What is "science"? According to Wikipedia: "a systematic enterprise that builds and organizes knowledge about the universe by virtue of testable explanations and predictions...Science is based on research, which is commonly conducted in academic and research institutions as well as in government agencies and companies." This simple statement includes both the history of its development, and the current range of procedures.
The historical summary in Wikipedia includes: "Isaac Newton and Gottfried Wilhelm Leibniz succeeded in developing a new physics, now referred to as classical mechanics, which could be confirmed by experiment and explained using mathematics (Newton (1687), Philosophiæ Naturalis Principia Mathematica). Leibniz also incorporated terms from Aristotelian physics, but now being used in a new non-teleological way, for example, "energy" and "potential" (modern versions of Aristotelian "energeia and potentia"). ...Where Aristotle had noted that objects have certain innate goals that can be actualized, objects were now regarded as devoid of innate goals... Leibniz assumed that different types of things all work according to the same general laws of nature," These historical observations capture two features of current practice: the use of specific terms for objects, and mathematics as a descriptive language to describe rules. During this time the pursuit shifted from exploration of natural events to practical applications. Much of thermodynamics comes from observations boring cannon barrels. WIKI: "In Bacon's words, "the real and legitimate goal of sciences is the endowment of human life with new inventions and riches", and he discouraged scientists from pursuing intangible philosophical or spiritual ideas, which he believed contributed little to human happiness..."
These are the two aspects of science: the methods of doing it, and its applications to the everyday world. The "scientific method" WIKI: "involves formulating hypotheses, via induction, based on such observations; experimental and measurement-based testing of deductions drawn from the hypotheses; and refinement (or elimination) of the hypotheses based on the experimental findings." This is vague and general because the procedures of inquiry vary from one field to another, but do usually start with a question which leads to an "hypothesis" or proposal from observations of the world. "Why does so and so....?" It must lead to answering the question by specific, repeatable observations. Scientists test hypotheses by conducting experiments or studies. And the outcome answers the question posed.
There is a general agreement that science method is 1)a statement about the natural or material world 2)evaluated by a procedure requiring observation of an event(s) related to the statement. Some examples may help:
In astronomy, this may be a prediction of the occurrence of an event, the location of Jupiter from a vantage point in October, related to statements about the motion of objects in the solar system of objects in view, which can only be observed at infrequent intervals. In astronomy, the events are observed but not controlled. By contrast, physics can usually create an experimental controlled procedure for repeating the event described in the statement to make observations to compare with the predictions. A simple example might be observing an object rolling on a surface and discovering the effect of friction. A complex example might be the acceleration of particles in a cyclotron which strike a target and deflect in predicted ways, (or do not). The complex apparatus of the cyclotron is engineered according to other physical procedures to produce defined effects (forces) on the particles. The ultra-nuclear particles can only be observed by indirect measuring techniques, subject to their own experimental methods. Many observations in early studies in chemistry resulted from combining substances and observing changes of color or state. But modern chemistry depends on highly complex reaction models and the identification of complex molecules based on response to electromagnetic radiation. In modern scientific inquiry, a complex chain of assumptions and observations underlies the outcome of the experiment, which are subject to variable influences. Even more complex and indirect procedures are needed for studies in biology, and with each level of complexity, the potential for error increases and measurements must include statistical techniques for separating the observations from random statistical noise introduced in the observation procedures. The idea of testing a statement about the world by techniques of observation does not change, but the techniques become increasingly complex, and the observations are the result of complex features of the experimental situation. This is not a theoretical statement but a practical statement about how modern science is done. If an effort is made to repeat the observation, the precise system of observation is required for the replication, and the replication may fail if the conditions are not met. Individuals who have never actively participated in scientific research beyond the controlled "labs" of high school or college courses do not appreciated this complexity, and assume that whatever is “discovered” by science is agreed upon by all scientists immediately or soon after, and disagreements only occur when someone is “wrong”.
The observations discovered in “science” are applied in the applications of engineering: the breakdown of crude oil into petroleum products in refineries, the effect of air flowing over curved surfaces to lift airplanes into the air, the creation and testing of pharmaceutical chemicals, etc. In engineering applications, allowances are usually made for a margin of safety, i.e. lift has added capacity for the unexpected dangers in flight. Crashing airplanes illustrate the reality that valid scientific principles can be applied in ways that are sometimes not reliably applied to achieve the predicted engineering outcomes. This does not disprove the scientific principles but shows that they can be incorrectly applied, or that other factors intervene.
The repeatability of observations in astronomy depends on the recurring events of various components in the solar system. But many events of material objects on earth are not reliably repeatable. An example are the algae tides that sometimes “bloom” on ocean shores. They recur often enough to observe the correlation with other variables, but not often enough to predict their next occurrence. The repetition of the geyser “old faithful” has repeated on schedule for centuries, but the timing of another geologic phenomenon, the eruption of Hawaii volcanoes cannot be predicted with accuracy. The inability to predict does not make science “fake”, but shows it's limitations. When the observations cannot be made in a controlled laboratory setting, determination of variables is more difficult; when events do not recur at regular intervals, prediction of their causes is more difficult.
An example is the assessment of global warming. Observations of the rising temperature of the earth are confirmed by
multiple sources, and “global warming deniers” are generally persons
who wish to deny the reality for personal economic reasons. Evidence from tree and plant growth, and geologic formations indicates that variations in the configuration and climate of the earth have varied over eons. The current changes in global temperature do not validate hypotheses about what variables contribute to these changes. Theories about the role of loss of the ozone layer, and green house gases trapping heat provide possible explanations, but because global warming is not been a recurring event, and because the vast increase in green house gases is not known to have occurred on earth in the past, these cannot be validated with assurance. Experiments have been designed which support the possible effects of these variables, but do not quantify their role over the earth. It is stupid to deny the evidence of global warming, but defensible to claim the evidence for green house gas effects may not be the entire basis of the events. May countries' scientists have concluded that this is likely to be a major cause and so there is strong advocacy for reducing the production of these substances. But this raises two problems: 1) how to generate the necessary energy for economic purposes without generating these gases, and 2) will the reductions slow or reverse the changes in climate or is the process too far progressed into some cycle to be reversed? Neither of these has a current scientific answer. Efforts to promote one or another opinion are influenced by fossil fuel companies on one hand, denying their role to prevent curtailing the use of these fuels, and on the other by environmentally sensitive scientists who believe that the data available is essential to saving the planet.
Nothing in this debate invalidates the accuracy of science as a method, but indicates that scientists are human beings who can be influenced to state opinions about ambiguous information biased by their source of support. As a group, scientists are disappointed by the loss of respect that has emerged in the general public, but this is the result of participating in political dialog, advocating for a position, based on limited data, and not acknowledging the advocacy position. When scientists behave like politicians they are perceived to be politicians, and science as an endeavor is degraded.
Other examples can be found in sciences involved with human life and behaviors. Is economics a science? Economists can measure aspects of human behavior, including the distribution of money as a variable in society. Traditional theories define the expected human behavior of a group of individuals “rationally”(logically) attempting to benefit themselves. But observation of real human behavior does not consistently fit these models, and a modifying theory of “behavioral economics” has developed and been given a Nobel prize for recognizing the perturbations that must be considered. Some of these effects can be predicted in advance, but others are only identified by the failure of humans to perform according to abstract models. The entire US economy has been modeled by computers, which give fair predictive accuracy for short term future values, unless a significant unexpected event influences the economic situation. As they say in investing “past performance cannot predict future results”. Does this mean that economics does not have predictive value? Sometimes the predictions are useful. The unknown factors cannot be added in advance, including certain human behaviors. Economists are called on to advise governments about changes in policy, and not surprisingly different ones provide advice that fits the expectations of different parties. This reveals the ambiguity in data that does not have repeatability to confirm predictive statements. Responsible economists would be clear about the limitations of their proposals, but here again, when scientists behave like politicians they are judged as politicians.
If scientists want to "rehabilitate" scientific activities as non-political events, they must be clear about the limitations of data in political situations, and avoid taking advocacy positions clearly supported and compensated by different advocacy groups or they must clearly state that their opinions are being supported by these groups. The requirement for this disclosure has recently been applied to medical research reports funded by commercial sources. With the current polarization of our society, it is possible that leaders may refuse to fund research that might not support their objectives, or unrealistically fund projects that do with little hope of success. So long as researchers are clear about the sources of funding and the observations made in research the scientific method is protected.
There are many ways of distorting scientific methods to create observations that are biased or limited. Because the conditions of modern science often require complex arrangements of observations, manipulating this situation may result in entirely different results. Here the scientific procedure is being carried out, but the influencing variables are being controlled in advance. For example, study of the effectiveness of a treatment on humans can be biased by how the patients and controls are selected. It is very difficult to detect and manage this kind of manipulation of scientific procedures, and results in "noisy" collections of studies pro- and con- a given determination which can only be sorted out by detailed analysis of the different procedures. This INTENTIONAL misuse of experimental procedures to manipulate the outcome of studies is a fraudulent use of scientific method but hard to suppress. It can only be managed by the combined efforts of the scientific community in the area of research being reported.
Advocacy in science will not go away. The current COVID-19 epidemic is a good example. Applying scientific results to practical situations is important and misuse is destructive and must be minimized. Advocating positions based on ambiguous data is sometimes necessary and acknowledging the personal advocacy will clarify the interpretation. Political leaders must make difficult decisions that impact the country, and they should be interested in the best scientific accuracy they can obtain to help make the decisions. Balancing the risks of infection, disability, and death, with the impact on the economy is a difficult problem. Other epidemics have occurred and data collected, but this infection has its own special features, so there are no recurring events. Balancing epidemiology and economic data on a real time basis to chart a course through the problem would be the rational approach. One that our country has been unable to accomplish through failures of leadership, despite the availability of the resources to do so. Attacking scientists in this situation reflects the political need to avoid blame for failures, but scientists must be careful not to overstate the validity of their advocacy.
No comments:
Post a Comment