Sunday, September 26, 2004

The Blue State of Science

I usually avoid politics in this weblog but I cannot totally ignore the US presidential election happening slightly more than a month from now. But nothing I will say would make much difference; nearly all computer scientists, and I expect most scientists, will vote for Kerry (or would vote for Kerry if they were US citizens). It's not based on a single issue. If we never went to war in Iraq, the economy were booming, Osama Bin Laden was behind bars and Bush supported a woman's right to choose, you would all still vote for Kerry.

What makes scientists so liberal? Why doesn't a field like computer science draw from a wide political spectrum? Does this liberal attitude stifle real political debate in scientific departments or even worse discourage those with more conservative views from entering our field?


  1. What makes scientists so liberal? Why doesn't a field like computer
    science draw from a wide political spectrum? Does this liberal
    attitude stifle real political debate in scientific departments or
    even worse discourage those with more conservative views from entering
    our field?>

    I believe that a statistical majority of conservative voters are left brain dominant. Neuropsychology of left brain/right brain shows up something quite interesting. The right brain is capable of spreading itself into both left and right brain capacities while the left can only refer to itself. Sciences, especially systems sciences attract left brain thinkers too. Even a logical encoded context of science builds in awareness of the duality aspect and spectrum of reality. Within this the left brain thinker is given a helping hand from the tendency towards reduced awareness from issues that are not local to his specific converging needs and desires.

  2. Oof... my left brain hurts! What does the above mean?

  3. I myself have found that computer scientists are either liberal or on the libertarian-end of the republican party. Needless to say, even in mostly-conservative San Diego, the Kerry bumper stickers in the department parking lot overwhelm any Bush/Cheney bumper stickers.

    Either way, I don't think conservatives are really discouraged. I know a couple of republican PhD students in the department who are mostly stealth about their views, but every now and then are willing for a debate on issues.

    Perhaps part of the answer you're looking for lies in the fact that faith is generally not associated with the Left. Scientists can recognize the place for religion in society, but they generally have no need for it in their research, nor do they see a need for faith-based public policy. Often in political struggles the right assails scientists, either for teaching evolution or for conducting research on stem cells.

  4. As noted, there's a strong current of libertarianism among computer scientists - but I suspect many of them don't have Michael Badnarik campaign stickers (or, indeed, support the Libertarian Party). This is also the kind of attitude that tries to treat politics as damage and route around it, so you don't usually note them as politically active unless it comes up in conversation. Going by stickers in the parking lot may be misleading.

    That being said, there are people with conservative, Republican attitudes in computer science. I don't actually know the political beliefs of most of my colleagues, however. I do know that it rarely comes up in the course of work, with some exceptions like electronic voting. So I'm not sure where you'd see "real political debate" -- are you talking about informal discussions between colleagues, or something else?

  5. If liberal=informed, then yes most scientists are liberal.

    Not many scientists bought Bush's story on WMDs with no evidence.

    Not many scientists thought that eliminating the police force and local government in Iraq after the takeover was a good idea. Bush's reason was that they were "Bathist." History tells us differently. WWII reconstruction in Japan went well, even though local government was still intact.

    Most computer scientists are aware that our trade can be practiced anywhere in the world. Outsourcing is a reality, an d Bush's immigration policies and watch lists hider bringing talent to the US.

    Most computer scientists are aware that health care costs are strangling our economy. Bush has shown no plan to bring them down, and instead is raising them by increasing government payment to providers. We need to strengthen competition within the health care industry or socalize it. The current oligopoly does no good for US citizens.

    Most computer scientists are appaled at $48 barrel oil prices. We need to increase wind, nucular, ethanol, biodisel, and coal power production. Bush has done very little.

    Most computer scientists hate to fly with the TSA frisking them at every airport. The simple additon of steel cockpit doors would have prevented 9/11. Bomb problems could easily be taken care of if passengers were asked to file through a lexan enclosure that gave off lots of static electricity.

    Most computer scientists are appaled at the Patriot act. Allowing the CIA/NSA to spy on US citizens does little to protect them from political forces. Search warrents without judicial review give the executive branch an unchecked power. Holding US citizens without charge and without legal representation goes against many founding priciples of our country.

    Most computer scientists are aware of the gigantic debt Bush has accrued. By no means is this fiscal conservitism.

    Most computer scientists are aware of the 40 year conflict in Gaza and the West Bank. Bush has done nothing to create a viable Palistine.

    I would label myself a conservative. I have a respect for human life from conception. As a scientist I know about evolution. Basic research on fetal stem cells would not be impeded by studying embryos from our fellow primates. Any tissue you can grow for a chimp should only take a few months of research to grow for a human.

    I think most computer scientists, even conservitives vote Democrat for one reason. Democrats fund the NSF, and the NSF gives us fat paychecks. I am a sellout and I admit it, but this election I am voting for Kerry because Bush is incompetent.

  6. "I think most computer scientists, even conservitives vote Democrat for one reason. Democrats fund the NSF, and the NSF gives us fat paychecks. I am a sellout and I admit it"

    Well said. Most people are sellouts, whether they admit it or not.

    Along the sellout line (and abstracted away from the specific case of Bush vs. Kerry), I can think of two additional explanations for why we see disproportionately few conservatives in academia.

    1. Conservatives tend to value trade and commerce more than pure research and feel more at home in the "real world".

    2. At the risk of mudslinging, conservatives seem to have a stronger desire to strike it rich -- something that is much more likely to happen in industry than in academia. I imagine the situation would be different if there were billionaire computer scientists.

    Lance, I understand that politics has a limited role in a computational complexity weblog, but I think many of your readers enjoy political posts.

    I particularly enjoy the scientists' perspective on politics. The audience for this weblog naturally lends itself to nice abstract discussions that (usually) don't get bogged down in specific details like Kerry's 8th stance on Iraq or how literate Bush may or may not be.

  7. The more interesting question is: why would a true Republican vote Republican? Interventionism, bloated government: the man who claims to stand for "principles" has betrayed his party's principles.

  8. I do not believe that people with
    conservative beliefs will stop becoming scientists
    because their views are contrary to the vast majority
    of scientists. This is especially so in natural
    sciences and engineering where the work itself is
    disassociated from the politics. One cannot say a
    similar thing about social sciences.

    I strongly disagree with a couple of comments that
    suggest that NSF funding is the reason that most
    scientists are liberal - I think this reasoning is
    unsound. To think clearly on this question we should ask: if
    both Republicans and Democrats give the same budget
    for NSF, would scientists be in equal numbers liberal
    and conservative. I don't believe that the numbers
    would be much different than what we have today.
    Also, many systems people get large amounts of
    money from DARPA, ONR, and Air Force grants and Republicans support these more.

    I think there are many reasons as to why most
    scientists are liberal. It would actually help
    us all if the few people amongst us who have
    conservative views articulated their views. I don't
    think that the divide is based on particulars - in
    fact many scientists might believe in fiscal responsibility, or the importance of family, for example, it is based more on a high level view.
    I think scientists are liberal because of a psychic
    leaning towards a utopian view of progress
    that animates their work. There is a desire to be
    left alone to do research and be funded for it even
    though its real world impact might not be clear.
    The conservative view point seems to go against this
    at a fundamental level and it is difficult to reconcile the two.

  9. First, a disclosure: I am voting for Kerry.

    To address anonymous: "why would a true Republican vote Republican?". I think this is a matter of trust. Those who vote Republican based on policy should be a lot more comfortable with Kerry's spending caps plan and fiscal responsibility. But I think many republicans have it in their head that the opposition party is the socialist party. Even after eight years of Clinton's great, pro-business and pro-people economic reforms, the republicans still like to act like they have a monopoly on good economic ideas. Even though reality shows that's not the case, the conception still remains.

    On the other side, some vote Republican because they are single-issue voters on one of the contentious issues that the Republican party happens to hold a position favorable to them.

    Another anonymous poster stated that "Most computer scientists are aware that health care costs are strangling our economy." I don't agree with that statement. If you look over the history, healthcare costs are rising, but we are also getting much more benefit from it. It used to be if your hip was bad, you would limp and suffer the rest of your life. Now, you can pay for a hip replacement, which by all means is a bargain. The costs are increasing, but it's a service that we want to consume. It would almost be like saying "the purchase of music is increasing so much it's crippling our economy"; the only difference is that we have a moral obligation to give care to the poor. (And a moral obligation not to encourage free-riders, but that's a very tricky problem; a similar problem was addressed wonderfully by Clinton when he increased the benefits of the Earned Income Tax Credit, a plan started by Republicans).

    "Most computer scientists are appaled at $48 barrel oil prices. We need to increase wind, nucular, ethanol, biodisel, and coal power production." Actually, I disagree. Oil is going to "peak" in the next decade, which shows that it's a finite resource (the remaining oil supply is measured in decades, not centuries). For all of those environmentally concerned, high oil prices are a good thing. In fact, I think we should tax oil further, to encourage conservation and the purchase of more fuel efficient vehicles (the Hummer is a negative externality, which every citizen of the world is subsidizing by living more polluted lives).

    Indeed, Bush has done very little with the coming energy problems. Kerry's plan is a step in the right direction.

    Another quote: "Most computer scientists are aware of the gigantic debt Bush has accrued. By no means is this fiscal conservitism." I disagree: Deficit spending in a recession is a perfectly legitimate plan. The problem is not the budget deficit, nor the rising debt; the problem is what Bush chose to do with the money. The "jobless recovery" was due to Bush's anti-job policies: The economy came back; but the economy is not the same as jobs. (This point has been elaborated in detail by economist Paul Krugman. Typically when I discuss this with conservatives they amazingly don't realize the difference between jobs and the GDP.)

    Finally, I respond in detail like this only because of the following whopper: "Democrats fund the NSF, and the NSF gives us fat paychecks. I am a sellout and I admit it". This statement is disingenuous. The lack of long-term research is a market failure best addressed by the government funding programs like the NSF. For example, String Theory could pay wonderful dividends in a hundred years; or it could be a dead end. It's too risky for business to undertake seriously (though the Big Guys like Microsoft do fund some researchers), but it has great societal value. Thus, it makes perfect economic sense to fund research. In fact, for the long-term health of the economy, we should increase research funding (even doubling it would create only a slight increase in the budget, because research spending is drowned by entitlement and defense spending).

    You do make one point, though: There's no such thing as a free lunch, except for when the government hands it to you. (Or, in the case of arbitrage, that's another "free lunch," but finding $100 bill on the ground is not the same thing as corruption and cronism.) On the whole, the peer reviewed system of the NSF appears to be working and we are getting our money's worth.

  10. Here is something I have been wondering about. Are science departments outside of North America also overwhelmingly liberal? In my own country, Latvia, one of our computer science professors is also one of the leaders of the Latvian pro-life movement (despite pro-life movement being far less popular among the general population than in USA). One of prominent Latvian nationalist leaders of late 1980s/early 1990s was a physics researcher prior to going into politics.

    Since I do not talk with my colleagues about politics that much, it is hard to judge what the majority views are. But I would guess there is a wide political spectrum among scientists in Latvia, without a clear-cut liberal or conservative majority.

    Any opinions from elsewhere in the world?


  11. I'm not so sure that if Bush achieved some kind of a perfect record (balanced budget, captured Bin Laden, etc) then everybody would vote for Kerry anyway. There are things in the realm of _imagination_ if not _probability_ that could lead me to switch my vote. People aren't that inflexible. But they are pretty inflexible, and will give their man a lot of slack while nitpicking any fault of the other guy. It's kind of like how one treats the faults of family members and the faults of outsiders.

    Why do scientists skew so liberal? Well, I'll throw some gasoline onto the fire: There is a sizable anti-intellectual presence in the conservative grass roots. Not all conservatives are this way by any means, and there is certainly a rich tradition of conservative intellectualism (and a good part of that at the University of Chicago). But it is significant and enough to send many scientists and intellectuals to the other side.

    However, the grass roots doesn't sit around reading Russell Kirk or Friedrich Hayek. At the grass roots level, it seems there are both industrial types who oppose scientific studies that lead to "the wrong conclusions" (eg. conclusions that suggest industrial practices are harmful and should change) and the Christian fundamentalist types railing against teaching evolution in schools (the former being more influential and the latter being more vocal). It doesn't strike one as a population where free inquiry is encouraged. These groups alone (even if they are minorities in the conservative camp) is enough to turn off many scientists.

    There are fundamental cultural issues at play. Science and introspection are not universally respected in the USA, and the GOP seems to attract the elements most hostile. I grew up in rural "red America", and there is a definite population there that views thinking too much as a bad thing. It is better to get a gut feeling and stick with it than to roll things around. Changing your mind is a sign of weakness. And scientists? Well, they're always changing their minds, and some even contradict the Bible.


  12. To answer Andris' questions, I think that in Italy most professors in the Sciences are to the left of the general population, and I think that this is true of most of western Europe.

    I suppose that in Eastern Europe, in the 1980s,
    being "progressive" meant being anti-communist, and
    the current leftist parties maintain a connection
    to the former communist parties, so it would
    make sense that "progressive" professors favor
    "right-wing" politics. I would suspect that in the
    future (and perhaps already now in Poland) most
    Eastern Europeans academicians will be "leftists"
    while their countries have "conservative" governments,
    in the western interpretation of the words in quotes.

    Replying to one of Lance's questions, I think that
    for sure political debate is stifled if one of the
    sides is extremely under-represented. However, I don't
    think that political debates are important to the
    mission of computer science departments, and I doubt
    that conservative theoreticians would feel unwelcome
    because of their ideas (of course I may be wrong).
    Many factors come into play in steering a person to work
    in a certain area, including charisma of senior researchers, one's own taste, funding opportunities, job market situtation, the technical questions one
    overhears from officemates, and so on, with
    politics being a very low order term.

    -- Luca

  13. I want to second Luca re Lance's last two questions - political debate is a very nice way to spend lunch (it is in fact a favorite Israeli pastime) but is not really important for us in our function as computer scientists. I seriously doubt that there is anyone that decided not to become a computer scientist because of their conservative political views. If there was such a person, I am not sure they would find the high tech industry to be that much more conservative than the university.


    p.s. re Lance's first question I have no idea why most
    computer scientists are liberal, but I wonder whether indeed they are that much more liberal than the pool from which they are drawn (i.e., CS majors with good grades from top universities).

  14. I agree with the point made by others that the liberal tendency is based more on broad based "cultural issues" than on disagreement with conservatives on specific issues.

    An example is the role of religion in formulating policy. I suspect there is a smaller proportion of the highly religious among computer scientists(and other scientists) than in the general population. Even those who are religious mostly don't believe that public policy should be conducted in a "faith-based" manner but instead it should be based on "secular" considerations.

    Bush is seen as being closely allied with groups like the Christian Coalition and Focus on the Family that believe differently and most computer scientists are probably not very comfortable with that way of thinking.


  15. Well, my experience is that the CS department (and other exact science ones) is one of the least left-leaning academic departments. Where I live (Israel), in most humanistic departments the political spectrum mostly ranges from extreme left to ultra-radical left, and the general (unscientific) feeling is that conformance on this is actively (though informally) enforced. I believe that the situation around the world is not that different (as measured for example by the relative success of an academic boycott on Israel in humanistic circles).

    So - is "why are CS scientists left-leaning" actually the correct question?

    - Eldar.

  16. I am a computer scientist and definitely *not* a liberal. I do not want government deciding how to spend my money when I believe I can do a better job. I do not think everyone is equal and despise socialism. Blah, blah...I think saying that computer scientist's are mostly liberal is just hilarious. Dream on. It looks great on paper!

  17. I think this is just a derivative of the fact that most of academia is more liberal than the general population. That view is promoted by conservatives who present it as a pseudo-populist anti-elitist argument against liberalism, but I think it's fairly accurate. The parallel argument that the media are particularly liberal is more debatable; the left argues just as convincingly that the corporate ownership of the media is inherently conservative. But reporters and editors certainly are more liberal than average.

    As a liberal, my answer to these conservatives has always been, "So you're saying that the majority of the smartest and the most knowledgeable people in the country are liberal; what does that imply about conservativism?" :-)

    I think that science is a creative endeavor, and that creativity implies a certain willingness to challenge orthodoxy. Liberalism is generally, although by no means always, a better fit for creative sorts.