Tuesday, February 16, 2010

The Invisible Beaker of the Laboratory

I, as usual, have three dozen problems with this article, but I'll narrow my objections down to one item. In the article, the author states:

"Yet in a wider sense of the term, the decisions of a scientist choosing a problem and pursuing it to the exclusion of other possible avenues of inquiry may be said to have an economic character. For his decisions are designed to produce the highest possible result by the use of a limited stock of intellectual and material resources. The scientist fulfils this purpose by choosing a problem that is neither too hard nor too easy for him. For to apply himself to a problem that does not tax his faculties to the full is to waste some of his faculties; while to attack a problem that is too hard for him would waste his faculties altogether. The psychologist K. Lewin has observed that one's person never becomes fully involved either in a problem that is much too hard, nor in one that is much too easy. The line the scientist must choose turns out, therefore, to be that of greatest ego-involvement; it is the line of greatest excitement, sustaining the most intense attention and effort of thought. The choice will be conditioned to some extent by the resources available to the scientist in terms of materials and assistants, but he will be ill-advised to choose his, problem with a view to guaranteeing that none of these resource be wasted. He should not hesitate to incur such a loss, if it leads him to deeper and more important problems."

The author earlier used the example of a giant puzzle that all scientists were trying to solve. The method the author claimed that science used was as follows: each scientist following their comparative advantage in particular puzzle piece preparation and permutation. But science isn't like that!

We don't know if the universe even has a solvable puzzle. Physicists are currently searching for a 'theory of everything'. This ToE, if it exists, will combine quantum mechanics and general relativity into one giant 'solved puzzle'. This understanding will enable us to peer into the early nanoseconds of the universe and understand why things are the way they are. But it is possible that there isn't a theory of everything. It is possible that we won't be able to find all of the pieces because there isn't a complete puzzle that exists. It's possible that the best humanity can do is to get a practical, workable, fuzzy, general understanding of the nature of things. This might be enough for our continued survival as a species.

The author acts like each individual scientist has the ability to choose from a number of unsolved scientific problems, and that the scientist will know in advance which might be too easy or too hard. The scientist will then choose the goldilocks problem, the one that is just right. BOGUS. It implies that a scientist knows the level of difficulty of each problem before he or she begins to solve it. It also implies that scientific problems can be broken into solvable chunks like puzzle pieces.

You know what you call a problem that is too easy? You call that problem solved. Too easy... BAH. The only reason that the problem exists is that someone hasn't figured it out yet. It's possible that no one has tried. A scientist won't look at it and think, I can solve this one in two weeks, so I won't even try. Every time a scientist discovers something it's an opportunity to publish his or her findings in a peer reviewed journal, and publishing is a major motivation for the scientist to figure things out. If anything an ambitious scientist will be motivated to find the 'too easiest' problems out there and solve them and publish the results.

Also not every puzzle piece is available to every scientist. The fields of specialization are so narrow that at the frontiers of scientific discovery there may be only a few people qualified to even know where the particular pieces are. In those cases those small groups of specialized scientists are actually working in 'isolation' like the author mentioned in the beginning of the essay. Anyway, I can't wait for our meeting this Thursday.


  1. I'm not sure I think the puzzle metaphor is as completely worthless as you seem to.

    1) I'm sure Polanyi recognizes that "the puzzle" is likely unfinishable, but that doesn't negate the point about cooperating to attempt and assemble the available pieces. The puzzle metaphor helps illustrate the importance of collaboration and interdisciplinary studies.

    2) Scientists face incentives just like everyone else.

    a) If a problem is outside of their area of specialty (too hard) they will likely not pursue it. To attempt to do so would require years of costly re-training. Scientists have an incentive to perform research within their disciplines as determined by their comparative advantage. Scientists do tend to approach their area of study selectively according to specialization (i.e through the division of labor).

    b) scientists may not always be aware of exactly how long their research might take but they do have a general idea about the depth of a research problem and the amount of work involved in testing their hypotheses.

    c) If a scientist anticipates that a problem is too simple there can still be a variety of reasons they choose not to pursue it. For one if the problem was as easy as it seemed the scientist would assume one or more competing researchers were already working on the problem at hand and nearly ready to publish their results. To attempt to pursue the simple problem as well would be a waste of time if the researcher was certain another was also on the verge of publishing. Economists call this the "No Cash on the Table Principle".


    d) Also if the problem was simple enough the scientist might simply choose to defer it to one of their starving graduate students for a quick research project so the fully fledged scientist can devote their efforts to more advanced and prestigious problems.

  2. While you manufacture certainty about your feelings as to the usefulness of the puzzle metaphor, let me make it more useful by adding the following:

    It's a group of giant puzzles that we are all trying to solve at once. It might be possible that it's just one super giant puzzle, but at the moment it seems unlikely. So far, in the Indiana Jones warehouse where we are keeping the completed puzzles we have about a half dozen groups of connected pieces that may or may not join together sometime in the future. Adding pieces to one completed bit doesn't really help the guys working on the other bits.

    We have an unlimited number of puzzle pieces in the warehouse with us, but in order to handle any given piece, and examine whether or not it fits in our completed section we have to spend anywhere from a few tens of thousands of dollars to millions of dollars. There are a few groups that are willing to loan or give us the necessary funds to play with our puzzles, but we are in competition with all other scientists as to how much of the puzzle fund we have access to.

    Some of the already completed sections of the puzzle have cute pictures of puppies and babies, and the jerks that control the puzzle money seem to favor completing those sections.

    Other sections of the puzzle look like treasure maps, and private companies occasionally chip in to the puzzle fund to see if they can find exactly what spot 'x' marks.

    Some partially completed sections of the puzzle don't look like too much of anything at all, but the pieces themselves are intriguing and they capture the attention of some of the scientists. These sections don't get much money from the puzzle fund, but these sections have what appear to be the borders of the bigger puzzle.

    Often times what determines what sections of the puzzle get the most effort is the charisma of the people working the puzzle, and whether they can sweet-talk the keepers of the puzzle fund out of some money.

    One more thing, the people who hold the purse strings to the puzzle fund, are usually color blind, totally blind, and often brain damaged as well. They couldn't tell the difference between a puzzle and a pizza.