I made my contribution to the discussion and it is on record. Other speakers were Dr Garrett Fitzgerald, R Kelleher, Noel Farrelly, Michael McGinley, Dr Miriam Hederman O'Brien and Joe Durkan. In the SSISI minute my status is incorrectly upgraded to that of Professor. I give my contribution in full, and add one background comment.

"I am impressed by the way in which Professor Bradley has managed to get the data for his modelling; the two-State situation has raised almost insurmountable barriers to this in the past. I tried to do something along similar lines in or about 1973, and I developed a sketch for a 'what if' model, in Fortran, in which I put emphasis on the opportunity for cross-linking via the sector manufacturing equipment for productive process, which has maximal import-substitution potential. I ran it with dummy but plausible data, and got a feel for the sensitivities. (I remember discussing it with Alan Matthews at the time, and he gave me encouragement. RJ Oct 1999).

"However, when I went to seek real data I fond it so difficult that I just gave up. So I am glad that in the end the data have been obtained, in co-operation with Northern researchers. I look forward to future developments, particularly at the micro level, in the regional hinterlands which were carved up by Partition (and which were studied by Joe Johnston, Ned Stephens and others in the context of the Boundary Commission).

"Dr Fitzgerald commented on the need for an all-Ireland approach to academic research. May I add some anecdotal evidence for how the two-State situation has put barriers in the way of this happening, in a situation which illustrates to cross-linking potential with which I was concerned in the early 70s, and which underpins Professor Bradley's synergetic non-zero-sum game.

"In or about 1977, when I was managing the TCD Applied Research Consultancy Group, we were seeking to market the results of academic research as a service primarily to high-technology industry. We had a micro-electronics unit, which was able to do useful things with 'microchips'. We were also into physics-based instrumentation, and had encountered the Queens laser group, then under Professor Dan Bradley, which had floated a business in Belfast manufacturing tunable dye lasers for the world laboratory market. This was a world-leading enterprise; the tunable dye laser was Bradley's invention. The idea arose of marrying the laser with the chip, so as to stabilise the output against temperature variations, and generally enhance the controllability.

"Being a small firm, and limited by cash flow considerations, the grant-aiding of research was of prime importance. We both made enquiries of our respective State agencies, and found that in the Republic there would be no grant for the R&D because the production would be outside the State, and in the UK there would be no grant for R&D done outside the State. The firm, regrettably, did not survive the move of Bradley to Imperial College.

"May I therefore strengthen somewhat Dr Fitzgerald's call for 'all-Ireland academic research' and urge the setting up of an all-Ireland agency for the support of near-market R&D in support of innovative enterprise, with the objective of enhancing the cross-linking and the synergy, and the maximising of the draw-down of EC funding for R&D which enhances cross-linking over political boundaries."

Econometrics and the Philosophy of Science

This was an in-depth critical review of econometric modelling procedures, and at the end the author made some remarks relating to the work of Newton and Einstein. These made me feel somewhat uncomfortable, and I responded critically. None of the comments is on record in the published proceedings, though Anthony Murphy (UCD), Patrick Waldron (TCD), Tony Kinsella, Fergal O Brolchain, Bob O'Connor and myself were minuted as having contributed. I submitted my text to the Secretary Donal de Buitleir on May 21 but for some reason it failed to get into print. I have it on record, however, and here it is.

"I am somewhat uneasy at the reference to Newton and Einstein. I don't think we have a valid comparison here. Newton was able to produce a model of excellent predictive value for the motions of the planets under gravitation. Einstein produced a very marginal improvement in the predictive value of the model, and also produced an underlying geometric reason for the law being inverse square; this for Newton was an ad-hoc assumption.

In physical systems, once you get the basic dynamics right, it is possible to model the motions of ensembles of basic units (atoms, solar systems, galaxies) statistically, and come up with laws of high precision (eg the gas laws).

If I may be forgiven for commenting as someone with a physics background among the economists, nowhere in economic theory does there appear to be anything remotely approaching this level of predictive precision. This is not surprising, given the complexity of the system under investigation, and the imprecision of the definition of the basic unit and its laws of motion. Economic theory is, I suggest, at the stage of chemical theory pre-Dalton, before the discovery of the atom, when it was dominated by abstract concepts like phlogiston.

The 'atom' in economic theory, I suggest, is the firm, the entity which makes the investment decision. The key dynamic effect is the capital investment process. The behaviour of an economy at macro-level is a statistical ensemble of consequences of capital investment processes at the micro level. Pareto's Law should come out of this analysis, like the gas laws come out of the Maxwell-Bolzmann statistical mechanics.

However I look in vain for indications of any approach along these lines any time I encounter an econometrics paper.

Perhaps economic theory (and its econometric representation) should be seeking to understand, and 'predict' with its models the behaviour of, simpler systems such as existed in the past. We are rapidly approaching the 150th anniversary of the foundation of the Society, which I suspect came into existence for the purpose of 'Inquiry' into the nature of the Irish socio-economic disaster of 1845-47 known as the Famine.

It might therefore be appropriate to address the problem of how to model, with modern techniques, the behaviour of the Irish economy prior to, and in response to, this catastrophe. It would be interesting to do this comparatively with other countries of comparable size (eg Denmark), where there was potato crop failure, but no famine. What policy levers could have been used in the Irish case, had there been an independent State?"

Science, Technology and Innovation: the STIAC Report

I have however on record the letter which I sent to the Office of Science and Technology, in response to their request for submissions:

Eugene Forde
Secretary STIAC
Room 124 / Dept of Enterprise and Employment
Kildare St
Dublin 2

April 18 1994

Dear Eugene Forde

Here finally are the substantive points I wanted to make, in response to the questions as listed.

(i) and (ii): The indications are that S&T policies are very far from serving Ireland well. We are producing plenty of good graduates; too many leave; too few are absorbed into innovative productive enterprise. Value for money from State expenditure is difficult to measure or to optimise. I would be glad to help to develop a measure if the opportunity arose. There is some experience elsewhere, none of it much good. This would be a good research project.

(iii) I advocated 2 decades ago in the Irish Times that all staff in teaching organisations should have as of right a research budget, to do what they like with. This would imply a separate head for research, distinct from salaries and overheads.

The total amount should be a policy-determined % of the overall State R&D budget, and the amount so allocated should not be the total funding available for basic research; simply a floor value.

An additional policy-determined % should be available in a lump, for allocation to supplement the pooled resources of collaborative groups who get together to put a reasoned proposal to an appropriate peer-review group.

The composition of such a resource-allocating group has been regarded as a problem. Up to now it has always been State nominees serving on some committee. There seems to be a basic distrust of the democratic process. I would prefer to see researchers themselves, via their voluntary organisations, having an input to the composition of the resource-allocating group. In other words, the recognition of the political process and its bringing into the open. This would help to get the issues discussed and understood. The AGMs of the voluntary bodies would become occasions of an interesting an impassioned policy-generating debate.

(iv) Investment priorities: I have no clear overall idea, but I think priorities are probably not right; again, this is a research project in which I would be interested in participating.

I look at the current state of our largest State scientific enterprise, what was the Agricultural Institute, now Teagasc, and I see a sad state of decline, in what should be a vibrant growth area, stimulated by the need to go for up-market quality and away from bulk commodity disposal operations. One negative factor was the way it was founded with a large initial recruitment of a single-age cohort, now all near retirement. The current decline was totally predictable given the initial establishment policy, and indeed I seem to remember predicting it. It is necessary to plan for a creative age-mix at all times. The key investment policy is in people, and in appropriate structures to enable people of differing ages to interact creatively.

(v) co-ordination: encourage cross-linking and formation of teams from the bottom up, by rewarding the process when it happens.

(vi) Planning in detail: the less the better. Simply provide the support and the structures, and let the people do the planning. Don't change the support system or the structures except in response to perceived demand from below.

(vii) Institutional structures are too centralised and too State-dominated. The voluntary sector needs to be upgraded and given a recognised influential resourced role. Let me outline what I think might be a good structure:

1. Construct a State S&T policy group out of appropriate panels of a reformed Senate, such that the revised electoral procedures reflected the suppliers of S&T personnel (ie the universities and colleges), the existing working S&T personnel (via their voluntary organisations) and the principal S&T users (industry, agriculture, services). Let this lay down the policy ground-rules for the S&T section of the budget.

2. Bring applied S&T delivery as near as possible to the end-user; in other words, encourage S&T graduate recruitment, and support the recruited graduates with R&D services nearby. This reconstruction would involve the existing centralist State R&D services being devolved into regional autonomous units, each closely associated with a 3rd-level education centre (RTC or university). Replace central hierarchical systems with networked autonomous units, linked by means of state-of-the-art communications; a 'post-industrial' type of organisation. The association of the source of R&D and consultancy with the source of the graduates would itself reinforce graduate recruitment, thus enhancing the uptake of innovative ideas in firms by people who understood their implications.

3. Encourage basic research with college staff unconditionally funded for this purpose at a basic level, and with supplementary funding available for collaborative groups who propose projects for funding (see (iii) above).

(viii) The measure of the innovative capacity of a firm would also be an interesting research project; I doubt if we have a good generally-accepted measure. The work of Kingston in TCD and Cogan in UCD may be relevant here. I can think of several dimensions in such a measure: graduate count, role of graduates in organisation; type of organisational structure, links with college-based R&D groups, uptake of EC-funded projects, uptake of projects funded under the old NBST university-industry schemes (which need to be analysed now in historic mode, as 'cases' in a case-file, generators of corporate experience).

(ix) Why are we less successful? There are deep historical reasons for this; for example, distrust of innovation is embedded in the national business culture. A large and successful firm which saturates the Irish market profitably and has cash for expansion will usually head off abroad and do the same thing: this was been the case with Guinness and Jacobs in the 30s and is currently the case with CRH and Smurfit. The tax environment must here be an important factor. The result is that it is very difficult for an innovative start-up firm to get early seed capital, such as to enable a prototype to be engineered and brought to the market. There needs to be the maximum possible tax incentive in this area, with the minimum of bureaucratic constraints.

There are other more subjective factors, like in the education system the emphasis on conformity and looking for a 'P&P job'. This is all part of our inherited fringe-imperial role in the British Empire. Those who generated modern Irish national identity in the early 1900s seem to have been largely unaware of this as a problem, and were consequently unaware of the important need to divert the objectives of the then Irish scientific Establishment (which was quite formidable by European standards, and certainly no less evolved than the Danish) away from the Empire and towards building the independent nation. This also is a research area, to which I am trying to give some attention in marginal time.

There was little or no explicit recognition by the Government of the role of S&T in economic life until as late as the 1960s, in the aftermath of the 1964 OECD ('Lynch-Miller') Report, by which time much erosion had taken place. The role of Government prior to the OECD Report needs analysis. Bradley in DCU has I believe looked at the Emergency Science Committee. The role of de Valera and the foundation of the DIAS needs to be analysed in relation to science thinking as it was then; this at the time was a relatively important investment by the State in science; its impact was largely sterile in the national context, although it did earn De Valera an FRS.

There are signs that the ANC in South Africa is not making the same mistake; they have set up an S&T Policy Group, and are aware of the need to win over the largely white scientific human resources towards building the new integrated nation. I can claim to have helped to point them in this direction.

(x) We appear to be good at pulling in EC-funded R&D money, but not so good at the follow-through. In some cases this is because the commercial component of the R&D consortium is abroad. Where the commercial firms in the consortium is here, and is small, the development of the results of the R&D into a marketable product is fraught with problems, mostly due to the underestimation of the amount of further money that has to go into the downstream development, and the difficulty in marketing innovative products from a source which may lack market credibility abroad.

Where the Irish component of the consortium is research-oriented, there appears to be a cultural block, based on negative experience, against dealing with an Irish firm in the exploitation; I have heard people say that they prefer to deal with firms abroad than with Irish firms. This must be a reflection of the lack of graduate uptake and innovation culture in Irish firms.

'Critical mass' in the S&T base is not necessarily a problem; the important factors are the quality of the team and its linkages.

[I am currently, in my day-to-day work with my prime client, concerned with finding applications for what has come to be known as 'Case-based Reasoning' (CBR) as a means of making constructive use of corporate memory; this is state-of-the-art software which has emerged from the AI area; we have it as a result of having taken part in an EC consortium with firms in France and Germany and a university in Germany. The problem of making use of the results of R&D such as this in the Irish context is embedded in the local cultural environment, and is an aspect of the overall problem being considered by the STIAC. State-funded S&T projects in the past would constitute an excellent 'case-file' in the CBR sense, which would constitute the corporate memory of the State in regard to its S&T funding experience. This could be a development area, and I have put a preliminary proposal in via Forbairt (cf Owen McBreen) and via the OST (cf Michael Fahy).]

(xi) There should be strategic networking when putting in EC proposals, if some means could be found for doing it which was not bureaucratic. If there were systematic use made of Internet, and an accessible knowledge-base of current activity and credible expertise, it would be helpful. Who knows how to do what, and where are they. If I wanted to put together a proposal, I could then see who else might be in the field, and ask them, with a view to joining forces.

(xii) I don't think we can legislate for future marketability of basic science. Let it just happen. Those with an eye to marketability will pick it up when it is ripe. It is important that such people should be around, and in intellectual contact with the basic science people.

(xiii) The biggest obstacle to development of linkages between industry and college-based research is dependence on bureaucratic hurdles. If a firm puts money into a college unit, there should be an automatic multiplier, and the money should be tax-deductible. It should be as routine as a headage-grant. It should not have to have committees looking at it, and should not have a failure rate such as to cause people to lose heart. The industry contact-point for a college should be an accessible location looking outwards, and should have someone in it who knows the research scene both in the college and in industry. We had this situation in TCD between 1976 and 1980 and we had the makings of a good model. It is a pity it was not continued, and it would be interesting to understand what factors contributed to its abandonment, and what factors if they had been present might have enhanced its role. I would be in a position to provide analysis of this experience in depth, in the context of a 'case-file' being set up as a tool for the work of the STIAC (see under (x) above). (xiv), (xv) Awareness of R&D at top level: a good start would be to reflect it into tax concessions; use the IMI conference and similar events to promote R&D and innovation and the new friendly tax environment; make it easy for small firms to start up with R&D-minded leadership and become bigger firms (what I may call perhaps the Mentec process) etc

There is also scope for the development of some sort of award system, on the Nobel Prize principle, such as to draw public attention to achievements of national significance. The RDS has tried various schemes in this direction, such as the Boyle Medal, instituted in 1898, but this has been under-resourced, and the awards have lacked impact; nor does the philosophy of the award appear at first sight to have been consciously linked to any national cultural objective. I am currently looking (in marginal time) into the history of this award, and expect to be able to write about it shortly.

The annual Aer Lingus Young Scientists Award would appear to have achieved a high prestige value, and there is a trend into regarding this event as a generator of innovative enterprise ideas, although in my opinion in this age-group such an emphasis is premature.

(xvi) There needs to be an active policy for encouraging research scientists not to stay on in colleges as 'post-docs' but to get into downstream research, via some sort of industrial fellowship scheme, encouraging mobility leading to career change. Career changes need to be made easy and normal; people should be able to move upstream and downstream without barriers.

(xviii) I don't think we are as bad as all that; look at the excellent international results of the Aer Lingus Young Scientists. There is excellent seed corn. The danger is that it becomes regarded in the culture as 'something for the kids to play at'. There is need for upgrading science in the national culture. This is an important area of research, and I would like to be able to do it in prime time. There is no university-based research group looking in this area. It could evolve out of historical studies. In Limerick they are introducing a technology for humanities module (cf Eamonn McQuade).

(xix) I don't think the Irish bureaucracy can pick winners. By facilitating people who know their fields they can however encourage Irish S&T people to position themselves so as to be associated with winning international teams, and sometimes lead them.

(xx) The less we have to do with current UK official thinking the better; the UK shows signs of terminal decline; we should rather look to the Continent and to the USA for role-models. We should perhaps look at the areas of damage and see if some good people from Britain, who might otherwise go to the US, might be picked up. Where research systems that are needed are in crisis, we should perhaps see if we could supply the need from here, via a local rehabilitation process, rather than in vulture mode.

*** This concludes my response under the suggested headings. If I were to add to it, it would be along the lines hinted at in the paragraph under (x) above which is in italics. There needs to be a systematic approach developed to the establishment of a 'corporate memory' in the State system of all 'cases' where State money has been used in support of a research project in science, or a development project in technology, or a transition from one to the other.

Such a case file could place a query-able abstract of each project file on record, including an assessment of its long-term outcome when this is known. With such a file, it would become possible to develop policies systematically.

The firm I work on contract with, IMS, has got the capability for supporting such a system, using software developed recently in an EC-funded project, in association with two university-industry development-type firms in France and Germany. The approach is currently in use on a specialist basis in the aircraft industry. We are seeking opportunities to assess the possibility of generalising the approach, in several high-tech area where fuzzy policy issues exist and require to be clarified by analysis.

We would welcome the opportunity to go into this question in greater depth, using an appropriate sub-set of the State's S&T project funding experience as a pilot-study. This, I suggest, constitutes an opportunity for the STIAC to contribute to heading (x) directly.

Yours sincerely

Roy H W Johnston PhD FInstP CIEI

Innovation Policy in Ireland

There was a prepared comments by Dr David Jacobson, who supported decentralisation and institutional diversity as a valid consequence of the intrinsic complexity of the science and technology scene. He was critical of the assumption that existing STI policy was determined neo-classically.

There was also one from Dr David McConnell, who identified industrial policy as being basically diffusionist (ie accepting innovatory technologies primarily from abroad), providing a cultural background within which it was somewhat difficult for science policies to help support an 'inventive economy'. He took the biotechnology industry as a case study, this being a derivative of an area of science to which he himself is a world-class contributor. He called for investment in the region of £200M per annum into basic science in Ireland, so as to avoid the dispersal of the best people abroad.

On this occasion it could be said that the substantive meat of the evening was in the prepared paper of the seconder of the vote of thanks, the paper being a peg to hang it on.

I quote my own contribution in full:

"I am somewhat uneasy at the level of abstraction with which the economics community attempts to examine the 'science, technology and innovation' (STI) process. I would be more comfortable with an approach rooted in the theory of the firm, and in the nature of the constraints experienced by a board of directors when making an investment decision.

I should suggest in passing that it should be possible to derive the laws of macro-economics from a statistical model of an ensemble of firms operating in an economy, by analogy with the way one can derive the gas laws from the statistics of the kinetic motions of the molecules. I take the liberty of suggesting this as an item on the agenda of the economic modelling community.

I noted at the time John Bradley published his economic model that he had allowed for several parameters which seemed to me to relate to, or perhaps even represent, the STI process, and I welcomed this as a link-point for discussions such as this. Later, when on behalf of the Irish Research Scientists Association (IRSA) I was exploring the possibility of persuading an economist to help them lobby the government (given that governments, for some reason, tend to listen to economists), I found it difficult to get support for the idea, and I did not pursue it further. Dr Kane's paper however constitutes a link-point around which this approach might be re-opened.

An initial approach might be based on a detailed analysis of the differences in culture and environment between Ireland and Finland, taking perhaps as pilot-studies the business and innovative histories of the firms Smurfit and Nokia. I recollect maybe a decade ago commenting on this in a letter to the Engineers Journal. Smurfit stays in wood-pulp and goes global in wood-pulp. Nokia begins in wood-pulp but at a certain stage, having saturated Scandinavia, decides to employ some good graduates and goes into telecommunication. It is now a world leader in a key high-tech sector.

We need comparative international studies of the STI process, in comparable-sized small states, exploring all the factors in the political, socio-economic and cultural environment which influence the innovation culture. The differential between the Irish 1% and the Finnish 3% for GNP dedicated to STI, as exposed in TRE Prime Time yesterday, needs to be explained.

The role of the central State has always been crucial to STI, ever since the beginning of the realisation by States the 'knowledge is power'. The funding of the Royal Society by the State at the end of the 17th century consolidated a relationship that can be traced back to da Vinci and earlier. The key motivation has always been to give a competitive edge to the military system. In Ireland we have not had this motivation, to our credit, and an opportunity exists to dedicate an equivalent motivation to the civil STI process.

There is need to teach about the STI process in third-level education, in the humanities and business schools, and to develop relevant teaching materials, specific to the needs of small 'fringe' countries like Ireland and Finland, and to ensure that STI 'best practice' is embedded in the culture."