Suggestions for improvement welcome

(A concept of elitism not understood by labour government ministers
in Tony Blair's government.)

Aaron Sloman
School of Computer Science, University of Birmingham.


Installed: 20 Dec 2004

Updated: 18 Aug 2009; 3 Aug 2014 (Minor revision -- reformatting)


This paper is
A PDF version may be added later.

A partial index of discussion notes is in


The thoughts expressed here were originally triggered by comments in the email discussion list of the CPHC (The Council of Professors and Heads of Computing). This was originally written on 20th December 2004. Some afterthoughts can be found at the end.

We should all be elitists in the first of two senses explained below.

Words like 'elite', 'elitist' and more recently 'snobbish' have been used from time to time in email messages, especially in connection with discussions regarding UKCRC (the UK Computing Research Committee 2009).

It seems to be assumed by many that being elitist is inherently bad (even 'snobbish').

I shall argue, on the contrary, that there is a kind of elitism that is valuable and should be vigorously promoted by ALL academics, namely

Failure of some politicians, including government ministers, to understand this seems to be based on muddled thinking about elitism, and when this infects academics also it can do damage to the quality of what we do and what we achieve, including the quality of discussions about funding and management, where name-calling (e.g. 'elitist') should not be a substitute for argument and analysis.

Compare the distorted semantics of the word 'liberal' in the USA. According to a news report this morning (20 Dec 2004) the new UK home secretary appears to have been infected by this tendency to use 'liberal' (admittedly preceded by 'woolly') as a term of abuse, when responding to criticisms of his plans for identity cards, instead of addressing the arguments. I don't know if that's just a one-off comment or a sign of a dangerous additional mode of subservience to the USA in the UK government.

As far as 'elitism' is concerned, muddled thinking can arise out of a failure to distinguish two common uses of the word 'elite' as an adjective (related historically to the adjectival uses of 'choice', and 'elect'), namely 'elite' meaning:

  1. best at some task or activity (e.g. art, music, athletics, mathematics, surgery, innovation, business, teaching, management, engineering design, nursing...)


  2. belonging to some social group or sub-group (e.g. the aristocracy, the rich, the powerful, the celebrities, etc.)

The noun 'elite' is often used in sense (a) (e.g. talking about 'funding for the elite' in the context of athletics), but the derived words 'elitist' and 'elitism' often tend to be linked with (b), usually with implied criticism or disapproval.

What has happened recently is that politicians and others have come to confuse caring about the elite in sense (a) with being elitist in a sense related to (b).

Examples are UK government ministers (including the UK Home Secretary in 2004, when he was previously in charge of education) who justify high A-level grades on the grounds that the students have worked very hard, and who descend to name-calling (using words like 'elitist') instead of producing arguments against people who point to evidence of lowered standards because grades are awarded on the basis of inappropriate criteria.

[We have all seen results of this in the intake to university degree courses, and we also now see a similar process in the grades we are being forced by various pressures to award to our own graduates. The last time I made that point some months ago on the CPHC discussion list it triggered a version of the 'elitist' response.]

We should all be unashamedly elitist in a sense related to (a), namely

[Digression -- quality judgements:
How to decide what high quality is without depending on personal whim, or prejudice, can be discussed elsewhere, for instance in this document on criteria for evaluating various kinds of computing research:

I recently added a note on using scenario-based criteria for evaluating difficult long term research programmes, such as some of the UK Computing Grand Challenges .

Judgements of quality need not be just personal preferences but can be related to agreed short-term and long-term research goals, though that does not always happen. As Stevan Harnad has pointed out, the growing use of citation-based judgements of quality presupposes the prior use by referees and citing authors of a more basic kind of judgement. Moreover, citations can be subject to fashion and we must be prepared to judge quality independently of fashion.

Note (added 3 Aug 2014)
In many practical contexts judgements of quality are related to what are labelled "Meta-requirements" in

Aaron Sloman and David Vernon, 2007,
A First Draft Analysis of some Meta-Requirements for Cognitive Systems in Robots,
Contribution to euCognition wiki,
(In retrospect I think 'Meta-functional requirements' might have been a better label.)

Why we need to be elitist (type-a)

There are contexts in which we would probably all be unashamedly elitist in the sense described above. For instance I expect we all agree that when people, including government ministers, need brain surgery, they should be operated on by surgeons selected on the basis of depth and breadth of knowledge and skill in judgement and performance rather than on the basis of good intentions, strong desire to succeed, hard work or the need to be fair to people who want to be surgeons.

Likewise designers of airliners, of software security systems, of bridges, etc.

So certain jobs should go to the elite (in sense (a)), for the good of all.

There is obviously nothing wrong with arguing that people on whose competence others will depend, sometimes for their life, should be assessed according to very high standards, and that when studying they should also be given teachers and learning environments of very high quality. Scarce resources and opportunities should go to the elite(a), namely those who are best qualified and best able to make use of them (especially where that produces benefits for all). But safeguards are needed regarding how the elite in that sense are selected as beneficiaries of public finance.

Sometimes we don't need to be elitist (type-a)

Where resources are not scarce or not very costly there is an argument for distributing them to all who want them. (This used to be one of the functions UK universities fulfilled by running 'extramural' education departments, which gave free or low-cost evening classes to anyone who wanted them. Many academics used to contribute some of their spare time for this, but now they cannot afford to.)

When resources that are widely thought to be valuable and a general right (e.g. heath services, or educational opportunities or opportunities to do enjoyable and interesting low-cost research) are scarce as a result of bad government policies, there is an argument for urging changes in those policies, rather than restricting the recipients. But there will always be difficult cases where the resources required are scarce or expensive.

How bad funding policies lead to distorted research fund allocation policies

It does not seem to be widely appreciated that in the UK the pressures on decisions regarding resource allocations for research are seriously distorted by inadequate university funding. That is because inadequate funding produces pressure on university heads, and heads of departments, to try to get as much external income as possible which causes many universities to require all academics to be grant-getters in order to keep their jobs. These forces also distort selection for academic jobs in favour of people who are more likely to get grants, even though many academics are perfectly capable of doing excellent teaching and research without external grants, and would have done so in the UK university system 20 years ago or more. (I started lecturing in 1962, and did not get a grant till about 1975, but nobody was threatening me with loss of tenure, etc.)

The distortion is extremely unfair because while there are some people who need equipment or collaborators to do their research, others, e.g. some of the theoreticians, and some younger lecturers who need time to continue broadening their education by attending seminars, reading, and meeting people from other departments, may actually be hindered in their research and development by the need to spend time on proposal-writing, recruitment, management and other tasks related to project funding. Such people may need time to get on with thinking and learning more than they need staff, PhD students, equipment, or even travel funds.

Of course, the distortion is worse in departments where most of the good research does not intrinsically require external funding, and where funding sources are very scarce, e.g. in philosophy and many humanities subjects.

The really nasty feature of this situation is that if everyone is required to get grants in order to be allowed to keep their jobs or get promoted, that makes it seem unfair not to distribute grants evenly. So that causes 'moral' or 'egalitarian' pressure to allocate grants on some other criteria than likely effect on quality of research. (EPSRC partly acknowledges this by having 'easier' access to research funds for people in their first teaching job.)

The answer is NOT to press for grants to be distributed evenly (anti-elitism), which would be a waste of public funds, but to remove the universal requirement for all university staff (especially young staff) to bring in grants.

But that requires a different funding model for universities. Making this work may require reducing the number of universities where most people are expected to do research, and where teaching is research-led, and increasing the number of polytechnic institutions where able and devoted teachers get on with their important jobs free from unwarranted pressures to publish or get grants. (That of course, is a distinction we used to have, but lost a decade and a half ago.)

I wonder how many university heads of department understand this point, and present the argument to senior managers, promotion committees, politicians, etc., instead of harassing young colleagues with threats of loss of tenure if they don't get grants, or even sacking them, as has happened in some UK universities (and has been commonplace for longer in American Universities.)

Part of the cause of the problem is something whose mention tends to trigger accusations of 'elitism', namely the disastrous abolition of the important and socially and economically useful distinctions of function between universities (where most teaching was research-based) and polytechnics (where most teaching was not research-based). That abolition (Thatcher's revenge on universities for not accepting her earlier proposal to distinguish R and T universities) artificially increased the numbers of people who were expected to get research grants, whereas in the previous situation they were doing perfectly good, and useful jobs that did not involve research -- and had higher teaching and admin loads accordingly. (Even before the abolition of the divide, some polytechnic staff were excellent researchers, and were able to apply for and get grants in competition with university staff, but the others were not required to.)

It seems that some of the 'new' universities have faced up to this point and started referring to themselves as 'polytechnic' universities, which will help to make clear to students and others what their goals are and which kinds of educational opportunities they offer.

Of course, where such a distinction of function exists, mechanisms are needed to ensure that people, especially students, are not trapped in the wrong part of the system, e.g. 'slipways' should enable students easily to move between institutions if their performance indicates that they would do better in a different one. (As argued here: http://www.cs.bham.ac.uk/~axs/gov/.)

Alternatives to basing resource allocation on quality of achievement

I expect nobody would argue that scarce resources and opportunities should be allocated to some elite of type (b) e.g. a particular subset of society defined by features that have nothing to do with the purposes of the resources, e.g. status, gender, race, colour, family history, connections with the rich and powerful, etc.

Yet when politicians (and others) cannot distinguish that sort of elitism from the sort recommended here, it can get in the way of sensible policy making.

There is, of course, a case for some positive discrimination where there is evidence that the 'irrelevant' features have temporarily impeded development of real potential to satisfy the requirements of capability and performance. People who have been held back by ethnic, gender, or regional causes of inequality should be given a chance to progress without the impediments.

However, simply putting more people into a situation where they cannot cope with what is required is no solution: it just causes general reduction of quality, as has already happened in UK universities because of a combination of drastically worsened teacher/student ratios and also because of reduced entry requirements necessarily adopted to fill places in order to maintain funding levels. That in turn produced pressure to lower standards in university examinations in order not to increase failure rates. An upper second class degree used to be an indication of a level of achievement that could justify admittance to do research on a PhD. Nowadays it would be foolish to treat PhD applicants as if standards had been maintained.

Instead of simply widening access to university places or to research opportunities, we need carefully designed mechanisms that do not produce lower quality all round, as has happened to teaching in UK universities over the last 20 years. (See http://www.cs.bham.ac.uk/~axs/gov/)

There is also a case for giving people a chance to establish what they are capable of doing where they already have qualifications and capabilities of a high order, but have not yet had time or opportunity to demonstrate that. EPSRC recognises this and has mechanisms with that purpose. This is all elitist in the good sense.

I am aware that this is a topic on which vast amounts have been written (e.g. in philosophy and social and political theory) that I have not examined.
(See Penny Egan's article in the Education Guardian, 28 March 2008, for example.
I don't have time to attempt a proper literature survey. But I am willing to accept criticisms and suggestions for amendments from people who have studied more of the relevant literature than I have.

NOTE added: 28 Dec 2004
I have just discovered this online article on elitism by "Ophelia Benson",
also critical of ill-informed varieties of elitism:
(Broken link fixed 4 Aug 2014)

Maintained by: Aaron Sloman