PUSH SHOVE AND THE GIFT OF THE GAB: The hidden techniques of equipment procurement

by Derek Law

The title of this paper comes from a remark made by a Somerset Maugham character. Asked what are the main attributes required of a successful politician he replies "Push, shove and a gift of the gab". That emphasis lies at the heart of most of what I want to say. Prising large chunks of money from the institution which employs you is a skill, but a skill embedded in the political process. And so it is with politics that I shall mainly be concerned. Institutional politics is probably the same wherever one goes and in whatever environment, which is why works such as'The Peter Principle' or 'Parkinson's Law’. 'Up the Organisation' strike common chords with us all and why Sir Humphrey from 'Yes Minister,' is seen as such a universal character. My context is the university system. Although I will argue that it displays particular features which make it unique, I hope that those of you from different environments will find areas of common experience. A university is, after all, just another form of bureaucracy. The important point is not the nature of that particular bureaucracy but the need which we all face to find the hidden pathways to the real seats of power. To paraphrase Karl Marx it is not enough to want to charge the world, you must first understand it.

The librarian as politician

My starting point is that politics must be seen, as enjoyable and not a distasteful chore; that the creation of alliances and coalitions, the forming of pacts and secret treaties, the use of a power-base-however small, the discovery of fools and knaves (and there are plenty of both), add an enjoyable spice to the otherwise dull round of photocopier maintenance contracts and leaking roofs. I follow James Thurber in believing that you might as well fall flat on your face as lean over backwards.

Some librarians espouse the view that their role is neutral and apolitical and that we should stand above the hurly-burly of the marketplace, simply offering the services decreed by our political masters. This seems to be at best wilful and at worst to misunderstand the nature of the political process within institutions. The trouble with the role of innocent bystander is that you're liable to get shot, whenever someone is trying a raid on the available money. As Clare Booth Luce put it, no good deed will go unpunished. Increasingly, the role of the librarian is that of a hustler raising funds not that of some impoverished bourgeois, scraping by in surroundings of genteel decay.

Of course, occasionally one finds the truth of Clausewitz's dictum, that war is the continuation of politics by other means. Fortunately, this is a rare occurrence, for open warfare, even verbal warfare, in Senate or Local Authority or Boardroom, is in effect an admission that politics has failed, and therefore that you have failed, and that can be difficult to accept. Politics is about compromise and settling for half a loaf. Just about the only remark I’ve ever heard Arthur Scargill make with which I agree is his comment that he's not an idealist -that implies you're not going to achieve anything. In my view there are rarely absolute principles that are worth going to the stake for. Library development is a constant shuffling progress towards Jerusalem, with many winding byways and wrong turnings and the intricate collegial community of the university or the labyrinths of most large bureaucracies are more suited to the patience of shuttle diplomacy than the grand uncompromising posture, which looks good but rarely gains anything. With God all things are possible said St Matthew -but then he didn't have to persuade a Faculty of Theology that they needed to upgrade the wiring of their buildings from 5 amp to 13 amp, in order to instal an automated library system, for which they didn't see the need.

Influencing the decision makers

Now if you remember only one thing from this talk it should be the title of this little book, the Microcosmographia Academica, first published in 1908. It is one of the great unknown treasures of civilisation. I would like to summarise one or two of its themes very briefly, because even after eighty years they are fundamental to an understanding of the politico-academic process. F. M. Cornford wrote this book and subtitled it 'A Guide for the Young Academic Politician'. But it has much wider applicability and is no more about academic life than Parkinson's Law is about the Admiralty. It should be mandatory reading for anyone who wishes to understand how organisations operate. In it, amongst other gems, he sets down for the first time the four arguments for resisting change. Since the only good reason for being in politics is to effect change and since all change is a threat to one interest group or another, it is important to understand how attempts are made to block change, so that you, in turn, are in a position to counter or side-step the arguments. You will. doubt have heard all four arguments used, perhaps without realising where they came from:

1. The Principle of the Wedge.

We should not act row for fear of raising expectations, which we have not then the means to satisfy. Those who wish to block change, but retain a reputation for radicalism can combine this principle with the .ore sophisticated argument that the present proposal would block the way for a far more sweeping reform at a later date.

2. The Principle of the Dangerous Precedent.

We should not act now for fear of binding our successors to a course of action, which will not meet their situation. Every public action which is not customary, either is wrong, or, if it is right, is a dangerous precedent. It follows that nothing should ever be done for the first time.

3. The Present System must be given a Fair Trial.

Someone is always tinkering with systems and it therefore follows that it can almost always be argued that more time is needed to assess the effect of the most recent set of changes.

4. The Principle of Unripe Time.

It is agreed that change is necessary, but in some mysterious way it is not quite the moment to introduce it. Usually because the public, the recipients of the change, are in some fathomless way unprepared.

So with all these arguments against change, how is business actually effected? It is effected by those who have influence. And how does one acquire influence? Cornford assured us that it is done in the same way as one acquires the gout and, indeed, that the two aims should be pursued simultaneously, thus earning one the reputation of being a good fellow and good fellows are "sound". Alternatively one can become a good business man, with contacts in and experience of the world outside. And how do those with influence conduct business? In Cornford's terms, business is done by those who walk the King's Parade in Cambridge, and use their network of contacts to give and receive favours and to make deals. You must understand that this is one of those irregular verbs so popular in 'Yes Minister'. It goes:

I make public-spirited proposals,

You promote narrow sectional interests,

He seeks promotion.

It is, of course, important that this squaring of individuals should appear to concern two quite unrelated needs, so that each party an appear to be offering the other disinterested support, to further the goals of the organisation What is required above all in this political arena is sincerity; once you can fake that, you have it made. It's known as 'making the world safe for hypocrisy'.

Every university has its King's Parade. For many years in my institution it was the Male Staff toilets, but it can equally well be a Senior Common Room or a local pub. Now, whether or not one approves of the system, it exists and it can be used. Due to the nature of the university as an organisation, it is unlikely to disappear or be destroyed, even by the pressures now facing us after the Jarrett report on the management of universities, which requires us to be more business-like -and that bears some examination. What marks universities out from business corporations is that they are not hierarchical pyramidal structures pursuing a single goal, but are really a loose federation or coalition of mutually supporting interests, which choose to come together while it benefits those interests. It is, of course, more complicated than that sentence depicts, but marks universities as clearly different from commercial structures. Now, that has political implications and so I'd like to spend a little time looking at the organisation of universities, and the pressures on them. For much of this I rely on the excellent recent review of the universities by Lockwood and Davies.(11) Although the world I know is that of the universities, I expect that many of the situations and structures which I shall describe will be familiar, at least in other publicly funded environments.

Understanding the organisation

Universities as a group form one of the longest standing set of institutions in the civilised world. They have a clear set of international links and an equally clear set of public expectations. Despite the pressures created by the Jarrett report, which recommended far-reaching reforms of university management and financial structures, universities are still highly collegial in nature. The effect of this is that any administrative act is only legitimate when it is validated by the agreement of the majority. There is therefore a need to reconcile different points of view and divergent interests, by the sort of negotiation which makes compromise or horse-trading possible. The academic role is still perceived as the dominant one and this is emphasised through the committee structure, which lies at the very heart of university life. Even here the perception of its importance is more apparent than real, the dominant power brokers are individual academics rather than officers, and much of the management of the institution is run by those academics as a spare time occupation mixed with their academic teaching and research.

As in many organisations, the committee structure in universities provides a forum for competing interests, many of which will cancel out. Committees are generally constructed with some notion of balance between interest groups and representation of all parties, which emphasises the bargaining process. Naturally, as with most committees, much of the real bargaining goes on outside the formal meetings. The university committee system is one in which recommendations are referred from one body to another for validation and in which the long vacations create a built-in delaying mechanism and cooling off period, which means in turn that inactivity prevails, participation in decision making is fluid and intermittent, while conflict and competition, are common. Worse, committees may share an overlapping membership, but it is quite normal to see one committee proposing a course of action which conflicts with the proposals of another committee and with both of them wholly ignoring the resource implications. Time in committees is at a premium- a Senate may have as little as twenty hours a year. As with most committees, much of the real bargaining goes on outside yet much of it goes on the formal meetings routine business. Even in universities which have been properly Jarretised and have a small Planning and Resources Committee, the PRC will still usually display the notion of a membership which balances the competing interests, while paying lip-service to the notion that the committee members do not represent those interests. It is common for the Librarian to be excluded from membership of this new source of power and influence which emphasises the need for the Librarian to be even more active in lobbying to achieve his goals.

The difference between power and authority within the institution is further clouded by the fact that different parts of the organisation may have different missions. There is, or can be, enormously high internal autonomy, for three reasons. Firstly, the department or unit has expertise or technical competences which cannot be challenged (although the recent exercise which gave comparative ratings to departments came close to this - hence the fuss); secondly, the department sees the peer group to which it affiliates as external, not internal, which is why academics and librarians have their conferences as interest groups not as organisations. A conference of Germanists, for example, seems sensible, while a University of London Management Team weekend seems somehow risible; thirdly, they have significant opportunities to raise resources externally and independently, which can in turn affect pecking order within the institution and accountability to it. At the extreme, successful research units can decamp wholesale to another institution, although, paradoxically, they have no place in the formal power structure.

In fact, it is often difficult to determine exactly where the boundary of the institution lies. There is so much externally organised funding and staffing that the edges blur, making resource management even more difficult. So, we have a picture of organised anarchy, limited manageability, ambiguity, complexity and diffusion, compounded by part-time senior management. It's certainly not a body where once a demand is recognised it is given a priority, resource allocated and then managed until the task is fulfilled. Indeed, it is a curious paradox of government policy that its desire for the universities to increase the proportion of non-government funding actually makes them less manageable at the same time as government policy requires us to improve management.

All of this provides a rich adventure playground for the aspiring politician. The micropolitical map bears no relation to the official structure, and although officers can rarely take decisions, they can certainly influence them. The mere fact that they may be the only stable element in a committee over periods of time, and in the case of the Librarian or Computer Centre Director, that they may command significant recurrent resource, gives the possibility of real influence. This is caricatured by saying that the decision making process relies on experience, expediency and the hydrostatic pressure of one's bladder. It is a constant source of astonishment to me how ignorant most members of the institution are about how the system works and how decisions are taken. Most of them don't know where the King's Parade is, far less how to walk on it.

The politics of automation

I'd like now to turn to the specifics of the politics involved in library automation and my theme now moves to the works of that consummate politician Niccolo Machiavelli, who commented in 'The Prince':

It should be borne in mind that there is nothing more difficult to carry

out, more doubtful of success than to initiate a new order of things.

As you can tell from a name beginning with Mac, he must have been a Scotsman. Looking at the most recent crop of Scottish politicians such as John Smith and Robin Cook, The Guardian suggested recently that they were much more devious. They were not Machiavellian -you always knew where you were with Machiavelli. He would have had a lot of sympathy for Mae West. When a reporter saw her diamonds and said "My goodness, how lovely", she replied "Goodness had nothing to do with it". The other old model to which all Scots politicians should aspire is T.S. Eliot's Macavity:

Macavity, Macavity, there's no one like Macavity

There never was a cat of such deceitfulness and suavity

He always has an alibi and one or two to spare:

At whatever time the deed took place

MACAVITY WASN'T THERE

In looking at the power structures to be addressed in attempting to create a new order, of things I want to divide these into three groups. Firstly, external; this can involve national funding agencies such as the UGC, validating agencies such as the RIBA or CNAA, other parts of a federal institution - although the University of London may be unique in this- or perhaps different institutions within the same funding authority. Secondly, institutional politics; the competition for resources, relationships with the Computer Centre and Department of Computing, and very importantly, relationships with the administration. Thirdly, internal politics within the Library; who will be first, who will be last (and there will be competition for both positions), and, if a retrospective conversion project is associated with the implementation, there will also be fears of deskilling and of new power structures.

The external agencies require little discussion, since they are the least amenable to manipulation. You push your effort in at one end and something-or nothing-comes out at the other end, but little can be done to affect the process. For example, if a group of RIBA or CNAA assessors is visiting an institution, there may be an opportunity to comment -either directly or through an intermediary -that library automation is desirable, but there is no control over whether that comment will appear in the visitation report. Similarly, the availability of UGC central funding for networking forces institutions to bid for it, since they cannot be seen to be giving up in any race. The size of the bid is important to the institution politically, and it may have to give the Library public guarantees about automation plans in order to make the bid appropriate to the size of the institution. Again, one can only do the groundwork, rather than influence the result, although if you look at how the UGC recently allocated money to libraries for communications you will see who are the smart operators. Finally, certainly in London, there is vicious competition for resources between the various Schools of the University. The Heads of Schools are prepared for battle, given objectives and shoved into combat. The real difficulty here is not the ability of the Heads of Schools to negotiate, but the fact that concessions in one quarter such as the Library, may gain the School larger concessions on other battlefields and again one has . control over this. In summary then, external politics is a factor which cannot be ignored, has to be managed, but one over which the Librarian can exercise only marginal control.

The second area is institutional. I have had a variety of different experiences here, but there is one common thread. It is that every self-respecting Department of Computing knows that creating an integrated library automation system is at the level of a postgraduate summer project, so why shouldn't it be one? If you have a powerful Professor of Computing, whose technical view, remember, cannot credibly be challenged, and an apolitical or, non-technical Librarian, there is a recipe for disaster. Although we all know the theoretical response, that the issue is not creating a system but maintaining it, I can actually offer from my own Library anecdotal evidence of a locally written system which I inherited. It was created locally and in good faith, then maintained economically by a member of the Computer Centre. But he then, moved to another institution, which, not unreasonably, charged for this service, albeit at a favourable rate. Finally, the gentleman concerned set up a private consultancy and charged a full economic rate for the maintenance of a system whose source code was now incomprehensible to all but him.

Another separate but common strand is the Library's relationship with the Computer Centre. This fails into one of three obvious types. One in which the two services both ignore and are ignorant of each other; a second where they are in competition, whether for resources or for roles, and finally and most helpfully, a type of relationship where there is a beneficial partnership based on mutual respect. This last is much the best because, if nothing else, both know that computers cost a lot of money and are not frightened by the fact. The Library and the Computer Centre together may account for over ten per cent of the institutional budget; if they work in tandem, it forms a powerful force.

That leads us to the Administration. Each institution will have one or two powerful administrators and their support, or at worst their neutrality, is required to get the resource to purchase an automated system. This involves presenting a case which is not only good, but which can be presented publicly as reasonable. The real reason for spending the money is probably quite different. There are all sorts of reasons for getting your computer system: a good track record for money management, a history of reasonable demands, being he who shouts loudest, being a "sound fellow", competition with neighbouring institutions, a means of doing down other parts of one's own, institution, keeping Senate quiet, mentioning that it will aid centralisation, that it will aid decentralisation, that the Vice-Chancellor saw one in Harvard, that it saved staff (or these days the more sophisticated truth that it lowers unit costs), that there is a visible crisis in the Library and even, rarely, that it forms part of a long-term strategy for information management. Whatever the final reason for the Administration agreeing to support the funding of a system, it is most .unlikely that virtue and logic will be the conclusive arguments. The difficult part of the process is winning the support of this group. Having done that, it is then merely a matter of operating the committee system in the proper manner to make the formal decision come out.

Whatever the reason for the Administration agreeing to support the funding of a system it is most unlikely that virtue and logic will be the conclusive arguments

Within any institution there is a basic need for two kinds of people to be involved in the implementation of a computer system. Of course you need the legman, the one who will install the system. This may be the Librarian, but is more likely to be of the library staff. But more importantly, you need a product champion. If he or she has achieved a position of institutional status, this may be the Librarian but it is more likely to be an academic who has the confidence of his academic peer group or a local authority councillor who has the ear of his caucus. This phenomenon is best illustrated by the caste system found in macaque monkeys. If you take a high caste monkey from the troop, you can teach it to put money in a slot machine, pull a lever and receive a banana. Put the machine and the monkey back with the troop and he will teach the tribe to perform the actions to get the banana. If a low caste monkey is removed from the troop, it too can be taught to put money in the machine, pull the handle and get the banana, but put this monkey back in the troop with a machine and the others will let it put in the money and pull the handle, and will the. steal the banana from the lower caste animal. In short, someone credible has to be the salesman.

Raising the money

It seemed difficult to get through this talk without mentioning the dreadfully vulgar subject of money. Errol Flynn put it nicely when he said that his problem was matching his gross habits to his net income. One of the frightful things about library automation systems is that they cost an awful lot of money. If one includes some of the database costs, an average system will require resource of about half a million pounds. It is my view that most libraries find sums of this sort difficult to raise because they regard them as some form of objective truth rather than as an optical illusion. Because our normal equipment purchases are things like kiksteps, cash of this sort becomes a major problem. Suppliers at 1east recognise this phenomenon of the optical illusion. One said at a recent conference "it is in fact a remarkable phenomenon that a product is sold with the promise that it will be consistently enhanced at the manufacturer's expense"."' So much for software maintenance charges! One library I know, frightened the wits out of its Administration by totalling all the worst case cost hypotheses and coming up with a requirement for several million pounds. Conversely, one colleague of mine, well known to many of you here, lists raising money as a hobby in the latest Who's Who in Librarianship and he has been successful in extracting money from a university which admits publicly to be on the verge of bankruptcy. I tend to his school of thought as well. When, asked what does two and two make, the answer is 'what do you want it to make?' Although I do think such large sums to be figments of the imagination, monopoly money even, at some point someone has to pay a bill for two or three hundred thousand pounds and that requires a sensible alliance either with the Finance Officer or with the Vice-Chancellor or preferably both.

Several hundred thousand pounds may sound a lot of money, but it is a besetting sin of British libraries to think small. Remember that for a midsize university it's about three days' cash flow and for a local authority even less. For a medium sized university library like mine, or a local authority library like Kingston-on-Thames, where I live, it may only be about two months' cash flow. Such obvious tactics as spreading payments over the financial year-end to take advantage of two financial years can make the problem seem trivial beside such emotive issues as senior:junior academic staff ratios or the allocation of space in university car parks. There seems to be a growing view that we must aim for appropriate technology rather than information technology. It's not quite 'come back edge notched cards, all is forgiven', but a willingness to give up the battle and begin by asking only for an inferior option. However, there is another way; I share the opinion, of Richard De Gennaro, who takes the view that "one should make no small plans which have no magic to stir the blood". Make big plans and aim high.

So how do these financial issues affect internal politics? In several ways. Again I'll use examples from my own Library but they are very much examples. Some parts of the system are separate cost centres such as the Medical School. They can be tackled separately and additionally for funds through the judicious deployment of arguments; these may depend on fear envy or greed as appropriate. In our case, the mere threat that another London Medical School might come on to our system ahead of our own Medical School was a powerful case. Most departments have their share of technocrats and, with luck, some of them may prove powerful allies. Certainly those academics who have some experience of computers are usually enthusiastic supporters of library automation. Most departments have access to endowments, grants and even research funds. Assistance can also take the form of planning advice, gifts of equipment, staff help and so on. Specific examples of this I have come across are: paying for the purchase of catalogue records and so jumping the queue; planning the wiring of a building; presenting the gift of a second-hand air-conditioning unit; installing and running the Library computer in an existing machine room without charge; paying for OPAC terminals within a department; paying for pad lines. The trouble with all this, of course, is that while it reduces the apparent cash outlay, each item requires separate and often protracted negotiation - a lot of time spent on the King's Parade and a lot of horse-trading which may offer unwelcome hostages to the future. Nor is it always obvious to one's own staff why some departments seem capriciously to be receiving favours, and that may in turn prove a source of difficulty.

So, having acquired the resource for the initial purchase of a system, how is it to be installed? Although the arguments for the automation of libraries, are now by and large taken for granted, the argument for linking it all together is less certain. For all those entranced with McLuhan's 'Global Village', there are others convinced by Mumford's less well known concept of 'Electronic Entropy'. (6) He argues that automation as progress is a dangerous chimera. The sheer size of computers coupled with their mechanical approach to problems leads to a loss, not an increase in information. Their logical end is therefore the decay, not the increase of knowledge. When John Ruskin (who was no lover of innovation and change) was asked for his views on the completion of the laying of the first cable from England to India he asked "What have we to say to India?" It does seem reasonable to be in a position to justify, if only to ourselves, why we want to spend all this money and what benefits it will bring. The classic rule of automation is to automate what you want, not what you have. I am firmly with that group who reject the mechanisation of library housekeeping routines as the goal of automation. I am with that group who have watched institutions pouring funds into those gold-plated mastodons, while a revolution in information technology and information management has begun to emerge around us. Librarianship is less and less about the contents of any one library and more and more about the ways in which libraries can link and co-operate.

PoIitics within the library

I'd like to move on briefly to the internal politics of the library when it comes to automation. The implementation of large systems implies priorities. First there can be the functional ones, say Circulation before Acquisitions, then there are the sectional ones. Most libraries these days are multi-branch organizations and this imposes its own political stresses. The scale of the problem will of course depend on the previous history of management of a distributed system. Fred Ratcliffe of Cambridge University has commented that "The centralised/decentralised library argument is one of the most enduring, controversial and futile themes of university librarianship of this century". This controversy is made more pointed by the fact that it is often a debate between users and librarians, with the former as the decentralises and the librarians as the centralists. It is true, as we all know, that academics prefer to have small, private libraries close at hand. When these exist, for whatever reason, the Library will prefer to control and staff them, rather than let them flourish in freedom. Very often the best of our subject specialists are there. 0ver time they develop autonomous systems which bear only sketchy relationships to the Main Library. We should remember, of course, that these arise quite legitimately to meet perceived local needs. In extreme cases the local library staff begin to identify more with the department or faculty than with the Library. Automation is then seen as a threat on two counts. It is a feature of academic branch libraries that they tend to contain discrete subject collections rather than the universal collections more usually found in the branches of a public library system. If the Library is in a clearly defined subject area, such as music, it is not always apparent what the benefits of automation are, and there may even be a feeling that the branch Library will be swamped by outsiders at the expense of the 'real' users. Secondly, the introduction of automation threatens autonomy and an independent style of management. The centre will often attempt to impose uniormity of approach, a standard view of fines policy for example, or a narrowing of the occasions when loan regulations can be over-ridden. At the opposite extreme to this are the fire-eaters and technobrats, the branch librarians who see automation, as the chance to pep up service, to prove the value of the Library - and by implication, to improve the status of the Librarian-and to open up new vistas of service. Neither of these polarised views is absolutely true, although both have elements of the truth. In reality, what determines such issues as the priority for implementation between sites are much more pragmatic and prosaic considerations such as the ability of British Telecom to provide a telecommunications link within the expected lifespan of the system. In practice, this debate is an unreal one. Libraries are the camp followers of the academic world, and wherever the academics pitch their tents, like the camp followers, we set out to service their needs.

You will all remember that Ranganathan said that wherever people habitually congregate, that is a site for a library. (8) Wilfred Ashworth wrote the classic works on multi-site libraries and came to the conclusion that no matter how wasteful and unhelpful he might find multi-site operation, that was the institutional pattern and one just had to make the best of it and accentuate the positive features of open management styles and the ease with which users identified with cosier and more comprehensive systems.") This has to be the correct attitude.

In reality, what determines such issues as the priority for implementation between sites are much more pragmatic and prosaic considerations.

Although Ratcliffe quite rightly sees the debate on centralisation or decentralisation as sterile, and it has certainly produced a large literature, it has been given a fillip in recent times by the appearance of the same debate in terms of institutional automation. Should there be large powerful central computing facilities or distributed power? The power available on the desk of an individual is almost frightening these days, and at the risk of repeating the point made earlier, I would suggest that librarians must come to terms with new technology, not just because it is a tool for improving efficiency, but because it is bringing about a revolution in the way the world is organised. We are like farriers at the turn of the twentieth century. We can stick with the horses as part of a respected trade or we ca, recognise the arrival of the motor car. This change in circumstances and the curious parallel with the problems of Computer Centres is important to the library politician, for in the right circumstances there may be possibilities for excellent political alliances. Although I said some time ago that money was a figment of institutional imagination, there is never enough of it. This means that most large systems will either be underfunded or phased in. Further, a

Librarians must come to terms with new technology because it is bringing about a revolution in the way the world is organised successful system will create its own momentum of growth. So there is not enough to go round. Not enough ports, not enough terminals, not enough data.

The management of change

There has been an unrelated tendency over the last twenty years to introduce at least degree of democratisati.n into the management of libraries. Most chief librarians would now pay at least lip service to the notion of team librarianship and everyone having a voice in debate, if not decision. This breaks down when priorities have to be set, and the Librarian, in a more or less democratic system has to arbitrate between demands. The trouble with arbitrating is that you may appear arbitrary. Veaner has emphasised the way in which automation is changing libraries and the consequential importance of library managers recognising the social forces at work in their organisations. This in turn implies that the problem has to be managed and the advantages of change must be sold. We have all witnessed the deskilling of cataloguing in the last decade or so, but we have also witnessed the simultaneous rise of computer-based information systems which we set up to benefit readers. Selling a system internally is important. The only quarrel I have with Veaner is his assumption that humans are always in charge of the systems they operate! The static nature of libraries is infamous. Mastering a new technology is more difficult than mastering a new task, and the quite natural fears, which surround change and lead to resistance, are encompassed in a growing American literature on QWL - the Quality of Work Life. (1 2) This is an interesting concept that bears further exploration. Implicit in it is the notion, of the library manager as the 'shock absorber' of a changing system, reducing change and breaking it up into manageable portions. One of the biggest problems I have found in producing change in a distributed system is that what one is in essence doing is producing a different culture, a different ethos and a different set of precepts for the organisation. This is hard enough at any time, but it is the basis from which all else flows. Yet that proselytising is an enormously hard physical task when it means, as in my by no means untypical case, getting round twenty libraries in a territory covering a hundred square. miles. Since most of these branches are subject based, one does not even always have the option, of reorganising staff so that the converted can be put in key positions. You can't just swop a medical and an education librarian at the drop of a hat.

Automation is changing libraries and the advantages of change must be sold

The most useful conceptual model here is Leavitt's classification of organisations as a diamond with four elements: People, Structure, Task and Technology. These are seen as interrelated and mutually adjusting, which is another way of saying that you can't change one element without affecting the others. Tasks are the day-to-day routines that make up a job and these can easily be changed. Technology is seen as the principles, which underlie tasks. Structure refers to how the responsibilities are distributed and roles co-ordinated. People is the most complex. and difficult part, and the most resistant to change. As well as the individuals, it is the set of values, norms, customs, assumptions and beliefs which characterise the organisation. It is potent, elusive and therefore difficult to charge. Johnson has noted that "Any innovation that individuals or groups perceive as intruding on their territory, limiting their autonomy, reducing their influence, or adding to their workload will be resisted and will cause staff fears". It is here that the notion of Quality of Work Life and how change affects it is usefully opening up a debate on social factors.

I don't propose to go into detail about the type of decisions which are made on implementations. These are always institution specific, and the much more important issue today is an understanding of the political process through which decisions are reached. I hope that I have managed to demonstrate the variety of political inputs needed, at all levels, in winning approval for the acquisition of an expensive computer system which will affect the working lives of all members of your university or similar organisation. This very universality makes it a potential political issue for all members f the institution, but it is part of the job of the Librarian to walk the King's Parade and ensure that Bacon's advice of four hundred years ago is followed: money is like muck -it is no use unless it is spread. And then, when you have got yourselves a system, when you have climbed your Everest, you should be able to repeat (at least under your breath) that memorable but little known remark from Sherpa Tensing to Sir Edmund Hillary as they planted their Union Jack on the top of Everest: "We've done the bugger".

I started this talk by quoting Somerset Maugham's view of politics, went on to Machiavelli in the middle, and I want to conclude now by quoting the great political theoretician Edmurd Burke, who said "All government, indeed every human benefit and enjoyment, every virtue, and every prudent act, is founded on compromise and barter.” That seems to me the perfect description, of institutional politics.