EDUARDO CÉSARFinding new strategies to improve and make the most of Brazil’s scientific and technological production is one of the chief concerns of Léa Velho, one of the country’s leading experts in the sociology of science. A professor at the Scientific and Technological Policy Department at the Unicamp Geosciences Institute, she evaluates, strictly and with balance, themes such as international cooperation in science and technology and the assessment of public policies for science and technology. She is critical, for instance, of Brazilians holding only lowly positions within major research networks and feels that it would be appropriate to resume the policy of sending Brazilians abroad to complete their doctorates, in order to create leadership on a par with foreigners. She also criticizes the indiscriminate use of conventional indices for evaluating scientific and academic production, which may generate confusion, often leading to injustice.
An agricultural engineering graduate, Léa Velho began her career as a university professor at Unesp, in the town of Jaboticabal, in the seventies. Her interest in scientific and technological policy dates back to 1978, when she moved to Brasilia, for family reasons, and was invited to work as a technician in the evaluation of projects in her field for the National Council for Scientific and Technological Development (CNPq). In 1985, she completed her doctorate in Technological and Scientific Policy at the Science Policy Research Unit at the University of Sussex in the UK. She left CNPq in 1991 to teach at Unicamp. From 2001 to 2005, she was a senior researcher and director of post-graduate studies at the Institute for New Technologies at the United Nations University in Maastricht, in Holland. Married and a mother of three (two sons and one daughter), she studies not only gender issues in the sciences but has also focused, together with her graduate students, on understanding why there are so few women in fields such as physics, mathematics and engineering. Below, some segments of her interview:
You have written several articles about Brazilian involvement in international research networks, such as those dedicated to bioprospection and to the Aids vaccine. How important is this involvement?
In the recent knowledge production movement, there is a major trend toward collaborative research. This is growing far more than isolated research; it is also mentioned more often and therefore has a greater impact. When one looks at European scientific production, however, one sees that international cooperation has been rising exponentially. Yet Brazil has had a modest level of cooperation. Although cooperation has grown, it has maintained the same level of about 30% of articles written jointly with other countries since the late eighties.
Thirty per cent of all articles?
Thirty per cent of Brazilian articles published in the Science Citation Index involved international co-authoring. What does this mean? It can mean many things, even that we are not very involved with this movement, less than we could be. I think Brazil participates little. My experience taking part in EU projects shows that partnering occurs much more when they are obliged, over there, to include a few Latin American countries according to the call notice. They know that you exist and they say: don’t you want to take part? And we end up involved rather marginally. You rarely get the local support you need, for instance, to travel and to take part in discussions held over there. Consider the case of the bioprospection networks that I have written articles about. The initiative is taken by a northern country and the funds come from there. The people from the countries that actually have biological diversity end up cooperating just to enable the foreigners to gain access to these resources, and not because of their competence.
What do researchers actually gain by belonging to a network?
They acquire skills such as negotiating the objective of the research, the methodology or the use of equipment. When I can do as I please, i.e., when I work alone with resources I myself obtained, I do not negotiate anything. At most, I tell my students to do certain things and they do them. But negotiating with your peers, who have their own ideas about things, is an advantage.
Aren’t there any national groups that are sufficiently mature to deal with foreigners as equals?
There are several mature groups in Brazil that probably take part on a more egalitarian footing, which has to do with their negotiation capabilities. Those who know what they want and have the resources to meet their requirements join the project out of real interest. The researcher joins because he/she has something to contribute. That is not the case of bioprospection networks.
Is this what happened in relation to the drug and Aids vaccine research networks?
In the paper I wrote about this I was trying to see how Brazil took part in the clinical tests for the vaccine’s development. Unfortunately, this is an area in which Brazilian competence is inadequate. From the standpoint of policy design, the Aids program has been very successful in Brazil. We do basic research well. There are very good doctors, but we still need to develop this in-between level, the capability of conducting clinical tests with laboratories that are suitable for doing all the analyses and capable of developing reliable protocols. Brazilian participation in these networks became rather questionable. We must start doing clinical tests for drugs that we ourselves develop; otherwise, we will continue being providers of services for multinationals.
There are virtually no drugs developed in Brazil?
Precisely. To conclude this network issue, we have to solve certain things. Do we want to join these international networks? If we do, we have to diagnose what is going on. That’s the first step. I think that the indicators are a warning. Successful countries, from the innovation point of view, have more dynamic cooperative activities. One of the things we are lacking is more human resources being trained abroad, now that we have strengthened our post-graduate studies. The path taken so far was the right one; I’m not questioning that, but the country can’t close itself off and say that now we have such strong post-graduate programs that we don’t need to train anyone overseas! Few people go abroad for a full doctorate. And many go abroad because Capes (the Coordinating Office for the Training of Personnel with Higher Education) decides to provide 500 grants and chooses those who stand out in each field of knowledge. This doesn’t work, because the few resources earmarked for foreign training are diluted and the country doesn’t take any quantum jump in any field. So we should specify the areas, make a list and send these people to be trained over there. I am referring to the policy of enhancing training in certain fields and of penetrating certain specific networks, since we can’t join all of them.
The complaint about doctorates abroad is that the most talented Brazilians receive good job offers and don’t come back.
Data from the US center that monitors foreign students indicate that most Brazilians do want to come back. If they don’t it’s because they don’t have the opportunity to take a qualifying exam. The truth is that only a few are going abroad.
What about the ‘sandwich’ doctorate, in which the researcher spends some time abroad? Doesn’t it fill this gap?
I don’t see much benefit in this program and it is certainly no replacement for the relationships established during a full doctorate abroad. Generally, the post-graduate students spend six months to a year abroad and frequently they don’t write a single paper in a foreign language. Social sciences researchers who do their doctorates in Brazil rarely manage to publish a paper in English. I think the ‘sandwich’ program may be good under certain circumstances. For instance, to do my thesis, I need to learn a specific technique that nobody is familiar with over here. So I go over there and within six months I learn the technique. But from this to generalizing the ‘sandwich’ as a training policy for human resources.
Companies in countries such as Japan and South Korea encourage their technicians to get degrees outside the country and return abroad from time to time, to keep track of new developments.
There were many Japanese companies, especially in the eighties and nineties, that sent their technicians to do doctorates at Caltech, at Berkeley or at the major US technological centers. When they completed the required credits, they were called back. They were not required to produce a thesis. They just went abroad to acquire certain types of knowledge, take some subjects and take part in the discussion of projects. The degree didn’t matter. The issue was training human resources to acquire knowledge. Having knowledge of knowledge is one of the very important competences that PhDs acquire. It’s knowing who does what and where. When you’re doing a doctorate abroad, you see researchers from a range of universities; you start figuring out who’s who; and you discover where these people publish. People who go abroad for six months don’t achieve this.
The number of PhDs in Brazil grows every year and some estimates indicate that in the next decade we will achieve a ratio of doctorates per hundred thousand inhabitants similar to that of Japan and of the US. The other side of this is that many of those who complete their doctorates don’t have a job. Is there any way out of this paradox?
In the OECD [Organization for Economic Cooperation and Development] countries, which produce half of the world’s wealth, most PhDs work in industry. There are exceptions, like Mexico, but this trend among the Asian Tigers, such as Korea and Taiwan, is even stronger than in some European countries. All studies about the role of qualified human resources for innovation very clearly see that companies achieve major gains when there are people with master’s degrees or PhDs working in their R&D units. These people know where to look for knowledge and interact with those who produce it. A company with no qualified personnel has difficulty with this interface. Well, that said, one cannot oblige a company to hire a Ph.D. Companies don’t fail to innovate because they think this is cool: it’s part of a company’s rational process to innovate when it sees that this will make it more competitive and help it gain or maintain a position in the market. In Brazil, almost as a rule, for a company to remain in the market, it is more advantageous to buy equipment or license technology from abroad. Because there is an excess of PhDs, doesn’t mean that companies will hire them. One has to think about what to do to encourage domestic companies or the subsidiaries of foreign companies in Brazil to create an environment in which it becomes advantageous to innovate using domestic resources and knowledge.
Many companies have started to wake up to the importance of innovation, because they need to export and be competitive abroad?
I think that this is clear for companies, that they must be competitive. It is a new fact, this awareness among firms that they have to innovate in order to remain in the market, to export quality products, or else nobody will buy from them.
But aren’t they changing their behavior?
It’s hard to tell. This line of discourse is recent and only recently have studies of this nature begun. What one does know is that nowadays they have the discourse. But we will only find out in the future whether things are actually changing.
What will be the long-term effect of training PhDs without having jobs for them?
I don’t see any sense in this. We train people to reproduce ourselves, not to perform new tasks. We train people to be civil servants. Many people enter a university, do their master’s degree and their doctorate and then take a civil service exam. Of course scientific production in Brazil is growing. It was bound to, with a large proportion of the ten thousand PhDs produced per year orbiting around academia and contributing to the production.
EDUARDO CÉSARCapes released a ranking that puts Brazil among the 25 countries with the greatest scientific productivity. What does this represent?
Everybody that works seriously with scientific production knows that it is difficult to compare countries’ performances, because they are not equivalent. One country may focus on medical research, another on agricultural research, a third one on engineering, while in another one the social sciences are important. It’s very difficult to compare countries in this way. One may compare Brazil to itself and say: Brazilian scientific production has grown significantly over time. That is valid. But I don’t think it’s as valid to conclude that Brazil’s production grows faster than India’s. What does this say in terms of developing local policy? Almost nothing.
Brazil’s position in terms of number of articles published is good, but when one assesses the impact of these articles, then the country’s position in the ranking is weaker. What does this mean?
Impact is measured by the number of citations. Here we start getting bogged down, because the motivations for mentioning a paper vary greatly from area to area. Physics, for instance, is very dynamic and international. There is a community that is always working at the forefront of knowledge and that has very few journals regarded as important. So it is valid to compare whether Brazilian physics has a large or a small impact versus US physics. However, regarding the issue of the impact of Brazilian scientific production as a whole, one really can’t compare things. There is a principle that all serious measurers of science repeat: “One can only compare similar things.” We aren’t like the others.
The most recent data show that we have almost 2% of global production, according to the Science Citation Index, making Brazil the 15th producer of science in the world. Can’t one draw anything from these statistics?
To my mind, no. We’re growing faster than the others. So what?
Doesn’t that mean that our science is more mature?
No. It means that we’re publishing more in the journals listed in the Science Citation Index, which are considered the best publications. We publish a lot, but the impact of our publications doesn’t improve much. Impact is a measure that gets in the way of making comparisons between areas. One can never say that the impact of a biotechnology paper is greater than that of a physics paper. It would be conceptually wrong; nevertheless, people make these comparisons every day.
And this is even happening in order to evaluate researchers individually.
This is the other principle: “One doesn’t use these data to evaluate researchers individually.” Never. The creators of the Science Citation Index did not devise the database for evaluation purposes, but as a tool for researchers to find out who was working in the same field of knowledge as they were.
Yes, it was created with the following spirit in mind: the world of science is getting very large; I work on a given line of research and can’t cope with looking at all the literature. How can I figure out what matters? Based on citations, the researcher can design his networks. But soon people realized that this could be used for evaluation. Then they started to play around with certain concepts and to find that those who were quoted more often were better quality. But this is very questionable. The reasons why a person quotes a given article, or not, are totally unknown. There is no theory of citations. People still don’t know very well what leads a researcher to quote or not quote a given article. From my experience, I know that second-hand citations are proliferating.
What do you mean?
The researcher refers to something that has already been cited. There are stories and more stories of mistakes that have been perpetuated across more than five hundred papers, because each one copied the other person’s reference. And there are other variables. I mentioned physics as a field in which one can say that international citations reflect impact. But take another area, the social sciences, for instance, in which the researcher doesn’t publish abroad. And that is not a Brazilian phenomenon. For instance, in Germany, France and Holland, social sciences are largely published locally and in the local language. This is related to the humanities’ characteristics, in that researchers write about their societies to be read by their societies. Additionally, the periodicals of many fields in humanities are not the main source of publication; the researchers publish much more in collected works or books. Take a country such as Brazil, in which a significant share of the scientific production, community and resources are allocated to the social sciences: the Science Citation Index doesn’t take these data into account. I’d say that eighty percent of the social sciences production is not covered by this figure that Capes is forever divulging, that is the growth of Brazilian scientific production. Using the impact factor of this periodical as a measure of the quality of a publication or, even worse, of an author is to imagine that anybody can evaluate Brazil’s scientific policies just because the database is available on the Internet: you key in a name, go click and find a number of citations for an author, an institution or a country. Of course that’s not how things work. Interpreting the data calls for a deep understanding of the database’s design, of its conceptual basis and of its methodological limitations.
How is this a loss?
Take an institute like ours, Geosciences, which consists of five departments. It’s the most varied thing in the world. One is Geosciences, another, Geography, which can be Human Geography or Physical Geography; there are people in Scientific and Technological Policy, in Teaching and History of Geosciences and in Petroleum Engineering. For all intents, because we are lumped under Geosciences at Unicamp, we are supposed to be exact sciences. When they do an institutional evaluation, they put us together with the exact sciences. Then we do poorly, because we have fewer publications and citations than Physics, Chemistry or Mathematics. And as a result of that, we lose out in the allocation of internal resources. But can you compare the Geosciences Institute with Physics. It’s like wiping the slate clean and saying: they have to behave like the others in the exact sciences. But we don’t have to behave that way. We are not in Physics. Scientific and Technological Policy is part of Social Sciences. We are here for historical reasons that they are very well aware of.
Are there other, more effective means of evaluation?
Bibliometry is an important tool for evaluation. But it must always be used in conjunction with other means of evaluation, based on a prior awareness of the dynamics of knowledge production in certain areas.
Is it possible to compare Brazil and other Latin American countries?
It’s complicated. How many countries have a post-graduate system like ours? None. Mexico produces the largest number of PhDs in its postgraduate programs, but even so, the evaluation of all these programs is only beginning now, as in Brazil. Evidently, a fair proportion of what we publish comes out of the postgraduate programs.
What about Argentina?
Argentina is far more complicated, because it has very few formal doctoral programs.
Do they have a scientific tradition?
Yes. They have three Nobel prizes. But the structure of scientific research over there is quite different. Argentina lacks this formal postgraduate system for training researchers. Another difference is that we have a postgraduate system based on professors who work full time. Here one earns a decent living as a professor, providing guidance and doing research at government universities, in both the state and in the federal universities. If one gets a grant, it’s a plus. If you take on a project, it’s a plus. But we don’t need this to live decently. In no other Latin American country is this the case. In Argentina, for a researcher to make a living, he must be a part-time professor at one place, a part-time researcher elsewhere.
Is free tuition common in other countries?
That’s another fundamental difference. Our best post-graduate programs are free. This doesn’t exist anywhere else. Postgraduate studies in Argentina have to be paid for. In Mexico, most of it is paid. The good schools training PhDs in Chile, for instance, which are mainly in basic fields, are all paid. I feel proud when I tell people that we do it free of charge. But there’s the other side of this story, which is: “Of course, it’s all free for the spoilt kids of parents who are well off and who studied their entire life in private schools”. True. At the end of the day, it’s true. Unfortunately, only a few trained researchers come from underprivileged backgrounds.
It is said that the reason why Brazil has never won a Nobel prize is the lack of original contributions from our researchers. Is this a problem?
The only way to get a Nobel Prize, if this is an objective, is really to try to examine some of the issues that are very particular to us and from a point of view that is also very particular to us. Renato Dagnino, my colleague here at Unicamp, usually says that Latin America’s only major contribution to scientific knowledge was the Cepal School. The theory of structuralism, of dependence, is a major Brazilian contribution to knowledge. And we got to it by looking at our own reality from our own standpoint. We lose out when we look at reality through the eyes of the North, because they have more appropriate tools than ours for working from that point of view. Perhaps if we could find out what is relevant within our physical, natural and social reality and describe this from our very own point of view we might produce a major contribution to science. I don’t think we should shut out the world. In some areas, we need to use our creativity and exercise a different thought process. However, in answer to your question, I’m not very worried about a Nobel Prize.
Though I think international recognition’s important, I’d like the production of Brazilian knowledge to be really useful in solving our problems, rather than being a mere reproduction.
Any specific area?
Several: malaria, bilharziasis and other neglected diseases, production systems for small farmers, unique sanitation systems. Even simpler things, such as political systems that don’t take it for granted that participative democracy, such as we engage in at present, is the only type of decent social organization. And it’s not just a corporate problem. I’m terrified when institutions such as the Campinas Agronomic Institute or Embrapa have to get money from outside and are driven into solving the problems of large agricultural concerns. What about the small farmer, who is unable to pay? Suddenly, the solution for this country began to be channeled exclusively through the private sector. At the Ministry of Science and Technology, the key word at present is ‘company’. As I worked for CNPq in the seventies and eighties, I can see how the discourse has changed.
What was it like, back then?
In my time, it was about strengthening the public sector. I started working at the CNPq in 1978. The current notion of innovation didn’t even exist. The term was ‘technical change’, a far more linear process, in which one invested in basic research, followed by applied research, then development; and all this would lead to innovation. The line of discourse was: “CNPq shelves are full of research that needs to be made available to companies.” As if the issue was lack of communication.
And wasn’t it?
No. Nobody does research, then makes it available for companies and sees the companies using it. Development for a company is carried out jointly. At that time, people believed in the process’s linearity. This changed when the key word became interaction. The idea is to put the players together. When they are together, they negotiate their requirements and their problem-solving possibilities.
There is another issue that you study, that of Big Science in Brazil. Are these major projects interesting for national science?
Brazil has chosen to move in the same direction as international science. One may question this option, but, once it has been made, I believe you have to join it, otherwise others will. What does joining Big Science mean? First, we must define what Big Science is. There are people who think that Big Science is merely science that spends a lot. Actually, it’s not. It involves the collective work of several fields of knowledge. Take the National Synchrotron Light Laboratory. There are several teams working together, from several areas of specialization. Big Science means working with very large, multidisciplinary teams, with several types of training. They have to negotiate the meaning of what they have done and the papers have as much as ten, fifteen or twenty authors. Maintaining all this costs money. We imagined that the said laboratory would generate revenue, that industry would pay to work with it. But actually, it seems it only manages to exist because there is a social organization and the salaries are paid out of public funds. Another Big Science project that Brazil took part in was the genomic networks.
For a long time, it was important to train human resources.
It was to train people to work in a different way. One can look upon the genomic networks as a Big Science project. It’s a way of working that requires that a molecular biologist acts as an IT specialist, so that he has to acquire bio-IT training. This obligation to work collectively, whereby one person is dependent on the next, is modern in science. Brazil decided to enter this type of knowledge production. The issue is how to absorb the human resources trained through these projects. When I was in Holland, I met three people who had come from this network and who were working in companies in Germany. Three women. They were identified by a firm in Germany and went to work there because here they had no job opportunities other than in academia.
How is Brazil doing regarding the difficulties for women to get ahead in science?
‘regarding access, women in Brazil account for almost half: in some careers more, in others less. Typically, women are still a minority in physics, mathematics and engineering. Here at Unicamp, they have accounted for about twelve percent for the last twenty years. Last year, I suggested to a master’s degree student: “Let’s see whether these theories we read about in papers apply here”. The literature indicated that women’s maths performance, up to more or less the sixth or seventh grade, was similar to boys. But after the seventh grade, which is when one starts teaching more complex mathematics, they start doing worse than the boys. Women lose interest and move into the biological or social sciences. One might think that it is a skill-related inclination. But teachers expect this behavior. They encourage boys more than girls. There are certain social expectations that drive women away from one area and into another. We interviewed girls who had just passed their university entrance exams and several confirmed that they had chosen to study physics against their parents’ wishes. As they are a minority, they have to behave like the boys or isolate themselves. They have to follow a career and behavioral model set by men, in order to last and become winners. This was confirmed by many of them, who said: “There’s a professor who says the following at the end of his lecture: the girls who didn’t understand this should ask the boys.” There are always certain stereotypes that make women’s lives difficult in certain professions. There comes a time when they say: “I refuse to spend my entire life fighting this; I’ve had enough. It is true that women largely avoid these careers. Another explanation is that men go in for technology because they enjoy it.
Is technology masculine?
Technology, as a social product, reflects social relations. From the tenderest age you say to boys: “Put the toy car together.” And they learn to enjoy dealing with things like this. With girls, you tend to say: “Come and watch TV with me; we’ll read a book.” They are not encouraged to play with technology. Many boys we have interviewed explain their option for engineering as follows: “I’ve always loved technology, building objects has always been my thing.” When you ask the girls, they say: “I think your job prospects are better.” But the problems faced by women in these careers are different. The scientific career was devised to be carried out by men who have a wife to look after their kids and their home.
What do you mean?
Science has always been conducted by white, upper and middle class men, right? In an ideal model, one expects scientists to have no set time to get home. They have to come into the lab to check on their experiment at four in the afternoon on a Sunday, or at three in the morning on a Saturday, because science is almost a form of priesthood. Few women are willing to conduct their lives in this way. Women have a lot of other interests, because that is how they are raised, culturally speaking. However, in Brazil, there are elements that result from social disparity that help. As here the women in science are middle class, they can afford to pay someone to look after their children and they can continue working. That’s impossible in Europe. If 13% of physics researchers in Brazil are women, in Germany they are 3%.
Is that the impact of nannies on science?
Not only of nannies, but of large families. Of the mothers, grandmothers and spinster aunts; there’s a social network. In Europe, you can’t rely on your mother. In Germany, school finishes at noon and children have their lunch at home. A day care center is extremely expensive. In Brazil, no matter how expensive it may be, what you earn as a researcher allows you to hire a nanny or pay for a day-care center.