A Perspective on Science 3 -- Development of science and the role of rigor

Status
Not open for further replies.
D

DrRocket

Guest
<p><span style="font-size:10pt;font-family:Arial">A Perspective on Science 3 -- Development of science and the role of rigor</span><span style="font-size:10pt;font-family:Arial">&nbsp;</span></p><p><span style="font-size:10pt;font-family:Arial">&ldquo;Truth&rdquo; in mathematics is nothing more than that which can be logically derived from the small list of accepted axioms.<span>&nbsp; </span>The axioms are supported only by intuition and are not and cannot be proved by experiment.<span>&nbsp; </span></span><span style="font-size:10pt;font-family:Arial">Euclid</span><span style="font-size:10pt;font-family:Arial">&rsquo;s axioms of geometry may be intuitively obvious, but it has been shown that other axioms can be used to obtain quite different and, on an objective basis, equally &ldquo;true&rdquo; results.<span>&nbsp; </span>There is no absolute truth in mathematics, but the level of rigor in tracing assertions back to the fundamental axioms is exquisite.</span></p><p><span style="font-size:10pt;font-family:Arial">Science seeks to explain natural phenomena, observed phenomena.<span>&nbsp; </span>Such phenomena do not depend for their existence on the validity of some set of assumed truths. Thus, science seeks a means of explaining and predicting absolute truths (for you quantum theorists out there this should be interpreted so that valid probabilities are absolute truths).<span>&nbsp; </span>Theories are accepted or rejected on the basis of agreement with carefully performed and validated experiments.<span>&nbsp; </span>Once a theory has been sufficiently tested and found to be accurate, it too becomes a yardstick by which new theories are tested.<span>&nbsp; </span>This is the correspondence principle.<span>&nbsp; </span>A new theory, seeking to replace an old well-tested and accepted theory must reduce to that old theory under circumstances in which the older theory is known to be highly accurate.<span>&nbsp; </span>Thus quantum mechanics reduces to Newtonian mechanics in the macroscopic world, and general relativity reduces to </span><span style="font-size:10pt;font-family:Arial">Newton</span><span style="font-size:10pt;font-family:Arial">&rsquo;s law of universal gravitation for moderate gravitational fields and moderate velocities.<span>&nbsp; </span>We demand this correspondence because of the close agreement of the older classical theories with a huge body of verified experimental data that applies under a known set of circumstances.</span></p><p><span style="font-size:10pt;font-family:Arial">To become an accepted physical theory, a proposed explanation for physical phenomena must agree with all observed data, within the bounds of potential error in that data, and comply with the correspondence principle.<span>&nbsp; </span>In addition the body of experimental data used to test it must be sufficient to provide confidence in the validity of the theory.<span>&nbsp; </span>Supporting experiments must be duplicated by independent investigators when possible.<span>&nbsp; </span>The data and experimental technique are subject to stringent peer review and comment.<span>&nbsp; </span>Often a variety of experimental techniques are applied.<span>&nbsp; </span>If even one experiment finds a discrepancy and the data of that experiment are found to be correct, then the theory must be rejected or at least viewed only as a n approximation applicable to circumstances in which it is supported by a large body of evidence.</span></p><p><span style="font-size:10pt;font-family:Arial">In those branches of science that involve complex systems, the role of theory and observation can become a bit murky.<span>&nbsp; </span>Often the systems involved are sufficiently complex that we do not have the capability to apply rigorous first physical principles effectively to produce meaningful predictions.<span>&nbsp; </span>We have not yet passed fully from &ldquo;stamp collecting&rdquo; to physics.<span>&nbsp; </span>In such branches of science there inevitably will arise differences of opinion as to the principles at work, the relative importance of those principles and hence on the validity of mathematical models used for prediction.<span>&nbsp; </span>The rigorous methods used in the development of physics cannot be effectively applied without at temporarily stopping the production of findings for use by the world at large.<span>&nbsp; </span>It is not socially acceptable to do so.</span></p><p><span style="font-size:10pt;font-family:Arial">When these complex problem demand that they be addressed, we are forced to rely on hypotheses, supporting evidence and debate.<span>&nbsp; </span>A participant in the debate or even a user of the potential outcome must approach the problem with a critical frame of mind and ask probing questions.<span>&nbsp; </span>Relying on &ldquo;expert&rdquo; opinion without critical thinking is a dangerous tactic.<span>&nbsp; </span>Questions to be asked and satisfactorily answered before conclusions are reached based on some model include the following:</span><span style="font-size:10pt;font-family:Arial">&nbsp;</span></p><p><span style="font-size:10pt;font-family:Arial"><span>1)<span style="font:7pt'TimesNewRoman'">&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">What physical processes are involved and how are they modeled?</span></p><p><span style="font-size:10pt;font-family:Arial"><span>2)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">What data is used to provide inputs and boundary conditions for the model?</span></p><p><span style="font-size:10pt;font-family:Arial"><span>3)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">How is the data measured and what errors may apply?</span></p><p><span style="font-size:10pt;font-family:Arial"><span>4)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">Do the physical processes involved and the data used correspond to one another?<span>&nbsp; </span>I.e. is the data used reflective of the processes involved and is it appropriate to the question at hand.</span></p><p><span style="font-size:10pt;font-family:Arial"><span>5)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">Does the model contain any adjustable constants; i.e. parameters that are adjusted to make the prediction of the model fit observation and that are not directly determined from first physical principles?<span>&nbsp; </span>If so, how are those constants determined and what is the sensitivity of the model to errors in them?</span></p><p><span style="font-size:10pt;font-family:Arial"><span>6)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">Has the model been evaluated for sensitivity to variations in the measured data, and what sensitivity has been found?</span></p><p><span style="font-size:10pt;font-family:Arial"><span>7)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">What competing models exist, and how do the predictions of the available models compare?</span></p><p><span style="font-size:10pt;font-family:Arial"><span>8)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">What uncertainties exist in the approximations used to describe the physical processes?</span></p><p><span style="font-size:10pt;font-family:Arial"><span>9)<span style="font:7pt'TimesNewRoman'">&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; </span></span></span><span style="font-size:10pt;font-family:Arial">What known conditions have been used to validate the predictions of the model against known results?<span>&nbsp;</span></span></p><p><span style="font-size:10pt;font-family:Arial">Note that none of these questions ask who or how many people believe the results of the model.<span>&nbsp;&nbsp; </span>Acceptable answers to these questions are simply direct answers to direct questions.<span>&nbsp; </span>Appeals to &ldquo;authority&rdquo;, repetition of unsubstantiated opinion, or evasions made at a high decibel level are not acceptable responses.<span>&nbsp; </span>Nobel laureates can be and have been wrong.<span>&nbsp; </span>Bohr and Einstein disagreed about quantum mechanics &ndash; they could not both be right.<span>&nbsp; </span>Pauling missed the helical structure of </span><span style="font-size:10pt;font-family:Arial">DNA</span><span style="font-size:10pt;font-family:Arial">.<span>&nbsp; </span>I won&rsquo;t even comment on the idea of awarding the Peace Prize to terrorists.</span><span style="font-size:10pt;font-family:Arial">&nbsp;</span></p><p><span style="font-size:10pt;font-family:Arial">Scientific rigor can separate good ideas from bad ones.<span>&nbsp; </span>The application of that rigor is, or should be, an objective process done without recourse to personalities or testimony.<span>&nbsp; </span>What matters is application of known, verified physical principles, quantified approximations, and detailed error analysis. <span>&nbsp;&nbsp;</span>In God we trust &ndash; all others bring data.</span></p><p><span style="font-size:10pt;font-family:Arial">A final postscript.<span>&nbsp; </span>This process of applying scientific rigor in the determination of the acceptance of new physical theories clearly requires an open mind.<span>&nbsp; </span>In fact open-mindedness is a characteristic of a good researcher.<span>&nbsp; </span>Why then, do those on the lunatic fringe find themselves often dismissed out of hand?<span>&nbsp; </span>The answer is that, as Feynman said when asked why he didn&rsquo;t work on anti-gravity, that the scientist must work with an open mind and creatively, but while wearing a straightjacket.<span>&nbsp; </span>The straightjacket is the recognition that new ideas must comply with the correspondence principle.<span>&nbsp; </span>A new idea that clearly contradicts known principles, in circumstances in which those principles have been shown to be applicable and accurate, must be rejected.<span>&nbsp; </span>There are too many worthy ideas to be evaluated to waste time on crackpot notions that have no chance of surviving the test of rigorous experiment.<span>&nbsp; </span>Unfortunately the supply of crackpots putting forth untenable ideas seems to be inexhaustible and the serious scientist simply has not enough time to politely debunk each item in a torrent of gibberish.<span>&nbsp; </span>To you category 3 folks &ndash; sorry to bruise your feelings, but that is the way it is.</span></p> <div class="Discussion_UserSignature"> </div>
 
Status
Not open for further replies.

Latest posts