Viki Weisskopf was a great director-general. I will try to illustrate his greatness, first by presenting the size and difficulties of the job that he faced, and the results that he and CERN obtained, and then how his special character and abilities showed up in his ways of working which made it all possible.

I will cover a period of about six years, starting more than 40 years ago in 1960, when maybe some of you were not yet born, when CERN was going through a rapid growing-up process under Viki's guidance. I am giving you mainly my memories, so some of the details may be incorrect and much is omitted, but not, I hope, the general spirit of what happened.

Let me set the scene with a summary of what happened at CERN during this period to show you the problems that Viki, as director-general, had to handle in less than five years, starting from scratch, and how I fitted into his work.

The starting point

Before 1960, CERN had effectively two separate structures: industrial - at the proton synchrotron (PS) machine run by John Adams, which was still under construction; and academic - in the early research programme around the SC machine with Gilberto Bernadini and others.

Early in 1960, PS experiments had just started and the future structure of the lab was being hotly debated. Then the laboratory's director-general, Cornelis Bakker, died in an accident and Adams was appointed director-general in his place. Adams instituted a structure for the whole laboratory, with 12 operating divisions and a staff-type board of directors to assist the director-general. The board had two members for research, one of them being Viki, one for administration and one for applied physics. I had been Adams's chief technical assistant all through the PS programme, so he knew that I was best fitted for a staff (not a line) job and chose me as the member for applied physics in his directorate.

These organizational upheavals throughout 1960 were then complicated by Viki's serious car accident, which sent him back to the US for several months, and by the UK insisting that Adams should come back to run the UK fusion programme. In mid-1961 the Council, still largely composed of the wise men who had founded CERN, appointed Viki as director-general, and I began to work closely with him in areas outside the physics programmes where he did not feel well informed, or immediately interested.

I did not know him at all well at that time, nor he me, and he had not chosen me for that job, so it was good luck that our ways of working matched so well. However, they did match, and I became his informal contact and assistant in many fields that stretched the term "applied physics" to cover accelerators, budget planning, computers, data analysis, European long-term plans, finance committees and on through the alphabet. Today I will speak as one who saw things as a floating planner, Kjell Johnsen will speak as one on the ground who actually did the work, without which plans are nothing.

Viki was faced with four major strategic jobs: to steer a potentially explosive research programme following the highly successful start of the PS machine; to see that the short- and long-term technical infrastructures were planned and built; to get the necessary resources; and to keep the member states and people happy.

We should not underestimate the size and difficulty of these jobs in a Europe that was still recovering from the war, with scientists in many countries lacking experience, with pre-war national traditions often strong, with the need to incorporate new staff rapidly into the expanding programmes and with high-energy physics as a completely new field for many people.

The research programme

This first job was the most exciting and satisfying for him personally, and it is a pity that there is not a physicist here to report on how he did it, with his lectures and discussions with physicists, his frequent contacts inside and outside CERN to identify and resolve problems, and bring in new people. In talking to him I could see that he had several principles: to include many physicists belonging to outside labs and give them the same or greater importance as the existing CERN staff; to encourage the build-up of collaborations, not of national groups; to prevent theoreticians and experimenters getting separated. In looking at a building plan with my assistant Gabriel Minder he said: "If they can't share offices, at least make them share the library and the lavatories."

The absence of infrastructure planning and of adequate resources were both urgent problems from the start. The annual budget was going to overrun a previously fixed three-year total, with the UK delegation as usual calling for a ceiling, if not worse. The success of the PS and Brookhaven AGS machines, along with development work in CERN, the US and the USSR, made it clear that large experimental equipment was needed soon and that much larger and more exciting machines were feasible at reasonable costs for the longer-term future.

To handle the budget problem, the Council set up a working party early in 1962 under the Dutch delegate Jan Bannier, which I fed with data on science expenditures in the member states and CERN programme needs for four years ahead. I was amused how, by plotting national-science cost forecasts on logarithmic paper, their straight-line 20-25% per annum growth rates surprised the group and made our proposed 13% look modest. The committee recommended, and the Council approved, the famous four-year rolling-budget procedure named after Bannier, together with figures for the first four years that allowed us to make a good start on the four-year programme of work that we had presented.

That programme implied that the original facilities at the PS machine and for experiments would be inadequate in the medium term, and that an improvement programme then being studied, with capital expenses starting in 1966, had to be funded from a continuing growth in the budget. In particular, the arrival of large bubble chambers, two to be built at CERN, required an adventurous step into data analysis and large computing power, with techniques and thinking not well understood by many physicists.

Long-term programmes and politics

The Accelerator Research Division, of which Kjell Johnsen was a leading member, had already helped to prepare a paper for the Scientific Policy Committee (SPC) in 1961, which outlined the possibilities of much larger PS machines and of colliding proton beams as ways of reaching higher energy interactions.

Similar design work was already under way in the US. To advance discussions and decisions on such proposals, Viki asked Edoardo Amaldi to run a European committee (the European Committee for Future Accelerators) with very active CERN participation, to analyse and propose a Europe-wide policy for new accelerators, both national and international. It reported in 1963 and its proposal was to build a 300 GeV PS machine and to build the intersecting storage rings (ISR) at CERN on the recent extension of the CERN site into France. This hotted up the existing vigorous discussions inside CERN, which Kjell will refer to in his talk. The ISR would require another source of money, and a supplementary programme had to be added to the CERN convention with the agreement of the member states, with the additional financial, legal and political work that this implied.

Achieving this extension was but one of the problems of keeping the member states happy. High-energy physics was moving towards a concentration of accelerators onto a very few sites, with the possibility of building the next big machine either in Europe or as a single world project. It was altering the functions of national laboratories, and replacing small university groups with large collaborations. All of these trends were difficult to understand and digest by physicists, administrations and treasuries, and needed continuous attention at high level to keep things moving on stably. The SPC had an important role in spreading the word, but also in informing the CERN management of the realities out there.

Viki was insistent in setting up collaboration and contacts worldwide, with fellowships for non-member-state physicists, and in particular maintaining active contacts with the USSR despite the Cold War.

In parallel, the rate of change affected CERN staff, who also needed and received attention from Viki to explain and encourage, particularly in service and administrative areas where staff did not see the immediate results and success of their labours. He made a point of visiting labs and workshops and talking with staff everywhere. He liked to tell the story of when he told a visiting bigwig that he could not see him then because he was going to visit a workshop.

At the end of Viki's term, in December 1965, the result of intensive work in all of these areas was a coordinated laboratory programme, with accelerators running well above design specification; a corresponding research programme with a growing international participation; the start of an equipment-development programme for future experiments; the lab's medium term assured by the addition of the French site and the decision to build the ISR; active European work towards a 300 GeV machine; a budget process satisfactory to CERN and to the member states; and, to support all of this, agreed budgets amounting to some SwFr 3000 million (€2000 million) at today's prices, covering the following five to six years. In parentheses, this gave an annual expenditure reaching SwFr 700 million in today's prices, to be compared with SwFr 1000 million today. CERN has been a remarkably modest demander of the member states' resources over the years since then.

How was it done and managed? The work was done by hundreds of highly skilled people in the divisions and managed in detail by the division leaders. They largely prepared their own work by participating in a system of subject-planning committees, with members from outside CERN where appropriate. The staff and manpower needs had to be presented over past and future four-year periods, which could be coordinated in my central planning office in a consistent format, with enough detail to see how each of perhaps 10 activities in a division had evolved and were evolving, along with statements on the progress of work. I did not ask for finer details, which were the business of divisional planners preparing the activity costs and estimates.

This was the material that allowed me to draft the four-year budget papers for review and modification in top-level management meetings inside CERN, which were then sent to the Council and its committees for decisions.

Where Viki came in

At last I can come to Viki functioning as director-general. In this long story that I have related, he appears only occasionally, and this is an important fact in showing what he did do and what he didn't need to do. The latter, in fact, involves most of the work at CERN during his mandate.

I exclude here the time he spent with the physicists in meetings and lectures in CERN and outside. This, I think, was what kept him happy amid the other problems.

As director-general he would, however, intervene actively for major policy decisions - such as choosing the ISR; for limited issues where personal problems or policy divergences could not be overcome in normal negotiations; and where he wished a particular idea of his to be adopted. He was much less directly involved in medium-sized problems, which he had delegated to the middle and upper management levels, leaving them in peace, provided that they kept in line with generally agreed programme policies, and giving himself time to think.

How was he able to do this? First, by assuring himself of the quality of his senior staff - he was intuitively and by experience an excellent judge of character. He would weed out or sideline, usually in a very humane way, staff that he did not think fit for the posts that they were holding. On the other side, to the others he would delegate real power, let them do their work unhindered, support them in trouble if needed, all the time showing his trust that they were working for the good of CERN and its programme. In this way he encouraged people to grow and shine. I think his title of DG - director-general - could equally well have been read as delegator-general.

He did of course keep himself informed on what was happening, where signs of trouble might be appearing, which he did by his informal conversations throughout the laboratory and outside. He would discuss with me the progress of planning and how well forecasts were holding up, so he had access to the picture at all times at my planning level, and he would talk directly to divisions and chair top-level meetings.

From time to time, when worried, he would talk to me about some point where he felt he had made a mistake or hurt someone, and my answer, which he has repeated to others, was: "Put your regretter on zero," which seemed to cheer him up.

By not getting involved in micromanagement and by referring problems first to the responsible senior staff, he had time to think and to carry out the real jobs of a director-general, which I have listed.

He was very explicit in publicly stressing the importance of all of the branches of CERN staff - physicists, engineers, technicians and services alike - in the success of CERN's work. He made personal friends at all levels that lasted after he left, whom he would meet when he came back to Geneva and to his summer house at Vesancy in France. He was made an honorary fireman there, and given a helmet.

One further quality of a great director-general that I have not mentioned so far is the ability to guess right at the critical moment and stick with his intuition, which Viki showed in forcing through the decision to build the ISR. Although the project was absolutely new and technically hazardous, he ignored a large fraction of his own staff and a negative evaluation made in the US. His comment on this was: "Why did we start CERN? Only to imitate what the Americans do?" I don't think he could have foreseen, rationally, that it and the extraordinary technical success of the project would lead to the replacement of large-proton-accelerator programmes everywhere by colliders, the present stage being the Large Hadron Collider (LHC) at CERN.

What would the future have been like if he had lost his nerve and faith in the ISR? Both continents would most likely have tied up their resources for years building bigger very-high-intensity synchrotrons, stopping at the 1000 GeV level - a not very promising long-term future for high-energy physics compared with what the ISR was able to open up for CERN with the proton-antiproton programme and now the LHC. The physics community owes an enormous debt of gratitude to Viki for his intuition, courage and success in taking what I believe was the most important decision in his whole scientific career. It would be nice to see it publicly acknowledged by the CERN Council and others.