Institute of Governance > Online Articles & Papers, by Author > Online Articles & Papers, by Date Published Online > Article: The Survival of the Democratic Intellect: Academic Values in Scotland and England by Lindsay Paterson

The Survival of the Democratic Intellect: Academic Values in Scotland and England

by Lindsay Paterson

Now published as: Paterson, L. (2003), 'The survival of the democratic intellect: academic
values in Scotland and England', Higher Education Quarterly, 57, 67-93.

• Home
• News & Events
• About the Institute
• Consultancy & Research
• Publications
• Parliament News
• Online Articles & Papers
• by author
• by date online
• Internship Programme
• Site Map
• Search
• Contact

• Scottish Affairs web site

• Acknowledgements
• Abstract
• Introduction
• Data and Methods
• Analysis
• Conclusion
• References

Read the Technical Note which accompanies this paper (80 kb)

Acknowledgements

I am grateful to Frank Bechhofer, Ross Bond, David McCrone and Michael Rosie for comments on a draft of the paper. The survey was funded by the Leverhulme Trust, as part of its 'Nations and Regions' research programme. The conduct of the survey was greatly helped by the Leverhulme project administrator, Margaret Macpherson. Population data for checking the representativeness of the sample were obtained from the Individualised Staff Records 1999-2000 supplied by the Higher Education Statistics Agency. HESA cannot accept responsibility for any inferences or conclusions derived from the data.

Abstract

Using the results of a sample survey of academics in higher education institutions in Scotland and England, the paper assesses attitudes to the civic role of higher education. It places these in the context of debates about core academic values, about the public accountability of higher education institutions, and about the devolution of political power in the United Kingdom. It finds that there is widespread attachment to a civic role for higher education, alongside strong attachment to traditional academic values. These values are not significantly influenced by individual academics' gender, age or social class of origin, and differences by academic discipline are not as strong as might be expected. Academics in Scotland tend to a somewhat more civic view than academics in England. This national difference seems to be a product of distinctive national systems, since academics of English origin in Scotland share in the majority Scottish views.

View the questionnaires:

• English Questionnaire - long (PDF document)
• English Questionnaire - short (PDF document)
• Scottish Questionnaire - long (PDF document)
• Scottish Questionnaire - short (PDF document)

Read the Technical Note which accompanies this paper

Introduction

The proper public role of universities has been a matter for public debate since universities were founded, and is being intensified at the moment by the rapid development of mass systems of higher education in most European countries (Bender, 1988; Clark, 1983; Kerr, 1990; Scott, 1990, 1995; Smith and Webster, 1997; Sommer, 1995). On the one hand is the view that universities are universal. Their academic staff owe their allegiance only to the disinterested pursuit of knowledge, and their primary loyalty ought then to be to the disciplines into which research and teaching have traditionally been organised. To pursue these goals requires that they have intellectual autonomy, and that the institution which employs them is insulated from political pressure. If a question of social purpose arises, the response is that freedom of research and teaching is ultimately for the good of a liberal society.

On the other hand, there is the unavoidable historical fact that universities are socially embedded. Nearly all of the higher education institutions currently in existence in Europe and North America owe their origins to explicitly political acts by public authorities - churches and cities mostly in the late Middle Ages or Renaissance, cities and states in the eighteenth and nineteenth centuries, and highly interventionist states from the early twentieth century onwards. This political activity has had primarily social goals in mind: educating a future ruling class, and providing equal opportunities of entering it; maintaining and developing the nation's culture; providing people and expertise that might help the national economy to compete globally. The history of the university and - even more - of the higher education college in the last century and a half has been at least as much about nationalism as it has been universal.

There is a tendency in writing about academics themselves to give greater emphasis to the aspirations to universalism (for example, Kerr, 1990), and there is a strong tendency for academic managers to try to persuade politicians that disrupting universal values would ultimately harm the nation's interests. In reply, there comes the accusation of betrayal: especially from nationalist and socialist politicians since the middle of the nineteenth century, there has been the allegation that academic autonomy is a dignified term for indefensible privilege. When the same view was taken by governments of the New Right in the 1980s, and when, moreover, this coincided with the first steps in the current rapid expansion of higher education, the traditional defence of disinterested autonomy found very few allies outside the universities themselves. The resulting crisis of identity in higher education has been characterised by Halsey as 'the decline of donnish dominion' - the unavoidable surrender to the demands of society and the state by people who feel this to be a threat to their whole way of life:

The decline of donnish dominion was written in the stars from the moment that the ancient civilisations of the Near East began to form an abstract alphabet. Literacy thereby became potentially a democratic possession. Access to human capital became intrinsically available to all. (Halsey, 1992, p. 258)

In the 1980s, 'the attack on academic autonomy ... has been conspicuously aggressive'. Halsey further comments, implicitly aligning academic culture with the straightforward defence of autonomy, for all that he acknowledges the legitimacy of the attempts to hold academics accountable:

The counterattack [by academics] has been surprisingly mild and perhaps this is a tribute to ... tolerance of hostile ideology and ... patience to pursue reasoned argument. (p. 270)

In fact, however, academics' own values have never been straightforwardly in favour of autonomy or even of universalism. They have acceded to political pressure for over a century, not only because it brings funds, but also because they have not been averse to sharing the rooted local, regional or state values that the political system and the social demands articulate (Neave, 1979; Neave and Rhoades, 1987; Paterson, 2001). The debate about academic values in Scotland and England since the mid-nineteenth century illustrates the multifarious character of academic identity quite well (Anderson, 1992; Paterson, 1998). In contrast to the essentially private traditions of Oxford and Cambridge, the four ancient Scottish universities in St Andrews, Glasgow, Aberdeen and Edinburgh were founded for public purposes and, by the time of the various inquiries into the universities in the Victorian period, had come to accept themselves as being public and national institutions which the state had every right to govern and reform. Academic work was a public service, the apex of a similarly public system of national schooling. According to the dominant epistemology of the Scottish universities, knowledge itself was public, a matter of clarifying and making rigorous the 'common sense' of society. The whole body of belief was later called 'democratic intellectualism' (Davie, 1961, 1986).

In these respects, the Scots' own image of their universities resembled that to be found in Germany, France and the USA a lot more than it did the older institutions in southern England (Anderson, 1992). But then matters became more complicated still, by the infusion of civic values - some coming directly from Scotland - into the new universities founded in mainly northern England in the nineteenth and early twentieth centuries (Scott, 1990). In both Scotland and England, the subsequent creation of colleges and polytechnics gave further impetus to the democratic intellectualism, just as the academic leadership of the mid-twentieth century British universities was apparently forgetting their origins in the rooted and the particular, and welcoming finance from the central state through the University Grants Committee that had been founded in 1919. By the 1960s, and the Robbins expansion, the values of the democratic intellect were being held up in Scotland as the measure of a truly authentic academic culture, against the allegedly anglicising - indeed English - staff of the original four universities. Inspired by the writing of George Davie, there was also the view that the humanism of the old democratic intellectualism had been debased by English utilitarianism, reaching a nadir, it was argued, in the succumbing of the universities to the vocationalism and instrumentalism of UK governments after the 1970s.

In fact, then, academic culture has been rather more complex than the straightforward opposition of autonomy and state control would suggest. Given this complex legacy, given the different recent ethos of some of the newer universities (whether dating from the 1960s or from the 1990s), and given the recent massive expansion, what does academic culture look like in Scotland and England? Is it resistant to accountability? Is Scotland still the home of democratic intellectualism, despite the ideology of universalism that seemed to engulf its universities in the middle of the twentieth century? By means of a survey of academics working in higher education institutions in Scotland and England, we investigate these questions in four parts:

1. What are the attitudes towards the various components of democratic intellectualism - a civic role for higher education institutions, their relationship to the state and to the national culture, the social usefulness of academic knowledge, and democratising access?
2. How, if at all, do these values differ between academics working in Scottish institutions and those working in England?
3. Where there are differences between Scotland and England, what can this tell us about the processes that might maintain democratic intellectualism? Is it a product of the rise to prominence of recently founded universities, or can it be discovered also in the older institutions? Are any differences affected by academic discipline or by the kinds of academic and non-academic networks in which academics operate?
4. Are the values that might be distinctive of Scottish or English higher education an expression of the institutions and higher education systems, or are they the result of the values inherited by the place of education of their academic staff? Is there evidence to support the claim that staff of English origin in Scotland have anglicised the academic values of Scottish higher education institutions?

Data and Methods

The data come from a survey of academics' views about higher education and national identity, carried out in spring 2001. Full details of the sampling, of the representativeness of the achieved sample, and of the questions asked of respondents are in a technical note.

The sample consisted of academics working in higher education institutions in Scotland and England, in roughly equal numbers (to allow reliable comparison) even though Scotland contains only about 12.5% of all academics working in Scotland or England (HESA, 2001). The original intention was to follow Halsey (1992) by selecting the sample from the Commonwealth Universities Yearbook (ACU, 2000). Since Halsey conducted his survey in 1989, this has been expanded to include the former polytechnics and colleges, and does provide a fairly thorough coverage of these higher education institutions. However, the Yearbook has stopped listing staff below the level of senior lecturer or equivalent. So the Yearbook was supplemented by institutional web sites to construct a clustered sample. In each country, around 100 departments were selected from the Yearbook with probability approximately proportional to the size of their list in the Yearbook. The full list of staff in these departments was then obtained from the web sites, and 7 staff were selected at random from each department. The target sample was therefore a two-stage cluster sample, with around 200 clusters and around 1400 units. This would be a self-weighting sample if the Yearbook had listed all staff (or if each department had had the same proportion of its staff listed); design weights are used in the analysis to compensate for the extent to which it is not self-weighting (Groves, 1989, p. 257). To allow for the clustering, the main results of the analysis reported below were checked by multi-level modelling, with departments as clusters (Goldstein, 1995, pp. 97-112). None of the results was affected by the clustering.

The topics asked about in the questionnaire covered views about the purposes of higher education, about accountability, and about general political matters, and also questions of a demographic nature. Some questions were replicated from Halsey's survey, to allow comparison over the decade since his fieldwork.

A full questionnaire of ten pages was sent out at the end of March 2001, a reminder was sent three weeks later, and the questionnaire was sent again after a further three weeks. This produced a response rate of 55.4% (58.7% in Scotland and 52.1% in England). A two-page version of the questionnaire was sent three weeks after the second sending of the long one, and this raised the overall response rate to 63.4% (66.4% in Scotland and 60.5% in England). A response rate of this level is respectable for postal surveys, and - for the long questionnaire - is almost the same as that achieved by Halsey in 1989 (55.8%: see Halsey, 1992, pp. 273-4). This yielded a total achieved sample of 830 (434 in Scotland and 396 in England). Fifteen individuals had removed the identifying code on the questionnaire, and so could not be assigned to departments. Of those which could be identified, responses came from 97 departments in Scotland and 100 in England (out of the 101 and 104 that had been selected from the Yearbook). These were from all the institutions that had been represented in the target sample: 14 in Scotland and 57 in England. Fifteen individuals were removed from the final sample because the design weights for them were very large: they were from departments where the Yearbook listed only one contact person. This entailed removing one whole institution from the Scottish sample.

The representativeness of the achieved sample was checked by comparing it with data supplied by the Higher Education Statistics Agency (HESA, 2001); full details are in the technical note. The sample was unrepresentative in two notable respects. It under-represented contract researchers, containing only about 13% from that category whereas the HESA data contained about 30%. Almost certainly, this is because the web pages that were used for the second stage of sampling did not list all contract research staff: those who were in the sample turned out to be fairly representative of contract researchers in the HESA data in the other respects listed below. The second source of unrepresentativeness was in England only, with respect to date of founding as a university: the institutions that were polytechnics or colleges before 1992 made up only about 15% of the sample but about 30% of the HESA data, because the response rate was lower in them than elsewhere.

In several other respects, the sample was representative of the population separately in Scotland and in England, once the under-representation of contract researchers had been taken into account. The sample and population were close in terms of age, gender, length of service in present institution, and academic grade (eg lecturer versus professor). The sample was broadly representative with respect to subject area, although scientists and engineers were somewhat under-represented, and arts and humanities scholars were somewhat over-represented.

For these reasons, all analysis had to be checked after controlling for grade of staff (contract researchers versus the rest), date of foundation (post-1992 versus the rest) and broad subject area. Few results of the comparison between Scotland and England were affected by this, probably because the extent of unrepresentativeness was much the same in the two countries.

Comparisons of views between Scotland and England are tested by loglinear modelling: that is the source of the levels of statistical significance shown in Tables 1-3, 5-8, 10 and 11. Where these differences are further tested controlling for various characteristics (eg academic discipline of respondent), the modelling is logistic. The dichotomous dependent variable is then the categories 'very important' and 'quite important' versus the rest in most of Tables 1 and 5-8, the two exceptions being the measures 'preparing people to challenge authority' (Table 1) and 'educate academic high-fliers' (Table 8), where the dependent variable is 'very important' versus the rest. In Table 2, the dependent variable is 'strongly agree' or 'agree, with reservations' versus the rest. In Table 3, the dependent variable is 'very loyal' or 'fairly loyal' versus the rest. In Tables 10 and 11, the dependent variable is 'yes' versus 'no'.

Analysis

Academic Values

Table 1 shows the respondents' attitudes to various civic roles which higher education institutions might take on - advising policy makers, engaging in public debate or public teaching, and preparing students to take on leading positions in government and society. The striking first point is that, for all these, very clear majorities in both Scotland and England were willing to say that higher education institutions do have a legitimate civic role. Indeed, in most respects substantial minorities (around four out of ten) regarded such civic contributions as 'very important'. The second point is that, where this is a difference between Scotland and England, academics in Scotland were somewhat more in favour of a civic role than academics in England. Thus academics in Scotland were more inclined than those in England to say that students should learn to contribute to the life of the community, and should learn how to challenge people in positions of authority. Academics in Scotland were also more inclined than those in England to believe that higher education should prepare its students to take on leading positions in government and business - whether that is in Scotland (a very clear contrast with government in the English regions) or, a weaker difference, at the UK level.

Table 1: Attitudes to Civic Role of Higher Education

• Unweighted sample sizes: * 372 Scotland, 340 England; ** 421 Scotland, 394 England.
• † Different versions of question in Scotland and in England.
• *** Tests whole response distribution, not just rows shown in Table.

There was also a readiness to acknowledge the legitimate role of the state in holding higher education institutions to account in certain specific ways (Table 2). People in Scotland were more in favour of government monitoring the expenditure of higher education institutions than people in England (75% as against 64%). Staff also accepted appraisal: only around a quarter were sceptical of it, and very large majorities accepted it (76% in Scotland and 70% in England). There was majority feeling in both countries that the management of higher education institutions ought to be made more business-like. So, although the respondents to this survey also followed those to Halsey's in 1989 in feeling that 'higher education institutions should be less under the control of central government', they did not reject accountability outright. Indeed, where direct comparison with Halsey's survey is possible - for the pre-1992 universities only - the level of resentment at central government control had fallen: in 1989, 52% (across the whole of Britain) strongly agreed that there was too much central control, whereas the proportion in 2001 was 27% in Scotland and 32% in England.

Table 2: Attitudes to Accountability of Higher Education

• Unweighted sample sizes: 372 Scotland, 340 England.
• * Tests whole response distribution, not just rows shown in Table.

Staff also showed strong elements of loyalty to society, and seemed to find no difficulty in combining that with loyalty to their own disciplines or institutions. The results are in Table 3. In both Scotland and England, there was clear loyalty to the local area, to Scotland or England as a whole, and to Britain, alongside clear loyalty to discipline and institution; note that, because the questionnaire asked about loyalty to Scotland or England separately from loyalty to 'the local area in which your higher education institution is situated', respondents in Scotland would not have interpreted 'local' to mean 'Scottish'. Academics in Scotland, again, however, showed more social attachment than academics in England, even though no less loyalty to their disciplines or their institutions. Thus 74% of academics in Scotland had some loyalty to their local area, in contrast to just 49% of academics in England. There was much stronger allegiance to Scotland as a whole (79%) than to England (56%). Respondents in Scotland were also very content that the accountability framework for Scottish higher education should be Scottish, whereas there was only minority support in England for any regional accountability (Table 4).

Table 3: Loyalty to Various Social Groups

• Unweighted sample sizes: 421 Scotland, 394 England.
• † Different versions of question in Scotland and in England.
• * Tests whole response distribution, not just rows shown in Table.
  

Table 4: Legislative Responsibility for Higher Education

• Unweighted sample sizes: 421 Scotland, 394 England.
• * The order of the categories offered in England was: UK Parliament,  Regional Assembly, English Parliament, European Parliament.

Given these views about local culture, it is not surprising that there were rather sharper national differences in views about the role of higher education in developing national culture, as Table 5 shows. Over half of academics in Scotland (55%) believed that higher education institutions should contribute to Scottish national culture, and 40% believed they should do the same for Britain as a whole. Academics in England had no more than minority commitment to English culture (28%), and not much more to British culture (35%). So the similarity of views with respect to the role of higher education in breaking down cultural barriers between nations (87% in Scotland and 81% in England) may conceal rather different views about what this might mean. The evidence would be consistent with people in England regarding the breaking down of barriers as a way of transcending national cultures. The evidence would also be consistent with people in Scotland - or at least a majority of them - taking the view that breaking down barriers is about encouraging dialogue between distinctive national cultures.

Table 5: Views about the Relationship between Higher Education and National and International Culture

• Unweighted sample sizes:  421 Scotland, 394 England.
• † Different versions of question in Scotland and in England.
• * Tests whole response distribution, not just rows shown in Table (apart from question about British culture).

In the light of all this, it is also not surprising to find widespread belief that the knowledge that is generated and transmitted in higher education institutions should be socially useful: Table 6. In both Scotland and England, clear majorities believed that higher education should contribute to local and national economic development. Conducting applied research was regarded as important, as were providing consultancy for business and placing the results of research in the market place. Respondents also believed that higher education should support the development of professional careers. Moreover, once again, Scottish responses tended to be more committed to the idea of socially useful knowledge, especially for local and national economic development and for commercialisation of research.

Table 6: Attitudes to Useful Knowledge

• Unweighted sample sizes: * 372 Scotland, 340 England; ** 421 Scotland, 394 England.
• † Different versions of question in Scotland and in England.
• *** Tests whole response distribution, not just rows shown in Table.

There was also a willingness to spread the benefits of knowledge widely, as Table 7 shows. In both countries, there was majority support for teaching mature students and students with non-traditional qualifications, for access courses, and for part-time courses. Scotland had more support than England for access courses.

Table 7: Attitudes to Widening Access

• Unweighted sample sizes:  372 Scotland, 340 England.
• * Tests whole response distribution, not just rows shown in Table.

Finally on academic values, this sympathy for higher education's social role was accompanied by persisting commitment to some of the conventional core elements of academic identity: Table 8. There were high levels of support for post-graduate and doctoral teaching, for educating an academic elite, and for conducting basic research. There was strong support for the old liberal idea of preserving humanistic culture, and for Matthew Arnold's principle of handing on the best that has been thought and said. People in Scotland may have been slightly less in favour of some of these things, but there was no majority antagonism to them. In neither country was there any basis for the claim that a mass system of higher education is culturally impoverished or is uncommitted to rigorous scholarship.

Table 8: Attitudes to Academic Values

• Unweighted sample sizes: * 372 Scotland, 340 England; ** 421 Scotland, 394 England.
• *** Tests whole response distribution, not just rows shown in Table.

The overall picture, then, is of a great deal of support among academics in both Scotland and England for a socially embedded role for higher education - for some of the central tenets of democratic intellectualism. What is more, in many respects Scotland retained this commitment more firmly than England, and this difference gives us an opportunity next to test some ideas about how such a system of values might be reproduced. So it is useful to summarise the ways in which Scotland was distinctive. Academics in Scotland were more in favour of higher education's local and national economic role. They were more in favour of students' being educated to take on leading civic roles, whether nationally or locally, and more inclined to educate students to be critical of people in authority. They were more in favour of providing access courses. They were more in favour of higher education's role in maintaining national culture, whether that is conceived of as Scottish, English or British, and alongside this they were more in favour of commercialising research. They were also slightly less in favour of higher education's undertaking basic research. And academics in Scotland were more in favour of government's monitoring of higher education, and more attached to their institution's locality and nation.

Institutions and Networks

These comparisons of Scotland and England, where there are differences, enable us to investigate some potential explanations of the prevalence of views about the social purposes of higher education. We investigate the relevance of demographic characteristics (gender, age and social class of origin), academic discipline, academic grade, history of institution, the academic and public networks in which academics engage, and the academic origins of respondents.

The proportions of the sample who were women were 30% in Scotland and 32% in England. Of the attitudes which differed between Scotland and England, there were in fact few differences by gender, and so the national differences were not explained by gender differences. Where there was a gender difference, it was much the same in each country. Compared to men, women were more in favour of offering access courses (by 86% to 74% in Scotland, and by 69% to 58% in England), and felt more local loyalty (by 80% to 70% in Scotland, and by 52% to 48% in England), but in each case the national differences were greater than the gender ones.

Age was grouped into 35 and under (18% of Scottish sample and 23% of English sample), 36-45 (34% and 31%), 46-55 (31% and 26%), and 56 or over (16% and 20%). Again, none of the national differences in attitude were explained by age differences. There was only one substantial age-related difference. People aged under 35 were more in favour than average of preparing students to take on leading positions in government and business in Scotland or the region of England (91% in Scotland and 76% in England, compared to the averages of 88% and 58% in Table 1), but the largest age difference (18%, among young people in England) was not significantly greater than the smallest national difference (15%, among young people).

Respondents were asked to say which social class their father had belonged to when they were aged 14: professional or managerial (61% in Scotland, 71% in England), intermediate non-manual (12% in Scotland, 9% in England), or manual (27% in Scotland, 20% in England). Yet again, none of the national differences was explained by this. There were only two large class-related differences. One concerned educating academic high-fliers, which people of professional background were more likely than others to regard as 'very important' (66% compared to 51% in Scotland, and 76% compared to 70% in England). The other was government monitoring of higher education expenditure, less favoured by people of professional background than by others (69% compared to 84% in Scotland, and 64% compared to 67% in England).

Thus none of the differences in attitude between Scotland and England may be explained by demography, and indeed the demographic characteristics of academics in the two countries were very similar.

Discipline is defined here in five broad categories: medicine and related subjects, science, engineering, social science (including business studies and education), and arts and humanities. Some attitudes where Scotland and England differed did also vary among these, but none of the differences between the countries is explained by the pattern of disciplines to which academics in Scotland and England belong, mainly because these patterns were similar in the two countries. In any case, the differences in attitudes among the disciplines were not great. Where there was majority average support for a civic role for higher education, in nearly all respects all disciplines still gave at least 40% assent. There were only three exceptions. Scientists and engineers gave very low support for preparing students to be critical of authority - in both countries, only about 30% of scientists and about 17% of engineers agreeing that this is very important, compared to overall figures of 33% in England and 46% in Scotland, as shown in Table 1. And social scientists in England showed only 37% loyalty to England, as against the average of 56% in Table 3.

The second potential academic explanation of differences between Scotland and England is academic grade, defined as four categories: contract researcher, lecturer, senior lecturer (or equivalent), and professor. Again, none of these differences explained the differences between Scotland and England, because the distribution of grades was similar in the two countries. And, also as with discipline, the grade differences were in fact small. Where there was majority average support for a civic role for higher education, in all respects all grades still gave at least 40% assent.

Rather more revealing results are obtained when we examine the effects of type of higher education institution. These were classified into five categories: (i) ancient Scottish, (ii) Oxford and Cambridge, (iii) English 'redbrick' universities (founded between the early nineteenth century and 1962), (iv) 'Robbins' universities (founded in the 1960s and 1970s), and (v) the post-1992 universities or other higher education institutions. Mostly these did not explain the national differences either. The pattern shown by the logistic regressions was, mostly, that the ancient Scottish universities were more in favour of a social purpose than Oxford or Cambridge, just as their nineteenth-century history would lead us to expect. The English redbrick universities had inherited something of this too, again following their history as civic institutions. The Robbins and post-1992 institutions in Scotland seem to have inherited the history of civic engagement established by the ancient Scottish universities. And in England, compared to their counterparts in Scotland, these institutions generally showed less of a civic commitment.

Four examples of these general patterns are shown in Table 9, concerning local economic development, preparing students to contribute to the community, offering access courses, and whether government should monitor the expenditure of higher education institutions. In all cases, the civic view is stronger, usually much stronger, in the ancient Scottish universities than in Oxford, Cambridge or the English redbrick universities. In England, the Robbins and post-1992 institutions show higher levels of civic commitment than Oxford and Cambridge, whereas there is not so much of a difference between these categories and the ancient universities in Scotland. The result is striking: with one exception, every category of English institution has lower levels of civic commitment than all the categories of Scottish institution, not just than the corresponding categories. The exception is attitudes to access courses in post-1992 institutions, but the difference there could easily be due to sampling error since the samples sizes are fairly small (the t-value for the difference is about 1.47, which has a p-value of about 0.08).

Table 9: Attitudes to Social Context of Higher Education, by Institutional Type

Unweighted sample sizes:

 * ancient Scottish 174; Oxford, Cambridge 47; English redbrick 166; Robbins 123 (Scotland), 45 (England); post-92 72 (Scotland), 75 (England).

** ancient Scottish 195; Oxford, Cambridge 53; English redbrick 193; Robbins 142 (Scotland), 48 (England); post-92 79 (Scotland), 92 (England).

Among all the ways in which Scottish and English views differed on average, there were only two exceptions to this uniformity of national difference across types of institution. One was the support for students' being critical of people in authority, where the English redbrick universities showed very low support, and thus explained most of the English distinctiveness. The other was the question of maintaining a common British culture, where the ancient Scottish universities showed exceptionally high support, thus explaining most of the Scottish distinctiveness. In all other respects, what seems to have been operating is a national culture, associated originally, perhaps, with particular institutions, but not being lost by these while being inherited by new ones during each wave of expansion.

We have two further tests that might suggest how the relatively distinctive academic values in Scotland might have been sustained. One is in terms of the kinds of academic and public networks which staff in the two countries are involved in. The patterns of these networks are shown in Tables 10 and 11. Table 10 is about academic networks. Scottish and English academics are involved in networks in Scotland or the region of England to similar extents, as they are in networks outside the UK. They also are involved similarly in editorial activities. But the Scots are rather less involved in networks within the UK but outside Scotland than English academics are in networks within the UK but outside their own region. This tends to suggest there are Scottish and English national networks which might serve to socialise academics into a common national academic culture. Something the same can be inferred from a question about the networks that people found most important to them in doing their research (not in the table). In Scotland, 22% named networks in their own institution or across Scotland; in England, only 10% gave their own institution or regional networks. UK-wide networks were more frequently named in England than in Scotland: 21% against 17%. European networks were equally common (14% and 13%), but global networks were more frequent in England than in Scotland (54% against 44%).

Table 10: Activities in Academic Networks

• Unweighed sample sizes:  372 Scotland, 340 England.
• † Different versions of question in Scotland and in England.

Scottish academics were not more likely to have engaged directly with government than academics in England in the last three years. For example, around one in ten had given advice to (respectively) Scottish or English government departments, and a similar proportion in each country had done so for UK departments, or had advised an international agency (such as the UN), or had sat on a quango board. As Table 11 shows, moreover, much the same can be said about engaging with non-governmental organisations or with the private sector: levels were similar in the two countries. However, Scottish academics were rather more inclined to take on the role of the public intellectual than academics in England: 57% as against 49% for speaking to a non-academic audience, and 33% against 25% for speaking on the broadcast media. Writing for newspapers was at a similar level in the two countries.

 Table 11: Public Engagement, other than Government

• Unweighted sample sizes:  372 Scotland, 340 England

So academics in Scotland have some academic and public networks that are peculiar to them. In fact, however, it turned out that the networks on which the countries differed had little relationship to the civic attitudes on which they differed as well. So, in the logistic regressions, most of these differences in attitude could not be explained by differences in networks - that is, the two such measures in Table 10, the two in Table 11, and the use of Scottish, English regional or institutional research networks. So it appears that the distinctive aspects of Scottish academic beliefs are probably not mainly sustained by distinctive kinds of network, except by the network implied by working in a particular national culture.

The other test of the origins of civic attitudes is to look at a potentially particularly interesting group: people who received their own initial education in England but who are working at Scottish higher education institutions, and to compare their views with the views of people who were educated in England and are working in England. Twenty-nine percent of the Scottish sample received their first degree from an institution in England (compared to 54% from Scotland). That yielded 139 respondents in Scotland who received their first degree in England. In England, 73% received their first degree from an institution in England, giving 294 respondents. (Unfortunately, there were too few Scottish-educated people in English institutions - only 5%, or 20 respondents - to allow the reverse comparison to be made reliably.)

All the attitudes which differed between Scotland and England were then tested controlling for place of initial degree. Without exception, the national differences remained in the logistic regressions. That is, the views of people of English origin in Scotland more resembled the views of people of Scottish origin in Scotland than they resembled the views of people of English origin in England. The typical pattern is shown in Table 12 for the same four variables as were used to illustrate the differences of institutional types in Table 9. In each case, people of English origin in Scotland were closer to the views of people of Scottish origin in Scotland than they were to people of English origin in England; indeed, in three of these four examples, the English in Scotland were much closer to the Scots in Scotland than to the English in England. Carrying out the same analysis with country of secondary schooling in place of country of initial degree led to the same conclusions.

Table 12: Civic Views, by Place of Initial Degree

Unweighted sample sizes:

* English degree in England 251; English degree in Scotland 125; Scottish degree in Scotland 183.

** English degree in England 294; English degree in Scotland 139; Scottish degree in Scotland 213.

• Meaning of columns: eg ' English degree, working in Scotland' means people who received their first degree in England and who are working in a higher education institution in Scotland.

It appears, therefore, that incomers to Scottish higher education institutions absorb the culture simply from the country in which they are located. This proposition can be further tested by examining whether the people of English origin in Scotland take part in the same kinds of networks as other academics in Scotland - the networks that were discussed above, and which turned out not to be able to explain the distinctive civic attitudes in Scotland. The results for the five kinds of network that were distinctive in Scotland are shown in Table 13. For giving lectures, attending conferences and finding particular kinds of networks important in research, the English in Scotland much more closely resembled the English in England than they did the Scots in Scotland. They were equidistant in relation to speaking to a public audience, and only for broadcasting did they more resemble the Scots in Scotland than they did the English in England. Thus the patterns of network cut across the patterns of values.

Table 13: Networks, by Place of Initial Degree

• Meaning of columns: eg ' English degree, working in Scotland' means people who received their first degree in England and who are working in a higher education institution in Scotland.

The evidence about networks is consistent with how the various types of respondent described their national identies, as shown in Table 14. The questionnaire invited them to choose one or more identities from the list British, English, European, Irish, Northern Irish, Scottish, Welsh, None of These. The proportions choosing British, English, European and Scottish are shown in the table. (No other specific identity was chosen by more than 6% in any country; 14% in each country chose 'none of these'.) The identity choices are classified by where respondents received most of their secondary schooling and their initial degree, but in fact the patterns are very similar for schooling and degree. The pattern that most closely reflects the results on networks is in relation to Britishness: people who were educated in England, and are working in Scotland, retain the same high level of British identity as those who were similarly educated and are working in England, a level of Britishness that is almost twice that of people who were educated in Scotland and are working in Scotland. The Scottish-educated are much more committed to a Scottish identity than to any other. Moreover, the migrants to Scotland are substantially less likely to claim an English identity than their similarly educated counterparts in England. That the proclamation of a British identity by the migrants is not just a desire to assert a less particularistic allegiance than the Scottish-educated is shown by the pattern for European identity, which commands around three-in-ten support in all three groups.

Table 14: National Identities, by Place of Initial Education

• * Respondents could choose more than one identity: see text.
• Meaning of columns: eg 'English school, working in Scotland' means people who received most of their secondary schooling  in England and who are working in a higher education institution in Scotland.

So academics coming from England to work in Scottish universities operate in much the same kinds of networks as the English in England, and yet they have an outlook on the civic role of higher education that marks them out from their compatriots who remain in England. Their activities look English, but their views are Scottish. They may retain a similar attachment to a nominal Britishness, but what they mean by 'British' is likely to have been coloured by their acquired national context.

Conclusion

We should not exaggerate the extent to which views in Scotland in this survey were distinctive. The first conclusion, in fact, is the widespread prevalence, throughout England and Scotland, of the core elements of the democratic intellect: that universities should serve society, both in their teaching and in their research, that knowledge is potentially socially useful, including of use to the economy, that governments have the right to hold universities to account, and that it is proper for academics to feel loyalty to a society as well as to their discipline or institution. These views are held alongside a persistently high level of attachment to the core elements of traditional academic values, such as advancing basic research and maintaining liberal culture. Donnish dominion may have declined, but the academics who now inhabit the universities where it might once have held sway have a coherent set of beliefs about the kind of higher education which they should now be providing.

Nevertheless, Scottish views were distinctively more civic than views in England, and so we can reasonably talk of the maintenance of a tradition of democratic intellectualism, not just the creation of new attitudes to suit an era of mass higher education. The tradition pervades all types of institution, and is shared across all grades and, in most respects, across all disciplines. It is held by academics of both genders, of all ages, and regardless of the social class in which they grew up. It is, moreover, a national tradition that seems to have been maintained by the very existence of a national system of higher education. The strongest evidence for that came from the views held by academics in Scotland who received their initial education in England, whose views were much closer to typically Scottish views than to typically English ones. Whether these incomers acquired their views once they had arrived, or were attracted to Scotland because they perceived a set of national academic values there that would be congenial to views they already held, cannot be settled from a cross-sectional survey; but, either way, the academic values of these incomers serve to confirm the validity of the tradition. Further evidence for the national character of this tradition is in the clear tendency for academics in Scotland to view Scotland in the same way as they, or as academics in England, view Britain, and in ways that are quite different to the views in England about English regions. We saw this in relation to preparing students for social leadership (Table 1), to economic development (Table 6), and to the appropriate legislative framework (Table 4). Indeed, there was in fact rather more loyalty to Scottish culture, and to maintaining it, than there was to British culture (Tables 3 and 5). These similarities in attitude to Scotland in Scotland and to Britain in England strongly suggest that Scotland is regarded as a source of academic values that is of a similar kind to Britain - a national cultural source, not just a regional one.

The significance of Scottish and British culture differs, however. Studies of educational and other attitudes in Scotland and England have found that Britishness in Scotland is associated with right-wing views, whereas Britishness in England is associat