1. The early years

In 1959, when the Linguistics Association of Great Britain was founded, linguistics already had a presence in several UK universities. In the School of Oriental and African Studies (part of the University of London), the Department of General Linguistics had existed since 1932, with J. R. (John Rupert) Firth as its head from 1941 and as Professor of General Linguistics since 1944 (Palmer Reference Palmer and R. E.1994) – the country's first chair in general linguistics (Marshall Reference Marshall2004, contra Halliday Reference Halliday, Brown and Law2002). Other chairs followed, including (in 1948) Angus McIntosh's Chair of English Language and General Linguistics at Edinburgh (Asher Reference Asher, Brown and Law2002), and (in 1951) Alan Strode Campbell Ross's Chair of Linguistics at Birmingham (Marshall Reference Marshall2004). Alongside these academic status symbols, there were active clusters of linguists in a handful of other universities including Durham (Collinge Reference Collinge, Brown and Law2002, Quirk Reference Quirk, Brown and Law2002) and Cambridge (Allen Reference Allen, Brown and Law2002), as well as the cluster in Hull out of which the LAGB emerged, as I explain below. These were all scholars who chose to call themselves ‘linguists’ or ‘general linguists’, thereby aligning themselves at least in part with the linguistics which was already firmly established in the USA. But at that time there were, of course, many others who did research on language outside linguistics – not least, philologists and phoneticians, including Daniel Jones's famous department at University College London (UCL).

However, although numbers were small, linguistics had a bright future. Speaking at the LAGB's 25th anniversary celebrations in 1984, R. H. (‘Bobby’) Robins gave an up-beat summary of post-war expansion:

In 1959 linguistics was taught mainly to postgraduates, with a handful of isolated modules for undergraduates, but within the next decade, it became a full undergraduate subject with single-honours and joint-honours degrees in a number of universities, starting with a joint-honours degree at the London School of Economics (Aitchison Reference Aitchison, Brown and Law2002: 7) and a single-honours degree at Reading (Palmer Reference Palmer, Brown and Law2002: 235); and the expansion continued through the 1970s. In 1959, the mood was buoyant as linguistics was clearly set to have a national presence – hence the need for a national association to represent its interests. This paper will end with the question whether this national presence has now reached its correct level.

The LAGB grew out of the Hull Linguistics Circle,Footnote ² whose character gave the association a somewhat complicated start. (Unfortunately, the available archive material contains nothing before 1963, as can be seen from appendix 2, so my account of the early years depends entirely on what the surviving participants have been able to remember.) The Circle was initiated and organised by Jeffrey Ellis, a lecturer in the German Department at the University of Hull who later became the LAGB's first President (as can be seen from the list of LAGB officers in the first appendix). The Circle's function was to ‘bring a number of linguists together for intensive weekend discussions once, or maybe twice, in a year’ (Halliday Reference Halliday, Brown and Law2002: 121). However, Ellis was also a member of the Linguistics Group of the British Communist Party, along with Michael Halliday, Dennis Berg (the LAGB's first TreasurerFootnote ³), Jean Ure, Trevor Hill, Peter Wexler and others (ibid.: 118). It is hard to unravel the relation between the Hull Linguistics Circle and this group, but it is fair to assume that since most of the Group members later joined the LAGB, they also attended the Circle meetings. On the other hand, those who attended the Circle ‘had various affiliations and various angles on language’ (ibid.: 121), which suggests a diversity of political views. Moreover, N. E. (‘Oscar’) Collinge, the LAGB's second President, recalls that ‘for four or five years the Hull conferences brought together linguists from London, Manchester, Durham, Newcastle and other centres’, but without mentioning the communist group (Collinge Reference Collinge, Brown and Law2002: 69). Indeed, according to Collinge, at least one conference was attended by Firth himself, whose political views were at the opposite end of the spectrum (Halliday Reference Halliday, Brown and Law2002: 119). It is true that Peter Matthews recalls rumours that the newly-founded LAGB ‘was at first more like a cell of the Communist Party than a normal learned society’, and that it was used to promote Halliday's controversial ideas (Matthews Reference Matthews, Brown and Law2002: 205); but these second-hand impressions do not easily mesh with the memories of those who actually attended the early meetings (Frank Palmer, p.c.). Whatever the truth may have been, it seems likely that other linguists at the time may have viewed the new association with the same suspicion as did Matthews.

Another complication at the birth of the LAGB was its relation with the already ancient Philological Society, which was founded in 1842 in the wake of earlier societies with similar names and aims founded in 1830 and 1792, and whose great achievement had been the ground-breaking Oxford English Dictionary (Marshall Reference Marshall2006). If isolated linguists needed a national association, the obvious move was to join ‘PhilSoc’ (as the Philological Society is generally called). Admittedly the name ‘philology’ suggested a much more antiquarian or literary approach to language than the new linguistics, but PhilSoc's aims included the study of ‘Structure, the Affinities, and the History of Languages’, and indeed its membership has for a long time included some of our most distinguished linguists, including Firth himself. ‘From the 1930s he himself [Firth] had played an active part in the Society and in particular had helped to make it the major British forum for the discussion and promotion of what was by then coming to be called structural linguistics, with a variety of characteristically distinct schools’ (Lyons Reference Lyons2006: 11). In short, from a purely academic point of view, PhilSoc at that time already provided a perfectly satisfactory forum for presenting work on both descriptive and theoretical linguistics, and still does.

The most important difference between the LAGB and PhilSoc has always been not their academic coverage, but rather the social, organisational and geographical structures. PhilSoc's annual programme has always consisted of seven separate meetings where a single invited paper is read and discussed, and most meetings are held in London, with annual meetings in Oxford and Cambridge (or, more recently, further north). At least in the early days of the LAGB, it was normal for papers presented at PhilSoc meetings subsequently to be published in the society's journal, so the presentation would be the reading of a publishable paper. In contrast, the Hull Linguistics Circle met twice a year for a whole weekend, giving opportunities for a series of presentations and much more social interaction; moreover, since the presentations were not intended for eventual publication, they could be less formal and (perhaps) more easily digested. In Halliday's words, ‘these meetings were for me, and I suspect for quite a few others, the first encounter with the phenomenon of the academic conference, with people presenting papers and discussion following’ (Halliday Reference Halliday, Brown and Law2002: 121). This pattern of meetings proved popular and is probably the main reason why the LAGB has survived alongside PhilSoc as a national association for linguists. The two societies now co-exist on a friendly basis, with a good deal of overlapping membership, but their early relations were somewhat tense. When the LAGB was mooted, PhilSoc was asked for its blessing (Frank Palmer, p.c.) but some of its leading members were opposed to the founding of a rival organisation (John Lyons, p.c.), and it is interesting to wonder why PhilSoc did not experiment with alternative arrangements for its own meetings (Palmer Reference Palmer, Brown and Law2002: 234).

It is unclear whose idea it was to turn the Hull Linguistics Circle into a national association. One suggestion is that Firth himself suggested it (Collinge Reference Collinge, Brown and Law2002: 69); but in view of Firth's loyalty to PhilSoc, this is unlikely. In any case, it was certainly Ellis who mentioned the idea in conversation at a Hull meeting (Frank Palmer, p.c.), so the LAGB, like the Hull Circle, may well have been Ellis's idea. What we do know is that the first weekend meeting took place at SOAS (not UCL, pace Collinge Reference Collinge, Brown and Law2002: 69) in London from Friday, 30 October, to Sunday, 1 November 1959, and that it was attended by 25 members and guests (Fiona Marshall, p.c., quoting material from the LAGB archives). The new association, which had 63 founder members, considered the name ‘British Linguistics Association’ but (in view of the unfortunate acronym BLA) adopted ‘Linguistics Association (Great Britain)’, which later changed to its present name (Collinge Reference Collinge, Brown and Law2002: 69). The natural choice for the first president (officially called ‘chairman’ at that time) was Jeff Ellis, the founder and organiser of the Hull Linguistics Circle and possibly the creator of the LAGB.

The remainder of this article takes the history forward by topic rather than chronologically. Section 2 discusses the LAGB's distinctive characteristic, its conference structure; section 3 is about the association's publication, the Journal of Linguistics; section 4 is about the internal organisation in terms of members and officers; section 5 deals with the LAGB's relations with other scholarly associations; section 6 describes the various ways in which the LAGB has represented linguistics in dealing with government and other outside bodies; and finally, section 7 raises some questions about the future.

2. Conferences

The LAGB's main activity has always been the organisation of conferences. Not surprisingly, the pattern of meetings has not stayed the same over the fifty years covered by this history, though some of the variation has to be guessed at because of gaps in the information that has survived.Footnote ⁴ The location of conferences has always varied, as have the frequency of conferences (between annual and bi-annual, i.e. twice per year) and their duration (three or four days) and date.

Since its inception, the LAGB has nearly always used cheap university accommodation for its conferences rather than the much more expensive alternative of hotels or conference centres (as in the USA's LSA conferences), and expenses have been kept down by the use of local staff and students as volunteer organisers. Costs have obviously risen with inflation, but an entire three-day conference package cost only £90 in 1995 and £176 in 2003, rising to £220 for four days in 2007. This is an important consideration because the association has always tried to encourage postgraduate students to attend meetings (as I myself did as a Ph.D. student in 1963),Footnote ⁵ and getting funding for attending conferences can be a problem even for salaried staff, let alone students. At first the meetings were tied geographically to Hull, but from 1962 one per year was held elsewhere and after 1964 the principle of choosing a different university for each meeting was established.

The number of meetings has varied according to demand. The inaugural meeting in November 1959 seems to have been followed by a single meeting in 1960, but from 1961 to 2003 meetings were bi-annual in principle, though there was no Autumn meeting in some years during the 1970s (1971, 1973, 1974 and possibly 1976), and the same may also have been true in 1964 and 1965. Conferences typically lasted three days, i.e. from midday or tea-time on the first day to midday or tea-time on the third. The demand for conferences has to be measured primarily in terms of attendance rather than in terms of papers offered, because each conference is meant to cover its own costs, which it can only do by attracting a certain minimum of participants. In the early years, there was a shortage of papers rather than of participants, so that one of the President's tasks was to persuade colleagues to offer enough papers to fill a programme without any parallel sessions – for instance, I gave my first LAGB presentation in about 1965 as a recent Ph.D. by invitation from my external examiner, Frank Palmer, who happened at that time to be the LAGB Chair; and in 1970, the programme included just 10 papers. Indeed, there were so few competing papers in those years that the second afternoon of a three-day conference tended to be left free of any presentations at all as ‘down’ time for valuable social activities. Since then, of course, the number of papers has shot up, reaching 29 in 1975, 48 in 1988, and 67 in 2008 (under the new four-day conference arrangement). Current LAGB programmes typically have three parallel sessions without any down time at all, and would-be presenters compete for time by submitting abstracts to be considered anonymously by the whole LAGB committee.

In contrast, attendances have not risen in the same way, though there are few attendance figures in the archive to document this. For some decades conferences attracted about 100 people, though this figure fluctuated between about 70 and 120 depending on the glamour of the hosting university. The large number of alternative conferences in linguistics, combined with some rather poorly attended conferences, led the LAGB committee, in 2002, to initiate a discussion about the number of conferences per year. The result of the ensuing discussion was a move to the present pattern, with a single four-day conference (at the end of the summer) in place of the two three-day conferences. The last Spring conference took place in 2003, and the first new-style annual conference lasted from 30 August to 2 September 2004. So far the results of the change have been encouraging, with attendances much higher than in the old bi-annual pattern. For example, the 2005 meeting in Cambridge had no fewer than 200 registered attenders. However, it is hard to separate the effects of the new pattern from those of the venue's prestige, and in any case, only a quarter of this number registered for the entire four-day event,Footnote ⁶ so the new pattern may hold the seeds of a different set of future problems.

The timing of meetings has always been problematic because of the difficulty of pleasing everyone. The first meetings of the LAGB continued the Hull pattern with meetings lasting over a weekend during term-time (typically in October or November), but the lack of cheap student accommodation and the conflicts with teaching commitments worked against this pattern. By 1963, one meeting was in the Easter break (i.e. late March or early April) and the other was in October or November. The last November meeting seems to have taken place in 1975. After that, a decision was taken to try to avoid university term time by scheduling meetings in September. However, an increasing number of universities have been starting their teaching year in early September, so since 2004 (and the start of the single annual meeting) conferences have generally started at the end of August.

These changes in the ‘external distribution’ of conferences have been matched by equally important changes in their internal structure. As mentioned above, the early conferences were if anything short of speakers, so all the papers were plenary. Moreover, in the 1960s some of them were longer than the present standard allocation of 25 minutes talking time and 15 minutes for discussion, so ‘one could easily speak for 40 or 50 minutes’ (Matthews Reference Matthews, Brown and Law2002: 206). This generosity was possible in part because the conference provided little but individual papers. The following paragraphs outline the process whereby this homogeneous structure evolved gradually into the present much more differentiated one, in which time is allocated to workshops, tutorials and invited speakers.

From the early days, it was official policy (as stated, for example, on the inside cover of each Journal of Linguistics volume) that ‘original papers are presented and special topics are discussed at symposia and colloquia’. However there is very little evidence in the archives of either symposia or colloquiaFootnote ⁷ and by 1977 the policy had been reworded as follows: ‘The Association holds annual Spring and Autumn meetings, at which original papers are presented in plenary sessions and in section meetings devoted to special interests’. So-called ‘section meetings’ were indeed a significant feature of meetings in the 1960s and the 1970s; for instance, all the following ‘sections’ were included in the programme for the Spring 1972 meeting:

• • Developmental and Clinical Linguistics

• • Language Teaching and Language Learning (Native Language)

• • Language Teaching and Language Learning (Second Language)

• • Semantics

• • Historical [Linguistics]

• • Stylistics

However, the name ‘section’ is misleading, suggesting as it does some kind of permanent administrative structure for organising these events. Rather, they were a very informal arrangement for classifying papers. A section might nominally have an individual as its convenor, but they all met only sporadically and many of them included just one paper. This was true of all the sections listed above for Spring 1972, and for the meetings we know about, around half the sections consisted of a single paper. Not surprisingly, perhaps, this informal arrangement petered out: the last sections were held at the Spring 1975 meeting, and the following meeting attempted to replace them with an organisation based on a single theme (‘The data of linguistics’) for the whole conference – an experiment that was never repeated.

In 1975 the LAGB celebrated a joint Spring meeting with the Societas Linguistica Europea by inviting André Martinet to speak, thereby establishing a precedent for an invited speaker at Spring meetings.Footnote ⁸ This role was eventually called the Linguistics Association Lecturer, to distinguish it from the Henry Sweet Lecturer (for Autumn meetings) established at the Business meeting in September 1982. The initial idea was that the former would be from outside Europe and the latter from inside, but this principle only survived for a few years. These speaking roles have been filled by a stream of very distinguished international scholars, all chosen by the membership at business meetings, and it is hard to imagine an LAGB meeting now without these keynote speakers.Footnote ⁹ The two roles have survived in the new one-conference pattern, with one invited lecture at each end of the conference.

A related development is the introduction of ‘workshops’ (sometimes known as ‘teach-ins’) on the topic of the invited lecture. The workshop fills the first afternoon of the conference programme, giving the uninitiated a chance to prepare for the keynote lecture in the evening. This very good idea has generally worked reasonably well, though there has been an understandable tendency for the workshop panel to treat their slot as an ordinary conference presentation for fellow experts rather than as a lesson for novices. A workshop in the first afternoon is now an established part of any LAGB conference programme.

Another successful innovation was the ‘language tutorial’, which was inspired by a tutorial on Mayali given by Nick Evans at the Australian Linguistic Institute. The LAGB language tutorials started in 1995 with a tutorial on Basque by Larry Trask. Since then every conference has included a three- or four-hour plenary tutorial (in two or three parts) about the structure and socio-historical background of some unfamiliar language. Those featured so far can be found on the internet history of LAGB, and include a diverse spread ranging from Kayardild, Chechen and Mohawk to Bengali and Welsh.

Finally, we must mention the idea of a ‘themed session’, which has surfaced repeatedly in the LAGB's history. A themed session is a slot in the programme – typically two hours long, enough for three standard papers – which is planned in advance around a particular topic. Such sessions were anticipated in the section meetings of the early years, but it was not until recently that meetings have included one-off events variously called ‘seminars’, ‘sections’ or ‘parasessions’ which have taken over a few hours (either during the meeting or immediately after it) for structured discussion of a topic ranging from ‘alternatives to unemployment’ to phonological theory. Themed sessions are an important way of guaranteeing a place for an area of linguistics which would otherwise not be represented in a normal LAGB programme, and the area where they have played this role most successfully is educational linguistics. The 1970s had the section on Language Teaching and Language Learning mentioned earlier. The 1980s had a new section on educational linguistics, launched with a panel discussion of ‘linguistic equality’ – a topic which attracted more than half of the conference participants (Hudson Reference Hudson1983). And the 2000s have had a regular two-hour session which is included in the remit of a more permanent Education Committee. The LAGB's commitment to education is discussed in more detail in section 6.

The picture that arises from this survey is of a constantly evolving basic pattern for conferences, which changes according to the current needs. The basic unit is still the 25-minute paper with 15 minutes of discussion, but various alternative strategies have been tried for airing new ideas and providing discussion – invited keynote speakers, specialist teach-ins and language tutorials, parasessions on topical issues, ongoing section meetings and linked conferences or workshops. Some of these strategies have become permanent features of every meeting, but even those which have not still remain as options – but only so long as the membership remembers that they exist.

3. The Journal of Linguistics and publishers

The decision to launch the Journal of Linguistics Footnote ¹⁰ was probably the most important one in the LAGB's history, and was taken within the first few years. It was discussed at the LAGB meeting at Bangor in the Autumn of 1963 (Collinge Reference Collinge, Brown and Law2002: 69), and the association gave its formal approval in March 1964, when the meeting was, appropriately, in Hull. The Memorandum of Agreement between the LAGB and Cambridge University Press, the Journal's publisher, was signed on 6 July 1964.

This decision was naturally preceded by negotiations between the LAGB and Cambridge University Press. The archives contain no formal records of these negotiations or their background, but according to John Lyons (p.c.), a great deal of credit for the initiative belongs to Michael Black at CUP. More generally, ‘Michael Black initially, and following him at CUP Jeremy Mynott and Penny Carter, was immensely important in the development of linguistics in Britain from the early 1960s’. It is important to remember the contribution of publishers to the well-being of our subject.

We do not know who first thought of launching a journal, but in view of Lyons's comments it may have been Michael Black's idea. The attraction of the journal for CUP was presumably the opportunity to break into the expanding field of linguistics (until then dominated by Longman, which had, for example, invested heavily in Quirk's Survey of English Usage – Quirk Reference Quirk, Brown and Law2002: 244). If this was the motivation for CUP, it was certainly successful, as the journal quickly established itself as one of the leading international linguistics journals and achieved a significant circulation. No doubt it also helped CUP to become a major player in publishing linguistics books.

From the LAGB's point of view, the new arrangement had obvious attractions. It put the association into the same league as not only PhilSoc but (more importantly) the Linguistic Society of America – an academic association with its own journal. The LAGB retained some control over the journal's policy by appointing the editors, and in my opinion the association should feel some collective pride in the way this policy has been directed. The first editor was John Lyons, who (by his own admission) had until then had nothing to do with the LAGB (Lyons Reference Lyons, Brown and Law2002: 190). In his own words,

Peter Matthews, who had serious reservations about the LAGB but was himself later appointed as one of the Journal's editors, agrees:

As an example of the intentional breadth of theoretical outlook, the first few volumes included an exchange between Chomsky/Halle and Householder by invitation from Lyons (who had met them both during his time in the USA – Lyons Reference Lyons, Brown and Law2002: 190), as well as a three-part article by Halliday, once again solicited by Lyons (p.c.). This theoretical openness has always been a hallmark of the Journal (as it has, in my experience, of LAGB conferences). Equally impressively, the Journal has always been allowed complete independence from the LAGB itself, so that (unlike the Transactions of the Philological Society in the 1960s) it is not primarily a record of papers presented at LAGB conferences (though it does carry a listing of LAGB conference programmes) and (unlike Language Footnote ¹¹) contributors are not expected to be members of the association. The LAGB's only control over editorial policy is through the process of approving the editors. Moreover, the editorial board has come to be selected on an increasingly international basis; indeed, ‘a majority of the current editorial board are non-British and based outside the UK. In contrast, in the late 80's all were British and only one based outside the UK (Bernard Comrie)’ (Bob Borsley, p.c.).

Since its first issue in 1965, the Journal has been fortunate in having a great deal of continuity in its editors:

• • 1965: John Lyons

• • 1969: Frank Palmer with Peter Matthews and David Crystal

• • 1979: Eric Fudge with Paul Werth and Nigel Vincent

• • 1983: Nigel Vincent

• • 1994: Bob Borsley and Ian Roberts

• • 1998: Bob Borsley with Nigel Fabb and Ian Roberts

• • 2000: Bob Borsley with Nigel Fabb and Caroline Heycock

• • 2004: Nigel Fabb and Caroline Heycock with Bob Borsley

In 1993 the role of Review Editor was created and filled by Maggie Tallerman until she was replaced in 2005 by Kerstin Hoge. In the early years, the editor had a great deal of personal discretion in selecting papers or even inviting them (as Lyons did so successfully). In many ways, this discretion matched that of the LAGB president in selecting or inviting conference papers. However, in both cases practices and principles have changed to cope with an increasing number of submissions (for instance, in 1999 (vol. 35) the number of issues per year went up from two to three) and higher expectations of neutrality and accountability, so that the editors now apply the usual international standards of blind peer-refereeing to submitted articles, with an acceptance rate of about 23%. No doubt these changes are inevitable and ultimately for the good of the discipline. On the other hand, the editors still have some flexibility, so that they allow themselves to encourage or even invite submissions (such as the present article), though even here peer review applies.

The LAGB's special relation to CUP has benefitted both sides. Most obviously, LAGB members receive the Journal for a considerable discount, but there have been smaller benefits for the LAGB as well. For example, in 1999 CUP made us a free poster advertising linguistics, for use on occasions such as open days and visits to schools. The downside of this comfortable relation, of course, is the danger of missing opportunities for cooperation with other publishers. A potential example arose in 1987 when (according to the minutes of a committee meeting on 25 September 1987) another publisher approached the LAGB with an offer to take over the Journal on terms that might have been more advantageous to the LAGB. This takeover turned out not to be possible as the Journal belongs to CUP, but no doubt the bid helped to motivate CUP to agree to the new and more favourable terms which were reported to the Annual General Meeting in September 1989 without reference to this other bid: ‘For some time we have been negotiating with CUP the terms on which JL is published’. According to committee minutes (April 1989), CUP had improved its terms considerably by, for instance, raising its discount for LAGB members from 25% to 45%.Footnote ¹² It would seem that behind-the-scenes competition may have worked in our favour, and although in 2008 CUP remains the publisher of our journal, the LAGB also has cordial relations with other publishers, who have been happy to display their wares at LAGB conferences since 1972 and even to fund the occasional wine party. Long may it continue so.

4. Members, officers and communications

The first record of a formal constitution is a revision made in 1976. This was further revised in 1987 and 1990, and another revision is under way in 2008; the current one can be found on the LAGB website. What the constitution says about membership is that it is allowed to ‘any person with an interest in linguistics or related disciplines’. Membership is not obligatory for attending conferences, though a small surcharge is applied to non-members; nor, as noted above, is it necessary for submitting papers to the Journal of Linguistics. However, it does carry enough advantages to compensate for the cost (which is deliberately kept to a minimum: £2 in 1980, £5 in 1990, £20 in 2004 but with a lower rate of £10 for students and the unwaged). Most members take advantage of the considerable discount for the Journal, which more than covers the cost of membership; student members are eligible to apply for conference grants; and of course members receive regular mailings about conferences and other matters. On a less mercenary and practical level, for many of us the main motivation for membership is our affection for the LAGB as one of the important forums where our careers developed.

The membership had already reached 672 by 1971, and has been fairly static for some decades at around 600 (606 in 1980, 660 in 1989, 583 in 2005), in spite of some variation in its composition over the years. For instance, the proportion of members who subscribed to the Journal fell from 74% in 1980 to 49% in 1998, but recent years have seen a steady rise through 57% (2000), 65% (2002) and 70% (2006). The proportion of members who are students (or unwaged) has also risen recently, from 16% in 1998 through 21% in 2002 to 24% in 2005; in contrast, the proportion who live overseas has stayed roughly constant during these years, at 37% in 1998 and 2002, falling slightly to 34% in 2005. Unfortunately there is no way at present of knowing whether these figures should be considered satisfactory or disappointing, either in themselves or in relation to earlier times. We don't even know whether our membership represents a roughly constant percentage of the nation's total population of academic linguists, let alone whether we have about the right share of the ‘market’, given the other alternative associations.Footnote ¹³

Turning to the committee structure, the 1976 constitution provided for the following roles:

• • Chairman (renamed ‘President’ in the 1987 revision)

• • Secretary

• • Assistant Secretary

• • Treasurer

• • Membership Secretary

These roles still exist, but others have been added, notably that of Meetings Secretary. This role was not added to the constitution until the 1987 revision, but it was present de facto much earlier – indeed, from 1977 to 1985 there was de facto even an Assistant Meetings Secretary. (From 1985 to 1997 the committee also had a place for the ‘BLN Editor’, a role which is explained later in this section.) The resulting six-role committee worked well and is still with us. The first appendix lists the individuals who have carried out the very large amount of work that these roles demand, and on which the LAGB depends totally. More recently, the 2007 meeting agreed to set up a postgraduate conference committee, whose chair will ex officio have a place on the LAGB committee. It will be interesting to see whether, and how, this addition changes the dynamics of committee activity.

One of the issues in any large organisation such as the LAGB is communication, and especially communication between the committee and the membership. Until recently, the only communications from the committee to all members concerned conferences, each of which generated two mailings: a first circular announcing the conference and inviting abstracts for papers, and a second one announcing the conference programme and inviting bookings. This pattern dates back at least as far as 1967 and still persists, though nowadays of course the circulars are distributed by email except to the minority of members without email addresses. However, other forms of communication have existed for some time. In 1981 the LAGB persuaded an enthusiast, Marion Owen, to run an ‘employment exchange’ with the aim of advising linguists on possible careers and listing job opportunities. And in 1984 the LAGB took over the British Linguistics Newsletter, a monthly news-sheet about linguistics which had been started in 1972 by another enthusiast, Jim Hurford, who later handed it over to a series of other volunteer editors. BLN eventually took over the role of the Employment Exchange until the BLN itself disappeared in 1997 with the arrival of email; then job advertisements were included for some years in conference circulars and finally disappeared when the internet made the remains of the Employment Exchange redundant.

The internet has now opened further possibilities which the LAGB has exploited reasonably fully – a website and a members-only email list where most traffic advertises jobs and conferences in the UK, but which is useful for consulting members on LAGB matters such as the present article. Paradoxically, one of the positive effects of the internet is to make communications not only faster but also more permanent, so we may hope that key documents such as minutes and conference programmes will be archived more successfully in the next fifty years than they have been in the first fifty.

5. Other linguistics associations

The LAGB has never been the only professional association for linguists in the UK. As noted in section 1, it was not the first such association. The Philological Society was already ancient when the LAGB was founded in 1959, but the LAGB is also pre-dated by a specialist association for phoneticians, the British Association for Academic Phoneticians (always known as ‘BAAP’ rather than ‘the BAAP’), whose websiteFootnote ¹⁴ lists meetings as early as 1958. Nor was the LAGB the last association. A number of other associations or looser organisations have become established since 1959.

The first addition was the British Association for Applied Linguistics,Footnote ¹⁵ established in 1967. The intellectual background was the growing interest in new methods for teaching languages which had culminated in the foundation of AILA (Association Internationale de Linguistique Appliquée)Footnote ¹⁶ in 1964 and in the potential contribution of general linguistics. The early history of what is again always called BAAL (rather than ‘the BAAL’, in contrast with ‘the LAGB’) is described as follows in its official history:

It is interesting to note the close links between BAAL and the LAGB, not only through the Language Teaching and Language Learning sections mentioned in section 2 above but also through the inclusion of two leading LAGB members, Michael Halliday and Frank Palmer, in the working party that prepared for BAAL.

BAAL has now grown somewhat larger than the LAGB, with over 800 members and conference attendances of more than 300. This can be explained in part by the large number of people who work in language education and the focus on education in the 1974 BAAL constitution:

However, although one might expect and hope that theoretical and descriptive linguists would choose the LAGB while those with more practical concerns such as education would choose BAAL the choice between the associations does not follow quite these lines.

Although the LAGB's constitution defines its remit as the whole of linguistics, without restriction to theoretical or even descriptive linguistics, in practice there are large areas of descriptive linguistics which are much more likely to be found in BAAL conferences because the LAGB is nowadays widely perceived (whether rightly or wrongly) as having lost interest in them since the 1980s and 1990s, when they all featured in LAGB programmes. To take the 2006 BAAL conference as an example, it included entire sections devoted to corpus linguistics, the ethnography of language, language and gender, and psycholinguistics, as well as individual papers on these topics and on bilingualism, discourse structure, first-language acquisition, attitudes to regional varieties and metaphor. All of these topics are in principle within the LAGB's remit, and the papers concerned did not appear to have any particularly close link to education which might have made BAAL more appropriate. In contrast, the same topics were totally absent from the 2006 LAGB programme, which in this respect was quite typical. LAGB conference papers tend strongly to focus on language structure rather than on methodology (e.g. corpus linguistics, ethnography), language variation (e.g. language and gender) or language use (e.g. discourse structure). Why the two associations have evolved in this way, and whether the resulting situation is healthy for our discipline, is a debate that the LAGB and BAAL should engage in.Footnote ¹⁷

BAAL is not the only organisation to provide coverage of more specialised research areas of linguistics. On the one hand there are international organisations which provide more specialised conferences, whether defined in terms of theoretical assumptions as in the case of Generative Linguists in the Old World (GLOW), the International Systemic Functional Congresses, or the International Cognitive Linguistics Association; or in terms of research areas as with New Ways of Analyzing Variation (NWAV), the Poetics and Linguistics Association (PALA), and the forty or so other conferences every month that appear on the Linguistlist's international listing of events.Footnote ¹⁸ On the other hand there are also more local conferences, ranging from conferences for postgraduate students to the UK Cognitive Linguistics Association, the Sociolinguistics Symposium and its offshoot, the UK Language Variation and Change conference, and the Henry Sweet Society (for the history of linguistics); particularly significant newcomers are the various associations for the linguistics of particular languages (e.g. the Association for French Language Studies, founded in 1981, and the annual Welsh Syntax Seminars dating from 1992) or language families (e.g. the Romance Linguistics Seminar, which has met annually since 1973). Given the range of alternatives on offer, it is perhaps not surprising that some LAGB conferences have struggled to attract participants.

This proliferation of specialised associations raises an important political question for linguists in the UK: who speaks for linguistics? This question has arisen repeatedly since at least as far back as the 1970s, and becomes urgent whenever the field needs a single mouthpiece or forum. As a result, a series of ‘umbrella’ organisations representing all the three main associations (and sometimes others as well) have opened up for a time before closing for lack of business. The first was the British National Committee for Linguistics (BNCL), founded in 1977 to represent PhilSoc, the LAGB and BAAL (Mitchell Reference Mitchell1997: 12). This committee was initiated by the British Academy, which also provided meeting rooms and a secretary, and its chair throughout its life was John Lyons; but according to the announcement on the inside cover of the Journal of Linguistics 13.2 (1977), it had an international rather than a national remit:

It met regularly twice a year and had its own budget (provided at least in part by the Academy) and high-level representatives – two from each association, three from the Academy and the Director of the Centre for Information on Language Teaching.Footnote ¹⁹ Its main business concerned the European Science Foundation, where John Lyons represented both the British Academy and the then Social Sciences Research Council (for whom he also happened to chair the panel responsible for linguistics). For example, the BNCL helped select participants at two ESF Summer Schools in linguistics (one on sociolinguistics at the University of Sussex in 1982, the other on psycholinguistics in Brussels in 1985), and also tried to arrange some funding for student bursaries (which resulted in bursaries from BAAL and a commercial organisation). It also appears to have provided some Academy funding for the subscription paid by the LAGB and PhilSoc to the International Permanent Committee of Linguists, and agreed (in 1983) to do the same for BAAL's subscription to the International Association for Applied Linguistics. However, the Spring meetings in 1983 and 1984 were cancelled for lack of business, and the last regular meeting was held in the Autumn of 1984. Possibly there was insufficient international business to keep such a grand committee occupied.

Meanwhile, a very different body had been set up on the initiative of a group of linguists who had a more specific interest: education. Some of these linguists were members of the LAGB and some of BAAL, so they proposed a joint committee of the two organisations, the Committee for Linguistics in Education (CLIE). This was founded in 1980 (on the basis of a steering committee set up in 1978 after a joint LAGB/BAAL workshop), and in 2008 it is still active, with three meetings every year, plenty of business and a website.Footnote ²⁰ (The next section gives more details about its activities.) It is interesting to compare CLIE and BNCL and to wonder why one survived while the other died. Whatever the explanation, CLIE shows that at least two of the associations can sustain an ongoing joint enterprise which can speak for both of them.

In 1988 a single body was needed to speak for linguists and phoneticians in reaction to the introduction of research assessment (at that time called the Research Selectivity Exercise, RSE, but later called the Research Assessment Exercise, RAE). One idea discussed at the July meeting of the LAGB committee in that year was to use a revived BNCL, but members expressed a rather odd ‘disquiet … at having LAGB representation swallowed up by the larger organisation’. The LAGB's Business meeting in September was similarly lukewarm about the BNCL; indeed, there was even some discussion of ‘pulling out’ of BNCL altogether but in the end the meeting decided to appoint new representatives and stay in. A single meeting of BNCL was called in November to discuss the RSE, but the meeting (if it ever happened) had no documented outcome.

The natural alternative, apparently, was to create a different joint body, described in a letter (dated 12 April 1988) from the LAGB secretary to the BAAL and PhilSoc secretaries as ‘some kind of integrated or coordinated approach to the question, for example in the form of a joint working-party’. The LAGB Business meeting accepted the idea, though this presumably carried exactly the same risk of being ‘swallowed up’ as with the BNCL. The LAGB selected six representatives for this body, which at first called itself ‘a standing committee of the linguistics societies’ but was later called the ‘Joint Committee of the Linguistics Societies’ and finally (and rather oddly) the ‘Committee of the Joint Societies for Linguistics’ (CJSL). However, when the new committee met in December it decided, first, to nominate just three individuals to represent linguistics: one for the LAGB and PhilSoc, one for BAAP and a third for BAAL; and second, to collect a great deal of data. It also turned out that the other societies had each nominated only two representatives to CJSL, and by 1991 the LAGB also had just two.

CJSL did a great deal of work in preparing for the first research assessment exercise in 1989; the main achievement was the production, in late 1988, of a monumental listing which tried to include every linguist in the UK, together with a brief note on their area of activity. (The LAGB and PhilSoc essentially each took on half of the work, taking linguists, respectively, inside and outside named linguistics departments.Footnote ²¹) CJSL also negotiated about the 1992 RAE and commented in December 1991 on the Economic and Social Research Council (ESRC) guidelines for postgraduate training. However, one of the limitations of CJSL emerges from a letter written in February 1990 about the costing of linguistics teaching. The president of the LAGB could write this letter on behalf of PhilSoc and BAAP as well as the LAGB, but could not include BAAL because (in principle at least) BAAL's members did not teach linguistics. This is presumably why CJSL is not even mentioned in the letter. The general problem is that different issues may be relevant to slightly different combinations of societies, so a single umbrella organisation needs to be sufficiently flexible to speak for different societies on different occasions. No doubt this can be arranged, but CJSL may not have found the right formula. At any rate, CJSL soon followed BNCL into oblivion. The last mention of it is in the first LAGB Spring circular for 1992, when the LAGB AGM's agenda included a CJSL recommendation for ‘a general national committee for linguistics’ – which the minutes ignore, noting that ‘it seems that on the whole the societies wish to act individually, for instance in the matter of submitting nominations for a panel of assessors to the UFC [Universities Funding Council]’.

The same meeting welcomed the establishment of the Association of Heads and Professors of Linguistics (AHPL), which represented linguistics departments rather than professional associations. While AHPL was chaired by its founder, Neil Smith, it met intermittently for some years – often as an appendix to an LAGB meeting, such as the Spring 1992 meeting where it is mentioned in the second circular – and produced a large database of linguists and linguistics department called ‘DABLING’; but it never established itself as a permanent mouthpiece for linguistics and may in some ways have undermined CJSL. After the chairperson changed it never met again.

The demise of BNCL, CJSL and AHPL left linguistics without the single voice that would have been so helpful in dealing with outside agencies, and that most other comparable subjects do have. The lack was filled by a series of ad hoc meetings called by other bodies; for example, the British Academy's linguistics section has brought the various associations together from time to time to discuss the composition of the linguistics panel of the RAE, and the linguistics officer at the ESRC has convened meetings to discuss its policy on research funding. Such ad hoc meetings are no substitute for a permanent body in which the associations can build relations and develop policy without fear of losing their individual voices. It is pleasant to end this part of the history by noting the possibility of a solution. Since 2000, the LAGB has been actively involved in the government-funded ‘Subject Centre’ for Languages, Linguistics and Area Studies (LLAS),Footnote ²² particularly through its specialist advisory group for linguistics. In 2005 the Subject Centre set up a group that came to be called the ‘Linguistics Strategy Group’, consisting primarily of representatives from all the linguistics associations, to consider strategic issues such as student numbers. The associations agreed to support this group and the Subject Centre offered managerial help and meeting rooms. The February 2008 meeting of the group agreed to propose to the associations a more permanent body whose provisional title is still being negotiated. It remains to be seen whether this will fare better than its predecessors.

6. Representing linguistics

In spite of these ambiguities about who speaks for linguistics, the LAGB has often assumed this role either by taking the initiative on some public issue or by establishing itself as a body to be consulted on matters to do with language. The issues covered have ranged from Value Added Tax (VAT, a tax which at one time applied to academic conferences), through research assessment, to the defence of individual persecuted linguists; but the most consistent area of activity is education. I review these activities briefly below.

The campaign to exempt learned societies from VAT was headed by the Royal Society, whose Joint Committee on Learned Societies contacted 300 such societies, including the LAGB, in 1975, in order to find out what problems they faced. Our archives suggest that the LAGB was a passive partner in this campaign, and that it merely received a series of communications from the Society. However, the VAT issue did not go away, and the LAGB found itself dealing directly with Her Majesty's Customs and Excise in 1990 in connection with the VAT status of one of its conferences. This issue illustrates the need for learned societies to work together at various levels of collectivity, including a level where their particular subject specialism is irrelevant.

Somewhat more relevant to linguistics was the LAGB's temporary involvement in the Association of Learned Societies in the Social Sciences (ALSISS), where linguistics is highly relevant as a bridge subject between the humanities and the social sciences. (In the British Academy, the Linguistics and Philology section is the only one which straddles the Academy's two-way division into Humanities and Social Sciences.) ALSISS, which had been founded in 1982, invited both the LAGB and BAAL to join in 1991, but although both societies accepted, giving our subject double representation, linguistics turned out to be a very small player in a field dominated by large subjects like geography and economics. In 1993 the LAGB nominated a member to the ALSISS committee, but soon gave up this place and active participation in ALSISS even when it turned into the Academy of the Social Sciences in 1998, and the LAGB seems to have withdrawn entirely in 2000.

The LAGB has also engaged in even more directly relevant issues. One such case was an approach by the British Academy in 1994 which was addressed to the LAGB's president (though, at least on paper, CJSL still existed at the time), asking him to recommend people for selection panels of the newly established Humanities Research Board. The invitation was hardly surprising given that linguistics was well represented on the HRB itself by its chair, the phonetician John Laver, and an LAGB member, Nigel Vincent; but it was still gratifying to be recognised in this way. Similarly, the LAGB has been consulted on a range of academic matters which called for our special expertise: subsequent research assessment exercises, the ‘benchmarking’ statements about the content of linguistics degrees, exercises in assessing the quality of teaching, and consultations by the research councils on research policy for social sciences and for the humanities. In all these cases, the LAGB's contribution was a reaction to developments that had been initiated by others.

During the 1990s, the association played a more active campaigning role. The main issue in the early 1990s was the status of British Sign Language as a natural language (Smith Reference Smith2005: 50), but in the late 1990s and early 2000s it was the study of foreign languages in higher education, where there was a steady decline in both research and teaching in all languages, including the major languages such as French and German that have historically been the backbone of foreign language study. In 1985 the LAGB joined a small (and unsuccessful) campaign initiated by PhilSoc to press for official intervention to protect less widely taught languages, and in 1999 it was included in a similarly unsuccessful attempt initiated by the British Academy. But in 2003 it coordinated a successful application in defence of the more widely studied languages. The background to this initiative was that the Arts and Humanities Research Board had offered six special Ph.D. awards per year (for three years) for the study of endangered subjects where extra support was needed. The proposal managed to persuade them, on behalf of a range of academic associations (including PhilSoc), that the study of French, German and so on was under threat; but the problem turned out to be even worse than it seemed at the time. In the third year of this scheme, there were so few suitable applicants for Ph.D. awards on modern languages that some of the ring-fenced grants could not be awarded.Footnote ²³ Promoting the linguistics of foreign languages, including the traditional mainstay of language study, remains an important challenge for the LAGB.

A different kind of activity was the defence of individuals and groups, which occupied a considerable amount of presidential time into the 1990s. The LAGB has made representations on behalf of at least four persecuted linguists in other countries,Footnote ²⁴ and in 1996 the President wrote in defence of a community of indigenous Indians in Brazil who (together with their languages) faced extinction at the hands of local landlords; in each of these cases our linguistics turned out to be relevant to the protest. It is hard to know how much can be achieved by interventions from a small academic organisation such as the LAGB,Footnote ²⁵ but it is equally hard to know how it can refuse to make them.

A different kind of operation was mounted in defence of linguistic departments in this country that were threatened with closure; indeed, in 1981 so many linguistics departments were under threatFootnote ²⁶ that the membership directed the Chairman to argue the case for linguistics with ‘the Vice-Chancellors and relevant Deans of all Universities where linguistics is under threat from University Grants Committee recommendations of cuts due to government policy’ (minutes of the Business meeting on 1 October 1981, item 5). Fortunately these general threats receded in a more favourable economic climate where linguistics departments in fact multiplied; but more recently a number of individual departments (Durham, Reading and East Anglia) have received LAGB support when faced with closure. Here, too, it is impossible to know how much effect the association's interventions have had,Footnote ²⁷ given its obvious vested interest in every case.

The last topic is education, the external concern with which the LAGB has had its most long-lasting, consistent and effective engagement. The LAGB has always included members who argued that linguistics has a great deal to contribute to language education, such as Michael Halliday, David Crystal, Katharine Perera, Michael Stubbs and myself (Halliday Reference Halliday2007, Crystal Reference Crystal, Brown and Law2002, Perera Reference Perera1984, Stubbs Reference Stubbs1986, Hudson Reference Hudson2004). These ideas have had a somewhat sporadic but long-lasting influence on conference programmes, starting with the ‘sections’ (see section 2 above) on ‘Language Teaching and Language Learning’. Not surprisingly, the founding of the British Association for Applied Linguistics offered a better forum for some papers in this area, but education continued to have some presence at our conferences; for example, in 1981 the LAGB invited a school teacher, David Cross, to describe his innovative language teaching to the full conference, and structured sessions on educational linguistics established themselves in our conference programmes through the processes mentioned in section 2.

Another interface with education is student recruitment, in which the LAGB has from time to time taken an active interest. In particular, in 1981 the association commissioned a member (Jim Hurford) to write a ten-page booklet about studying linguistics and paid for it to be distributed to 1200 schools. Similarly, the poster that Cambridge University Press produced for us (see section 3) was intended for use at open days and other recruitment events. However, student recruitment has not in general been a major concern for linguists (unlike our colleagues in foreign-language departments).

In contrast, the LAGB's main contribution to education has been made outside our conferences and recruitment activities in a number of initiatives concerned primarily with education in England and Wales; the school systems in Scotland and Northern Ireland, which are organised separately, have tended not to attract the LAGB's attention. The association has been represented on a series of external committees concerned with language education, starting with the National Congress on Language in Education. NCLE was founded in 1976 as an ‘association of associations’ with the aim of ‘drawing up recommendations for policy and action by Central Government, with administrative support from the Centre for Information on Language Teaching’ (Hawkins Reference Hawkins1996: 361). The LAGB was an important member of NCLE because one of the dominant new ideas in language education was a direct application of linguistics to classroom teaching called ‘language awareness’ (Hawkins Reference Hawkins1987). Language awareness is explicit understanding of how languages (in general) work and are related to one another, and it still plays an important part in government policy, especially in foreign-language teaching; for example, one of the five ‘strands’ in the syllabus for teaching languages in Key Stage 2 (upper primary) is called ‘language awareness’. The LAGB supported this movement from the start, and was represented on NCLE until it faded away in the early 1990s.

Meanwhile, the LAGB and BAAL had collaborated in setting up a permanent body of their own to manage the interface between linguistics and education. This started in 1978 with a joint LAGB/BAAL workshop on this topic, out of which emerged a joint steering committee which also included representatives from the National Association for the Teaching of English (representing school English teachers) and from the then Department of Education and Science (representing the educational branch of government). This committee's composition was important because it recognised the importance of building strong bridges from the academic world of linguistics and applied linguistics to the very different world of schools. After two more workshops in April 1980 (one of which was on the teaching of linguistics itself in schools), the associations, at the instigation of the steering committee, created the more permanent body mentioned in section 5 above, the Committee for Linguistics in Education. CLIE is still active, nearly 30 years (and 90 meetings) later, making it by far the most long-lived organisation created by either the LAGB or BAAL. It is also the only organisation in the country whose remit includes the whole of ‘language education’ – first-language literacy, foreign languages, English as an Additional Language and the ‘community’ languages of our new multilingual society. Its membership has gradually expanded so that it now includes representatives from thirteen other associations. There is no longer any representation from a government agency, but CLIE has built good relations with all the relevant agencies through commenting on their initiatives.

Communication between CLIE and its parent organisations has not always been as good – for example, CLIE minutes are meant to go to the LAGB committee, but when the flow stopped in the late 1980s, the LAGB committee took three years to notice the gap. It was partly in order to remedy this problem that in April 2000 the LAGB set up a permanent Education Committee,Footnote ²⁸ one of whose activities is to provide a verbal and written report on educational activities including CLIE at AGMs. The committee also organises the regular education sessions at LAGB conferences mentioned above and responds to relevant government consultations. The present climate in England is rather sympathetic to ideas from linguistics; for example, the Kingman Report (Anon. 1988) on the teaching of English has been described as giving ‘enormous prominence … to linguistics’ (Stubbs Reference Stubbs, Jill and Tom1989: 20), and more recent trends in schools have continued to encourage the explicit study of language (Hudson Reference Hudson2007). In this climate it has been important for linguists to do their best to support positive developments. A number of individual members have played an important part in promoting linguistics in schools by sitting on committees (Michael Stubbs, Gillian Brown) or even in helping to draft the National Curriculum for English (Katharine Perera, whose role is described in Cox Reference Cox1991). And the LAGB itself has also contributed to these developments through its Education Committee, which has responded on the LAGB's behalf to most of the recent consultations about relevant innovations.

The activities just described have all been directed at school-level education, but higher education has also attracted a great deal of attention recently. One issue has been the national crisis in both research and teaching of foreign languages mentioned above, which led to the creation in 1993 of the University Council of Modern Languages, replacing the National Council for Modern Languages; the earlier body had an LAGB representative, though the UCML only has a committee post for linguistics as a whole rather than for any particular association. Another issue has been the quality of teaching in higher education, and since 2000 the LAGB has been actively involved in the Subject Centre for Languages, Linguistics and Area Studies (LLAS), mentioned at the end of section 5 above, particularly through its specialist advisory group for linguistics. The LAGB is involved in this advisory group partly through the individuals on the group and partly through presentations by the group at LAGB conferences, and LLAS is gradually becoming recognised as a key ally for improving the teaching of linguistics in higher education.

7. The future for the LAGB at 50

The LAGB has achieved a certain middle-aged respectability after a rebellious and uncertain youth, and enjoys reasonably good health. Even the lack of memory which this paper tries to remedy is (paradoxically) a sign of youth rather than of age. But what does history tell us about our future? One conclusion is very clear in an otherwise rather cloudy crystal ball: the future cannot be simply ‘more of the same’, because the LAGB itself has already changed a lot and there is no reason to doubt that more changes are in store. Some changes will be imposed on us by external circumstances, but we still have a great deal of control over our own destiny. As I see it, we face a number of strategic choices.

One issue is the LAGB's identity in relation to the other linguistics associations, and especially to PhilSoc and BAAL. How, and why, are we different? A simple answer is that we distinguish ourselves from PhilSoc by organising conferences and from BAAL by discussing language structure; but this is too simple because we do a lot more than organise conferences (as emerges from section 6), and language structure is not the only thing that interests us as linguists. Indeed, one of the striking characteristics of modern linguistics is how much broader and more open it has become during the LAGB's lifetime. The LAGB's present identity is a rather messy product of our history, including some unplanned historical trends such as the narrowing of conference programmes. We should ask whether the outcome is good for individual linguists, for the LAGB, and for our subject. These trends in the UK have distinguished several kinds of linguistics – LAGB linguistics, PhilSoc linguistics and BAAL linguistics, as well as the various kinds of linguistics that are discussed by the UK Cognitive Linguistics Association, the UK Language Variation and Change conference, and so on – but without either a rational basis for the differences between the various kinds of linguistics or an effective interface between them. This kind of division is likely to harm any subject, but especially one which is as small as ours. The choice for the LAGB, therefore, is whether to plan future development more deliberately than in the past. If planning does seem a good idea, it will clearly have to involve all the other societies, which will not be easy because different members and organisations will inevitably have different visions of the future (as can be seen in the exchange in Borsley & Ingham Reference Borsley and Ingham2002, Stubbs Reference Stubbs2002, Borsley & Ingham Reference Borsley and Ingham2003). But however difficult, it will be an important debate.

The LAGB's relation to other associations also raises the perennial question of an umbrella organisation to speak for all UK linguists. Section 5 describes a series of attempts to achieve this aim, starting with the BNCL; but with one exception (CLIE), these cross-association bodies survived less than a decade before fading away for lack of business and interest. These organisations have typically been created from scratch in order to deal with some crisis such as research assessment, so they naturally wither away once the crisis disappears or turns into a new normality. But an organisation representing the whole of linguistics need not be purely reactive, nor need it be tied to a single issue. Instead, it might start to plan the subject's future strategically. It might consider, for example, how to remove some of the anomalies noted above in the relations between the societies. A number of organisational models are worth exploring, depending on the kind of goal selected. One model would be a strategic committee for linguistics consisting of just the chair and secretary of each of the separate associations, with annual meetings and occasional email contact in between; but no doubt many other models are worth considering (including the newly mooted body mentioned at the end of section 5).

Strategic planning for the well-being of linguistics in the UK is a third major challenge for all the linguistics associations, but perhaps especially for the LAGB. One strategic issue is the increasing tendency for linguistics undergraduates to come to us with an A-level in English Language rather than in a foreign language.Footnote ²⁹ Does this trend towards monolingual and English-dominated undergraduate linguistics matter, and if so, what should be done? Another issue is the institutional context of linguistics teaching, ranging from specialist departments of linguistics to departments or schools with a much broader focus such as media studies or education; what is the balance among these possibilities, and what ‘should’ it be? An even more important question is how large linguistics ‘should’ be. At the moment, it is a rather small subject with about the same number of specialist undergraduate students and academic staff as, say, German, Archaeology and Anthropology.Footnote ³⁰ But arguably language plays a similar role in the social sciences and humanities to that which number plays in the natural sciences, so might linguistics not expect to stand alongside mathematics, with a far larger presence both at universityFootnote ³¹ and at school? These are major questions which the LAGB should consider seriously.

In conclusion, the LAGB has evolved a comfortable way of life which suits some linguists and looks sustainable into the foreseeable future; but history teaches us that nothing can, in fact, stay the same. The association has already changed a great deal since its early days, sometimes under the pressure of outside forces but sometimes as a result of positive decisions by officers and members. It has taken the lead on many issues that affect all linguists, and considering the association's relatively small membership it has achieved a great deal both for the subject and for the wider society. But this leadership requires strategic planning for the future, and the fiftieth anniversary offers a good point for the association to take stock and imagine a range of possible new routes.