My feeling is that autism is actually many different things with similar manifestations. The source of my son’s autism may be completely different from the source of your son’s condition. I do believe, however, that most cases of autism are primarily genetic in origin. That’s just my opinion, pulled out of thin air.

I can see where you’re coming from, but I don’t think this is a useful way of comparing the two disciplines. Some issues are better addressed by philosophy, some are better addressed by science, with some overlap. The “restrictions” imposed by science are exactly what lend credibility to what would otherwise be statements of opinion. That credibility leads to trust, and trust leads to applications that attempt to improve our lives.

In constrast, philosopy bears no burden of proof or evidence. Most philosophical assertions have some link to observed evidence, but the link may be tenuous, and the field has no constraint that there be any link at all. This gives philosophy an advantage over science in some areas, but a disadvantage in others.

Take the brakes on my car as an example. I won’t buy a car whose brakes meet only philosophical standards of performance but haven’t been through any rigorous verification. I want a car whose brakes have been shown by scientific standards of testing to be highly likely to stop the car when applied.

Calling science a restricted version of philosophy downplays the power of science, power which is not found in unrestricted philosophy. Unrestricted philosophy has its own set of strengths and weaknesses, its own vital-but-different power. Philosophy is different, not superior.

To relate this back to autism, we need to back up to the steps of the scientific method:
Observe something
Make a guess about the meaning or cause of what you observed
Test your guess

The test is yet another observation, so usually this turns into a cycle where the test leads to another guess and another test.

The overlap between philosophy and science is primarily in the second step, interpreting what you’ve observed. There are often several explanations for an observed phenomenon, and even the observation itself can be questioned. In many cases the scientific community attempts to come to consensus (yep, consensus!) about the implications of an observation. Sometimes the consensus is right, sometimes it is wrong. Sometimes the consensus changes over time.

Does this mean that “all science is consensus science”? No! The type of consensus science Mr. Crichton is referring to is different. This dangerous strain of consensus science seems to believe that truth is whatever some influential group of people declares it to be, regardless of evidence. It is not based on observation. It is not verified with tests. It is conjecture. It is philosophy. It is not a good basis for a reliable set of brakes. Regardless of where Mr. Crichton may stand on any particular issue, he is spot on when he points out that we often pretend to know more than we actually do, and that misplaced trust in such “knowledge” can cause us to follow dead end paths as if they were the road to salvation.

This is so relevant to autism because of the prevalence of pseudo-science used to support various treatments. There must be well-defined criteria to discriminate between treatments which have been validated, invalidated, and yet-to-be-validated. Otherwise just about all treatments are claimed to be supported by science, including quack treatments. All treatments get some measure of the same trust you give the brakes that keep you safe in your car, just because some “expert” made a false claim to be supported by science when in reality the treatment only meets the loose demands of philosophy.

As you know, immense quantities of monetary and emotional resources are expended on autism treatment. In my opinion we can’t afford to allocate those resources based on guesses and warm fuzzies. We need science-based guidance.

Still, I acknowledge that science alone isn’t the complete answer. As I stated in my original post, “pure science and pure objectivity is neither possible nor desireable.” Some things can’t be tested for ethical or logistical reasons. Science tends to ignore or even reject some theories just because of the difficulties of formulating a test. And as John pointed out, humans are pretty complicated systems. Science may never hold all the answers, but I firmly believe science must play a role to at least keep us honest. It is science that provides a framework to enable us to do more than wander aimlessly.

Many of my relatives are graduates (primarily in the sciences) of Berkeley, one in particular being my civil engineer, bridge inspector grandfather.

A couple of thoughts regarding the remaining comments:

Michael Crichton has a medical degree, rather than a research one, and has not been active in actual research for decades, if he ever was. I’d be hesitant to view him as an authority on science, especially since he’s clearly wrong on a significant point: all science is consensus science. You can still find trained scientists who doubt the evidence that supports evolution or the link between HIV and AIDS. Their doubts can be phrased in scientific sounding terms (”insufficient evidence”, “lack of proof”); it is only the consensus of the scientific community, as well as the experts in the appropriate fields, that allow their critiques to be dismissed as attempting to hold science to unreasonable standards.

Having put a great deal of effort into understanding the science behind climate change, i personally find many of his arguments scientifically vacuous.

Separately, the standards for hypothesis testing vary from field to field. Autism is a human health issue, and that is perhaps the most challenging, because it is unethical to do many of the properly controlled experiments that are possible in model systems. Behavioral changes are doubly challenging, since no model system approaches the complexity displayed by humans. You can delete any gene you want in mice, but you can never ask a mouse how it feels.

As for science vs. philosophy: science is a specialized subset of philosophical inquiry; to an extent, you accept a restricted methodology in order to attain objective verification and accurate predictions. Designing tests that are empirically accessible and informative in terms of building/testing models is extremely challenging. Having taken graduate level courses in both science and philosophy, i would certainly not say that the best minds go into either; very sharp minds go into both, but get trained to do very different things.

As a graduate of Berkeley, i’d say they’d be doing very well based on their achievements in the sciences alone, including the Nobel Prize just awarded to George Smoot for his achievements in cosmology.

As a rule I am neither impressed with the cognition of most scientists nor with there gnosis and praxis either.

The best minds go into philosophy, because the very scientific method devolves from philosophically derived paradigms.

And where would UCal Berkeley be if it weren’t for the Bishop?

What we accept as natural law is (or at least should be) based on evidence derived from controlled, repeatable experiments. This is the scientific method:

scientific method
a method of research in which a problem is identified, relevant data are gathered, a hypothesis is formulated from these data, and the hypothesis is empirically tested.

The law of gravity, for example, is based first and foremost on the evidence supplied by dropping something. It falls down. This is an exercise which anyone can duplicate, one which provides evidence to support the hypothesis that the earth pulls stuff toward it.

Autism research should follow the same principle. There is a book which supports exactly this position, The Science and Fiction of Autism by Dr. Laura Schreibman. Her assertion is that the scientific method needs to be the foundation of all conclusions regarding the causes and treatments of autism. You should be able to back up any assertion with evidence. Not necessarily proof, but evidence. Don’t throw evidence away simply because it doesn’t support what you want the truth to be, and don’t embrace a theory without testing it out.

Dr. Schreibman’s book provides some history of how people thought of autism, its causes, and its treatments. That history makes a great case for why evidence matters. There was no evidence for the “frigid mother” theory of why some kids developed autism, but the medical community accepted it for years, adding guilt to already-burdened parents. Facilitated Communication is another example, where people actually went to jail based on accusations autistic people supposedly communicated via FC. It takes a carefully crafted study to determine whether FC is for real to determine whether those accusations are also real.

An unintended message from Dr. Schriebman’s book is that pure science and pure objectivity is neither possible nor desireable. Certain ideas seem to fall in and out of favor in the medical community with no real evidence. Some ideas are not fully investigated because it’s hard to devise an adequate science-based experiment. In spite of the ideals of a science-based methodology, which I whole-heartedly embrace, reality isn’t always conducive to the scientific method.

Michael Crichton has published a number of his speeches about hard science versus consensus science. Check out http://www.crichton-official.com/speeches/index.html. My favorite is Aliens Cause Global Warming. It’s extremely relevant to autism research, and particularly relevant to public policy regarding autism. Here’s a quote from that speech that everyone should memorize:

There is no such thing as consensus science. If it’s consensus, it isn’t science. If it’s science, it isn’t consensus. Period.