoting itself entirely to research methods). Hence, it is expected that the strategies to achieve the emerging targets will also undergo change.
The discussion above has its parallel in the methodology of (second) language teaching, and it is there that one may discover hints as to what may be happening to linguistics-teaching paradigms in general, and typology-teaching paradigms in particular. In an influential paper, Newmark & Reibel (1968: 145) suggest that ‘In his zeal to teach language students to produce well-formed sentences, the language teacher is in great danger of underestimating the importance of teaching students to use the language.’ They go on to argue in favour of emphasising the ends to which the language taught is to be used, and of partially lifting the emphasis from the formal aspects of said language. If we abstract their argument further, a valid claim should be that, in line with constructivist approaches, teaching ought to be about assisting someone in understanding how to do a certain activity, why they would do it, and only then about what breadth of knowledge should be sufficient for them to do so. A reasoning of this sort would admittedly be an attempt at legitimising methodology’s relevance to many fields, typology included, but it is also a more practical appeal to instructors: they are asked to be more conscious of the teleology of their work since that teleology is changing.
The emphasis on sufficiency in the paragraph above suggests that a shift away from comprehensiveness may be taking place in typology teaching. Early syllabi suggest a high degree of eclecticism (cf. Pencheva (2004), for instance), drawing on multiple fields, and discussing a plethora of phenomena in the hopes of familiarising the student with a wealth of data. This indicates a preoccupation with content which was understandable at the time when such courses ran. As Perkins (2001: 431) points out, and this is a statement largely valid for the field before the very end of the 20th century, ‘Typologists have invested considerable time and effort in developing the field but little time in thinking about issues from a statistical perspective, which would require another large investment of time and effort.’ This was not without its context: at a time when research was published mainly in hardcopy, data were difficult to verify due to insufficient resources, and doing so required a lot of networking and comparing multiple data points by hand, so it was not possible to use statistical data – among others – to their fullest potential generalisation-wise. The skills a typologist needed when older courses ran are far from being replaced (this can be inferred from the fact that newer syllabi have structural and topical parallels with older ones), but new ones have been added to them. Large-scale projects such as the CLLD and ODIN as well as the linguistic ontologies cropping up of late allow for facilitated access to, and cross-referencing of, multiple data points but their use also requires a skill set greater than that of the typologist of the late 20th century.
A new (meta)component is necessary to complement declarative knowledge and the standard evaluation procedures with which a student would be familiar from their work in the classroom: the ability to guide automated data processing (including data with predictive potential) with view to the often diverging traditions scholars work in. Hildebrandt and Bond (2009) discuss the use of WALS in two typology courses, noting some challenges and some successes. With them, WALS is used mainly for illustration, and its convenience as a resource with multiple data points all in the same place is highlighted. The typology lecturer may wish to go a step further than what Hildebrandt and Bond (2009) have achieved in their courses, and suggest that students test a new feature – one that does not exist in WALS but can be automatically cross-referenced with existing WALS features if data is correctly entered and researched. If interdependences are observed, the feature(s) with the greater predictive potential and greater potential for generalisability and cross-referencing with other features may be kept for the purposes of the class as a more convenient feature for prospective research. Work of this sort allows students to use existing data, and build on them, to potentially expand existing databases and work with material which appears homogenous but often comes from diverging research traditions.
Specifically with regard to linguistic typology, the emergence of skill-oriented syllabi involving work similar to that outlined above reflects a more self-aware teaching on the one hand (i.e. a reorientation of methodological practices towards a constructivist stance), and a change in focus within the discipline itself. Change within the field of typology is clearly observable in the emergence of multiple frameworks over the past decades, some of which are the Surrey Morphology Group’s Canonical Typology and Haspelmath’s stance on comparative concepts (which stand as distinct from features). It may well be that these developments are reflexes of deeper, further-reaching change, but that would be impossible to verify in the confines of the present text. Thus, below, I shall draw some conclusions on the basis of the observations already made.
3. Concluding Remarks
Discussing the why’s, what’s, and how’s of teaching linguistic typology should not be a trivial issue for a discipline which is still struggling for legitimisation. At the same time, however, this discipline cannot be equated with typology as a teaching subject. The two are intricately related, and approaches to each vary considerably. While total homogenisation of said approaches may be unfeasible, and indeed undesirable, it is important that there should be a degree of cohesion between what is taught and/or introduced in textbooks, and what aspiring typologists may need in their later work.
A largely constructivist methodology is beginning to take centre-stage in more recent typology courses. This suggests that the emphasis on comprehensiveness shifts towards an emphasis on sufficiency, and that on declarative knowledge – towards an emphasis on skills and procedures. Structural-ideological changes discernible in syllabi can attest to that. A discipline that is changing necessitates changes in its corresponding subject. It is expected that, as teaching typology grows more constructivist and as new frameworks and tools for big data analysis appear, more and more attention will be paid to the (meta)component of a typologist’s skill set. This development is expected to take a typologist’s work closer to that of the universologist and further away from that of the field linguist. In addition, it is reasonable to suspect that the necessity to operate with data coming from, and/or intended for, large-scale interdisciplinary projects will result in a consequent need for the prospective typologist to have a breadth of knowledge in disciplines outside of linguistics different from that of the typologist of the mid-to-late 20th century.
Childs 2009: Childs, T. Linguistics 410: Language Typology. [Accessed on 06.11.2016].
Comrie 1989: Comrie, B. Language Universals and Linguistic Typology (2nd edn.). Chicago: The University of Chicago Press, 1989.
Comrie 2009: Comrie, B. Linguistics 533, Spring 1997. [Accessed on 20.10.2016]
Corbett 2012: Corbett, G. Features. Cambridge: Cambridge University Press, 2012.
Croft 2003: Croft, W. Typology and Universals (2nd edn.). Cambridge: Cambridge University Press, 2003.
Goldberg 2014: Goldberg, L. Linguistics 125b. Linguistic Typology. [Accessed on
Haspelmath 2010: Haspelmath, Martin. Comparative Concepts and Descriptive Categories in Crosslinguistic Studies. – Language, 86 (3), 663-687.
Hildebrandt & Bond 2009: Hildebrandt, Kr. A. & Bond, O. WALS in the University Classroom: A review. – Linguistic Typology, 13 (1), 183-193.
Moravcsik 2013: Moravcsik, E. A. Introducing Language Typology. Cambridge: Cambridge University Press, 2013.
Newmark & Reibel 1968: Newmark, L. & Reibel, D. A. Necessity and Sufficiency in Language Learning. – IRAL – International Review of Applied Linguistics in Language Teaching, 6 (1-4), 145-164.
Pearson 2012: Pearson, J. M. Ling 328: Morphosyntactic Typology. [Accessed on 03.11.2016] http://www.reed.edu/linguistics/syllabus/lx328_2012_syllabus.pdf
Pencheva 2004: Pencheva, M. Linguistic Typology. List of Topics. Unpublished manuscript, 2004.
Perkins 2001: Perkins, R. D. Sampling Procedures and Statistical Methods. – In: Language Typology and Language Universals. Volume 1. Eds. M. Haspelmath., E. König, W. Oesterreicher, & W. Raible. Berlin/New York: Walter de Gruyter, 2001.
Richardson 1997: Richardson, V. Constructivist Teaching and Teacher Education: Theory and practice. – In: Constructivist Teacher Education: Building a World of New Understandings. Ed. V. Richardson. London/Washington: Falmer Press, 1997.
Ryan & Deci 2000: Ryan, R. M. & Deci, E. L. Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being. – In: American Psychologist, 55 (1), 68-78.
Toosarvandani 2015: Toosarvandani, M. Language Typology (Linguistics 124). [Accessed on 01.11.2016] <https://people.ucsc.edu/~mtoosarv/syllabi/ucsc/ling-124-syllabus.pdf>
Wenger 1998: Wenger, E. Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press, 1998.
 This text owes much to the meticulous and diligent work of Bozhil Hristov, Dima Serbezova, Katerina Stankovska, and Velina Dimitrova, who read drafts and commented on them. I am thankful to my family for all their help.
 In the sense of Ryan and Deci (2000: 70).
 When I first presented a version of this paper, an experienced linguistics lecturer asked me how one could teach typology if a typologist needed to know ten languages to do their job adequately. Disregarding some intriguing misconceptions related to this question, it is worth noting that it sheds light on the great extent to which a field of studies may be – and often is – identified with its corresponding course.
 Wenger (1998: 61-62) warns that ‘reification can give differences and similarities a concreteness they do not actually possess.’ In this sense, a typology lecturer should – among other things – take care not to leave students with a sense of homogeneity that does not pertain to the field.
 Here, skills is used as an umbrella term covering both subject-specific techniques and transferrable skills.
 Constructivism comes in many guises, and here it is used as a general term standing on the practical-philosophical tenets expounded on in some detail in Richardson (1997: 3-4).
 The CLLD project (Cross-linguistic Linked Data, website at http://clld.org/) publishes freely available cross-linguistic datasets, thus allowing for more uniform access to these across institutions. ODIN (Online Database of Interlinear Text, second version at http://faculty.washington.edu/fxia/odin/) makes use of interlinear glossed text found online to facilitate access to a wealth of cross-linguistic data by offering a single access point.
 Hildebrandt and Bond (2009) are aware that problems with WALS exist, and they use them to offer students an opportunity to evaluate data, which fulfils the critical thinking criterion of the (meta)component proposed here. Yet, tasks like those presented there do not provide an opportunity for informed generation of data and their subsequent cross-referencing as the data with which students work are pre-generated.
 See Corbett (2012: Chapter 6) and Haspelmath (2010) for these respectively.
 Cf. the last paragraph of section 1.
 Cf. the discussion of Childs (2009) and Goldberg (2014), and the move towards a non-comprehensive list of topics that favour the development of specific analytical skills (inferred from the assessment method in Golberg (2014)). These syllabi can be compared with Pencheva (2004), which offers a very rich set of topics but makes little mention of assessment, thus presenting an early stage in the move described above.
Georgi Georgiev is an MA student in Conference Interpreting at the University of Sofia (Bulgaria). He works as an adjunct lecturer in Introduction to General Linguistics and Syntax at the Department of English and American Studies at the same university. His interests lie mainly in the area of linguistic typology. More specifically, he deals with understudied morphosyntactic phenomena and marginal word-formation mechanisms.