Gee, I thought. Peter is right. There is change knocking on the front door. Should we let it in?
I am watching these developments with interest, but also with some concern. As a continuation of my earlier post on the Marketplace of Ideas, I want to elaborate somewhat on what the differences are between scientific research and the market in economy.
The Economic System
With regard to our economy, the free market has proved to be an enormously
useful tool, as long as its freedom is guaranteed by appropriate institutions. It works so efficiently because the mechanism that optimizes distribution of goods and money is strikingly simple. It is the individual who chooses where to invest and whether an investment is worth the money or not, which then provides feedback to those producing a good or offering a service. One has in this case a large group of potential buyers that judge on offers by accepting them or not accepting them. This leads the goods to obtain some value, a price that develops out of the interplay of supply and demand.
Buyers don't always act rational and/or are not well informed, so there is some error to this process. People eg might not want to spend arbitrarily much time comparing offers, or they are influenced by fashion trends. Decision science is a fairly recent interdisciplinary field that investigates how people make decisions under which circumstances. Still, it remains the fact that the marketplace works quite well. (Though some of it quirks are less than pleasant - eg a large part of the sudden increase in oil price is due to speculation). However, one important factor influencing people's decisions are advertisements which can substantially skew situations (and are thus subject to regulations).
If we are selling products on the marketplace we have sellers, we have buyers, and as long as the system is set up properly this should lead to an optimal investment of time, money and resources - and eventually free capital for further progress. One could say that a product will tend towards some natural value that optimally balances supply and demand.
This procedure of operation for profit has been so efficient and globally lead to so much progress that it is of little surprise some countries preach it like a religion - an 'invisible hand' guiding us towards wealth and happiness. Ideally, the mechanism of the free market converts our individual 'micro-interests' in 'macro-interests' that are to the benefit of everybody. (A lot can be said about the circumstances under which this works or doesn't work, but this is not the aim of this post.)
The Academic System
Systems with individuals who pursue interests, and develop strategies that lead to specific dynamics and trends can be found in many situations. The blogosphere is an example. Here we have the bloggers who are trying to increase the number of visitors, comments, and in-links. One of the results are multiple echos of topics that attract interest. Strategies that are useful are being fast, being brief, and being provocative. I leave it to you to decide whether the outcome is desirable.
The academic system is also constituted of individuals who pursue their own interests, and the result should be that the single researcher's strategies ideally lead to progress. As in all other cases this requires however that the system is set up appropriately and the developing individual strategies indeed lead to a desired trend.
For this, one has to be aware of several crucial differences between the academic system and the marketplace of goods:
- First, I dare to say that within academia the relevant factor that scientists aim for is not money but attention, primarily - though not entirely - attention of peers. Those who want to become rich wouldn't stay within academia to begin with. It is however the case today that financial support is closely linked to attention.
- Second, the most important assumption behind there being something like a 'Marketplace of Ideas' is that an idea has a natural value that can be identified in some simple way. It is here where the analogy fails dramatically. If a company produces a candy bar people will like it or don't like it, you will find out very fast. In scientific research, the value of an idea (or a research program) is eventually whether it leads to progress. Progress might not necessarily be an application but simply growth of knowledge and understanding. Developing and judging on the value of an idea is a complicated and time-consuming process, and the judgement itself is in fact one of the main tasks of scientists. It can however take decades until a research program is fully developed and until it is possible to figure out whether an idea is of high value. In academia, people still discuss today things that have been written hundreds of years ago.
The timescale it takes to figure out the value of an idea can depend very much on the field and also on the project. Theoretical physics specifically is a field that is very abstract and working out an idea takes at least 5-10 years. It can take decades for it to be tested.
- Third, between proposing an idea and there being an external indicator as to its value, there is no way to judge on the promise of a research direction other than peer review. And this is the most important difference between our economy and scientific research: Judging on the value of a research program requires an education in the field and expert's knowledge. Scientists are thus buyers and sellers likewise. They provide supply and cause demand. Until an idea is developed sufficiently and can be experimentally tested, we have only ourselves to judge on each other.
The Fall of the Ivory Tower
Especially because of the third point above, the academic system is extremely fragile and prone to be substantially distorted when under external pressure that influences researchers interests. It is for this reason that traditionally the ivory tower was meant to protect scientists. The ivory tower is now a word used quite cynically to describe the detached academic who doesn't know what people do in real life. But the intention of this detachment was to make scientists independent of financial, political, and social influence. This independence is crucial, and it is not easy to obtain. Rational and objective judgement is essential to science and is simple neither on the personal nor on the community level. The only way that researchers micro-interests can be as objective and well-informed as possible, and that these interests can lead to a desired outcome is to ensure researchers are able to follow their judgement on ideas, unaffected by what fashion, funding and the media says.
This however is presently not the case. There are most prominently three important factors that influence researchers
- Financial pressure: Nothing works without money and typically researchers are funded to work on certain projects, either through grants or because they are hired into specific groups. Projects are funded if they are considered interesting, often based on previous success. Tenure depends on grants obtained. The possibility to hire people depends on grants. The reputation of the place depends on the people and thus on the grants. Now we have a situation where people go where money goes, and money goes where people go, and interest goes where money and people goes - attracting more money and people. A perfect setting to produce bubbles of nothing, based on people telling each other and funding agencies how great their research program is (also called: positive feedback). Money is an extraordinarily powerful tool to direct interests and one has to be extremely careful with using it.
- Peer pressure: Scientists strive for attention and appreciation of peers, which opens the doors to social problems that one has to be aware of and that potentially need to be counteracted by providing incentives, or ensuring appropriate management. Especially in cases when sub-fields specialize there is the trend to misjudge shortcomings of the own research fields, and to neglect criticism by out-group members (this is a well documented phenomenon in sociology). Also, scientists might hesitate to work on topics that potentially could damage their reputation, might prefer topics their peers consider interesting thereby creating fashion trends, and there is the risk that a lot of time is invested in improving social connections which goes on the expenses of time invested in research itself.
- Public attention: Scientists are part of the society they live in and are influenced by what is discussed in the public. It is also not surprising that researchers like to work on topics that cause attention and are appreciated by the public, which is not necessarily a bad thing. However, this delegates quite some influence to the mass media and journalists. In a recent post I mentioned survey results according to which 46% of all scientists consider media contacts to be beneficial to their career. It is a small step from this for researchers to decide working on topics that are of interest for the media is beneficial for their career.
Unfortunately, especially theoretical physics presently suffers from a large competitive pressure that leads people to pay attention to these factors. I am very afraid this significantly affects our ability to judge on each other because we are investing too much time in worrying for our career, and because following our own interests might be in conflict with what is strategically a wise decision. Time pressure and lacking future options produce a tendency to dismiss or just ignore new approaches. This is not a problem that originates in funding agencies who in my impression seem to be well aware of the need of 'transformative research' - it is a problem that originates among the researchers who are afraid of wasting time or money in approaches that will not produce presentable outcome for several years. The 5-10 years in which a research program needs unsolicited support to be fully developed is often not available.
Some suggestions follow directly from the above:
- Don't tie researchers to topics, but let them chose freely. Scientists should be hired and promoted based only on their ability and creativity, period. No other factors should play any role, and then let them do what they want. This holds true also for younger researchers. Cooperation with senior researchers happens naturally because it is beneficial for both sides, but should not be enforced.
- Support researchers for an appropriate time period. In research the weight shifts strongly towards short term positions. From 1973 to 2005 the share of postdoc positions in academic research in the USA grew from 13% to 27% (numbers: NSF). People who are on short term contracts and under pressure to produce outcome (papers/patents) will hesitate to start projects that do not fit within that timeframe. This totally stalls any progress on topics that take longer to be worked out. Have some faith. If a researcher has done well, just fund him and let him do what he wants. In some fields of science (like theoretical physics), writing a proposal that plans ahead several years and then needing to stick to it is in many cases a procedure that is nonsensical and constrains scientists freedom.
- Don't fall for shortcuts. Under time pressure, people tend to look for easy criteria to judge on other people or their projects. Such might be the citation index, the number of papers, prominent coauthors or grants obtained. In all instances scientists should judge on other's work by themselves. This requires that scientists have enough time for this judgement. Time is essential, but too short in today's research atmosphere. For a thought on how to cut back on researcher's duties, see previous post on Research and Teaching.
- Counteract specialization: In the present system it is almost impossible for a researcher to change fields without risking severe drawbacks for his/her career. One of the reason is that researchers are hired into specific tasks, and for these nobody would be hired who hasn't previously worked on the field. Another problem is that grants typically require having former publications in a field to document expertise. It has been realized already decades ago that progress very often comes from interdisciplinary exchange. It is quite ironic that lots of funding goes into such new inter-disciplines while the possibility for researchers to just change between fields (or even sub-fields!) is hindered. Solution again: Have faith in people who have proved to do good research and just let them follow their interest.
- Reward internal communication and criticism: Science lives from criticism and dialogue. Unfortunately, presently there are about no incentives for researchers to invest time into obtaining and communicating an overview on their or other's research fields. It is basically a waste of time since the thing to do to obtain a position is to make oneselves an expert in some niche where one is the only person. Give credits to people and/or institutions who fulfil this task and create an atmosphere in which the relevance of criticism is acknowledged and its importance appreciated.
- Don't reward advertisement: Advertisement of the own product, in this case an idea or research program, is a tactic widely used to sell goods. The aim is simply to influence buyers. Such a procedure is completely inappropriate for scientific ideas that have to be critically assessed, in papers as well as in proposals. Nobody should be punished for openly presenting and discussing drawbacks of the own research program, eg because he or she might appear not enthusiastic or optimistic enough. Neither talks nor papers should overhype promises or forget to mention problems, because these are 'well known'. The latter also causes a significant problem when it comes to communication with science journalists.
I recently had an interesting exchange with Garrett Lisi, who said:
- "It's just that the [academic] system seems locked in a poor state right now, so it seems easy to think of steps to make it better."
I am afraid that the more people realize the present system doesn't work well, the more likely they will think it's easy to improve it and fall for panaceas - cheap solutions offering miracle cures. However, putting emphasis on other easy criteria than the previously used ones is not going to solve the problem, it just moves the problem elsewhere. There are no panaceas.
Consider we'd put an emphasis on the 'independent' researcher who works on an 'alternative' approach, who has plenty of single-authored papers and doesn't like mainstream topics. Consider we'd set up the system to preferably reward this type of person. What would we create then if we leave the selective pressure behind the system? People who strive to fulfill these new criteria. To me it seems pretty obvious that this only cures the symptoms, not the disease: One can also have too many 'independent' people, too much emphasis on 'alternative', and cooperation is as far as I am concerned a beneficial trait. The same is the case for over-emphasizing predictions or phenomenology. Just look at the arxiv and see what the outcome is. Money goes into phenomenology. People go where money goes. More money goes where people go. Interest goes where people and money goes. In two decades from now, somebody will come and say: Hey, there is trouble in physics. They are working on all of these cheap little pheno-models, how is ever something supposed to come out of this?
Therefore, the suggestions I had above are aimed at ensuring the system can self-optimize its outcome, by lowering disturbing influence that deviates researcher's work from their actual interest.
On the long run, the only way I see to ensure progress is to scientifically investigate the situation and incorporate asap offered solutions.
If change knocks on the front door, ask what it wants.
Re-reading what I wrote, I was just reminded of a paragraph from Lee's 1st book
“All there is of Nature is what is around us. All there is of Being is relations among real, sensible things. All we have of natural law is a world that has made itself. All we may expect of human law is what we can negotiate among ourselves, and what we take as our responsibility. All we may gain of knowledge must be drawn from what we can see with our own eyes and what others tell us they have seen with their eyes. All we may expect of justice is compassion. All we may look up to as judges are each other. All that is possible of utopia is what we make with our own hands. Pray let it be enough.”
~Lee Smolin, The Life of the Cosmos (Epilogue)
My FQXi proposal to investigate some of the aspects I discussed here was declined, main reason that it seems to "overlap with studies of the sociology of science, which is a well-developed field, but one in which the principal investigator appears to have no direct professional experience or training". That is entirely correct. I am just a physicist who has had too much time thinking how the academic system sucks, wondering why nobody in it seems to listen to what the sociologists say, and why said sociologists don't come up with practical advices (interdisciplinary research, anybody?). To those of you waiting, this also means there will be no financial support for grad students to participate in our upcoming conference. I am genuinely sorry about this.
This post is part IV of our series on Science and Democracy. See also: Part I, Part II, Part III.
TAGS: ACADEMIA, MARKETPLACE OF IDEAS, SCIENCE AND SOCIETY