Science Integrity

A Column on Scientific Integrity - Dr. Lenny Howdt - ONRRI scientist www.howdt.com

Wendy McElroy has a good article at LRC on questioning scientific authority. She gives a series of questions to ask to see if the science is on the level. This is a good first step for lay people, but if she thinks that scientists can catch faked data in the peer review process, she misses the boat. When asking for proof through evidence, she really misses the boat, because as a scientist, I can design a good experiment that would produce great data, get a rock solid conclusions and be completely wrong. It comes from the fact that scientists are exploring unknown phenomena and making things up to see if they fit observation. We propose a theory and then attack it.

With the advent of applied engineering, science now is product driven. The entire medical research staff is pushing drug discovery to line their pocketbooks, not specifically to make any improvement in the standard of general living. The assumption (not yet sufficiently validated) is that that standard will trickle up. There is some base information in our set of assumptions that is wrong - that doesn't explain how something works that we know but don't really know. We have to validate the entire set of science theory again, to make certain that we are pursuing the right goal - an explanation and understanding of how things work.

Modeling is our current best tool. We collect enough data from various sources to paint a picture, a mosaic of things that are relevant until we see a correllation where things fit. Then we check the data against the reality and again ask to see if it fits. But what if we bias out sample collection - how can the data be right, if the sample has a contaminant? Depends on what you are looking for.

The two key events in the 1990s that broke the back of the historic science process were 1) the Pons/Fleishman cold fusion fiasco and 2) the change that allowed universities to collude with corporations on patented research. The former stopped a very promising energy field very dead in its tracks because of the practitioners method of information distribution - the mainstream media. The problem seemed to me to be that the vested interests were tying up the paper politically in the peer review process by insisting on changes than P&F refused to make. Knowing they were on to something, they release the info to the press without considering the backlash of the very offended (and very wrong) science community. Attack dogs can tear anything apart - but look at the status of current theories on cold fusion.

The second issue meant that there was no well funded public interest watchdog looking at science. The few people with means and integrity to start speaking out were quickly ostrasized from the community. While scientists seem to be enamored with the peer review system - this absolutely necessary double check has been terribly corrupted by personal pettiness. There is no doubt that many scientists hate sending their stuff out for review before publication because information in rejected papers always seems to come out in the accepted papers of reviewers, later in time. Since the reviews are conducted by the editors top picks of people in the field, there is a total vested interest in publishing only agenda supportive materials and burying any information that is not consistent with the science philosophy currently being pushed.

The only way to get a straight answer from a scientist is to insist that he explain the concepts in layman terms. Unless you understand what she is talking about - keep asking questions. People who know what they know will spend all the time in the world getting the message across in terms that makes sense to each person interested enough to ask. The depth level shifting really helps the scientist cement the ideas.

Scientists are supposed to be the greatest critics of their own research - you have to know you are correct before you publish, if you wish to inform properly. But that's the problem - science is not to inform - it is to make money by selling things to people as later, greater or better. If you take the sales out of the laboratory, you will take the wind out of the sails of most of the untrue science effort.

The best way to understand science, is to listen, write your questions down, and ask a different scientist about them. Without giving away context, you can ask any question of any true scientist and get a real answer about how that scientist sees the world works. But all scientists do not agree - in fact science, like life, is not a zero sum game. Things change.

Our quality control data approach exerts too much control - it forces data and rejects valid data on the basis of arbitrary criteria. If you observe something, then you tend to believe it. If you observe it twice, you are pretty certain and the third time is a charm. But if you have an expectation of the result and you set up to verify that expectation, any test that doesn't quite give the same result can be invalidated methodically until the results converge to support the expected answer. So science gets jobbed, people get bad information, bury it in the knowledge base and wait for some other scientist to study the problem again and invalidate the data.

But - big caveat - but science has become so expensive at the cutting edge and so competitive that nobody really can repeat anybody else's experiments properly any longer. The total transparency of science, that required simplicity enough that anyone anywhere setting up the same test could get the same result (Newton's apple) is now gone to details of submicroscopic theory and very low or very high energy measurement to support speculation based on speculation. Time to go back and check premises. And do calculations. And develop theories that can be supported by experimental evidence. That anybody can verify readily at any time.

Oh - can that be done? Can i buy a chemical anywhere to reproduce a chemistry experiment. Well - as a chemist, i can buy chemicals - because i have a track record and an official entity (employer) that has jumped through the proper government hoops to allow me to do my job. Of course, I am self-employed and knew which hoops to jump through - i doubt an average business person could have made the connections of the proper dots. Plus the three letters that allow doctor lenny to be called doctor help. But these letters can be bought as well as earned.

The only way to tell if science is valid is through due diligence. If you don't wish to learn science, that's okay - just know that the method used most is called trial and error. But find someone you trust who knows science and is willing to answer your specific questions to the best of their ability, and is willing to say i don't know, when they really don't know. If she says, let's find out, then you know you have a good friend that knows science. After all - if you are reading this, you found howdt for yourself.