The word science is bandied about in many ways. Many of the great achievements of our society are attributed to science, as Jawaharlal Nehru (The first prime minister of India upon its independence from Britain) is reported to have said:
Engineers are prone to attributing much of their success to a practice of science. Thus the Structural Engineer, the official journal of the British Institution of Structural Engineers, defines its subject as Structural engineering is the science and art of designing and making, with economy and elegance, buildings, bridges, frameworks and similar structures so they can safely resist the forces to which they may be subjected. (2).
This attitude is often evident in speeches and statements in engineering magazines by presidents of engineering societies which are often intended to boost morale among the members of their profession. In these utterances they can get into extravagant statements, of which the following is an early example.
We are the priests of material development, of the work which enables other men to enjoy the fruits of the great sources of power in Nature, and of the power of mind over matter. We are the priests of the new epoch, without superstitions (3).
Engineers are not likely to call themselves priests today or bother to claim lack of superstition, but otherwise this kind of sentiment can still be heard, in various forms, from some engineers today. In the following, a brief look will be taken at what science really is, as discussed and described by two prominent scientists and a historians and philosophers who have examined science from the outside.
What is meant by the word "science"? Here we will look at an answer given by John Ziman, a physical scientist, in his book Public Knowledge: The Social Dimension of Science (4) He reviews four popular alternative definitions:
He thinks the first definition is the most common one, the "vulgar definition", as he puts it. The "mastery" of the world is also a notion popular in engineering. This definition identifies science with its products, such as penicillin or artificial satellites. However, Ziman objects to it because it confuses science and technology. It emphasises the application of scientific knowledge, not the processes by which that knowledge is arrived at. This definition does not distinguish between science and magic, and it gives us no reason to study cosmology, the theory of the origin of the universe, for how can that lead to mastery of the environment? This definition confuses ideas with things. Penicillin is not science, any more than a church or mosque is religion.
The second definition, the study of the material world, is also popular and familiar. There is some truth in it, but it is restrictive. Does it include psychology and sociology, where the subject is humanity and human beings, not the material world? Does it include pure mathematics (It may be significant in this connection that departments of psychology, sociology, and mathematics, may be found in Faculties of Arts in some universities, or may belong to both a Faculty of Arts and a Faculty of Science.)? Perhaps mathematics should be considered a discipline separate from science, but that is not the usual practice. In fact there are many branches of science devoted to theoretical abstractions of the real world; are they not science?
It was a big step forward when the importance of experiment was recognized, as in the third definition. Much earlier science was based solely on speculation on topics where, in fact, the ideas in question could have been tested by experiment. Modern science indeed owes much to the development of the experimental method of testing hypotheses. However, there are branches of science where experiment is impossible. Again mathematics is an example. Others are astronomy and much of geology. The role and place of experiments and the way they are really planned and conducted are indeed topic which have evoked much interest among historians and philosophers of science.
In considering science in relation to engineering, we should consider the particular way experiments are conducted with engineering work in mind. If we, as engineers, want to test a hypothesis by experiment we are likely to confine our tests to the range of variables of interest to us at the time for the practical application we have in mind. If we find that the hypothesis is obeyed reasonably well in those circumstances we go ahead and use the hypothesis. By doing so we may not be aware of other circumstances in which the hypothesis fails.
The fourth definition of science has been favoured by many philosophers, perhaps because it is similar to the logic cultivated as a sub-discipline of philosophy. The main problem, perhaps, is that much of science does not happen like this, even though there is a tendency for scientists to portray their work this way, after the event. James Watson (5) has provided an amusing account of an example of a major scientific advance, the discovery of the double helix structure of DNA, which bears little or no resemblance to a logical process.
Watson and his collaborator, Francis Crick, were junior research workers in Cambridge. Watson was effectively a post-doctoral fellow and Crick a graduate student. Both became interested in the structure of DNA, then known to be key element in biological material and under study in a number of laboratories, including one at King's College in London. The King's College team, consisted principally of Maurice Wilkins and Rosamund Franklin. Franklin was working under Wilkins and they were devoting most of their effort to applying X-ray crystallography to DNA and related molecules. At that time the evidence obtained from this technique was in the form of film showing a pattern of dots, lines and arcs, produced by the scattering of a beam of X-rays from crystalline structure. The trick was to interpret these pictures, i.e. come up with an arrangement of the atoms in the crystal which would produce the pattern seen, bearing in mind the known connections between the atoms concerned. Chemical analysis tells you something about the groups in which the atoms occur, but not how the groups are arranged relative to each other in the crystal. Simpler crystal structures, like metals and inorganic salts, had long before this been studied by X-rays and explained this way. Indeed the boss of the Cambridge laboratory, Sir Lawrence Bragg, and his team had pioneered much of this work. The interpretation of biological structures was a much greater challenge and the London group focussed on DNA, while the experimental work at Cambridge was on other substances.
Watson and Crick did not do any experiments on DNA. They simply put their minds to thinking about the possible structure and building paper and metal models of their ideas. In short they played with ideas. The evidence they tried to fit their ideas consisted of the experimental results from London, which they collected more or less surreptitiously. These had mostly been produced by Franklin. They also had the known chemistry of the molecule to take into account. Crick knew more about this than Watson.
If we are to believe Watson's account, their discussions on the structure were between games of tennis, movie shows, dinners and parties, plus a couple of visits to the London laboratories. The idea that the structure was basically helical was not new. Linus Pauling, a famous Nobel Prize winner, had found that another protein had a helical structure called the alpha-helix, and was working on the same lines for DNA. His son was in the Cambridge group, so they had some idea of Pauling's progress. It seems, however, that the London group was not so sure that DNA was also helical.
The breakthrough for Watson and Crick came when they hit upon the idea that DNA consisted not of one molecule but of two parallel long molecules in a helical arrangement, joined together so that certain groups paired off between the chains always in the same way. Their ideas were confirmed when they built a scale model, with the groups joined as they should according to chemical knowledge and they found that the structure would produce the X-ray patterns observed in London. For this solution they shared the Nobel prize with Maurice Wilkins. Tragically, Rosalind Franklin had died before the Nobel prize was announced.
The discovery of the DNA structure certainly does not fit the notion that science arrives at truth by logical inference from observations. It is not unusual that Watson and Crick did no experiments in this area, many scientists devote themselves solely to theoretical speculation on the basis of other scientists observations. What we can see in this case is that different groups of experienced scientists drew differing conclusions from the same experimental evidence. The process was mostly guess work and there was no obvious single path from the relevant observations to the correct model, that is to the model which the scientific community has accepted.
Ziman's answer to the question: "What is Science?" is that Science is Public Knowledge. In this he emphasizes the social aspects of scientific activity, as an enterprise conducted by scientists interacting with each other as a subset of society and within the context of the ideas and assumptions of their contemporaries.
The goal of science is not just to come up with observations and theories, but do to reach a consensus of rational opinion on the matters addressed. Most people, he says, may think of the experimental work and the theorizing in private as the essence of science, with the public discussion an auxiliary activity. To Ziman it is the other way round. The discussion between colleagues, the social interaction, is the key to science, the essential part of science. In many instances it is through discussion that the ideas appear that are subsequently developed into theory and tested through experiment.
Scientists often reach conclusions on matters which are important in science on very little evidence. This is true of the DNA structure. It is true of many so-called laws of nature. Once the hypothesis has been published, it may be that others will do experiments, apparently setting out to confirm the reported observations and hypothesis, but if the truth be known, most who do so are going to happier if they find that their observations do not fit the hypothesis they are testing, because then they have the opportunity to come up with a different hypothesis and get some glory from that. It is therefore not surprising that some recent great breakthroughs, such as the supposed evidence of cold fusion of hydrogen, have not been substantiated to a generally accepted degree by other scientists. Sofar the evidence for this phenomenon has been limited and contradictory. It is also a problem for many of those concerned was that the announcement was made to the lay media rather than through normal channels within science, i.e. the announcement was not made initially within the social institution which is science.
Other ideas on Science
Numerous scientists, historians and philosophers have tried to decide what science is and what the scientific method is. It would be impossible for me to deal adequately with this, other than to mention a couple of well-known names in the debate. Karl Popper was a Professor of Logic and Scientific Method at the London School of Economic and Political Science. He is associated with the idea that scientific laws and hypotheses are never proved by experiment, no matter how many experiments seem to fit the law and hypothesis. Laws and hypotheses are disproved by just one experiment which does not fit the law or hypothesis. Scientists usually set out in their experiments to try to disprove ideas or hypotheses. Doing an experiment which produces results which fit a hypothesis does not really prove the hypothesis, because you will not know it there is some pertinent circumstance which does not fit the hypothesis. For instance suppose your hypothesis is that all four legged animals are dogs. If you look around and keep seeing dogs you may think your hypothesis is true. However, as soon as you see a four legged animal which is a cat you know your hypothesis is not true and has to be changed. Of course, this example does not relate to an experiment, only to observation, but it makes the point that hypothesis are best tested by setting out to disprove them.This however, is only one aspect of Popper's thought on this issue. In a discussion of "Science versus Non-science", he says:
Thomas Kuhn is more charitable than Popper. He refers to the more usual practice of scientists as "Normal Science", that is scientific investigation which is guided by accepted paradigms (i.e. ideas and theories in a general sense), such as using the paradigm to explain aspects which have yet to be considered by scientists, or doing work to test the paradigm. The aim of the work is not unexpected novelty, it is an extension of accepted ideas. This is much like the standard Ph.D. research in a university. Before getting very far into her work, the graduate student often has to tell her supervisor and other professors what she expects to get from the research, i.e. surprises are not expected. In Kuhn's words, the work does not aim at unexpected novelty. However, every now and then a crisis emerges. Some observations do not fit the accepted paradigm or various awkward compromises are adopted to make the anomalies fit the paradigm. Eventually someone has a bold new idea and the new paradigm emerges, a scientific revolution (7).
Science for engineers
To return to what engineers mean by "science" and the use they make of science, when the Structural Engineer says that engineering is science, what does it mean? According to Petroski, when the structural engineer conceives his idea for a bridge that concept is a work of art. After conceiving the basic ideas for the type of structure to use, the engineer then becomes a scientist as he analyzes the structure rigorously. That is the engineer calculates the strength of every member of the bridge to determine that the bridge is safe and will do the job when built according to the ideas adopted for its structure. The analysis may also lead the engineer to change the detail of the design, or even the concept, in order to arrive at a better design, for instance one which will be cheaper to build, or perform better in some way.
Is this sort of calculation the work of a scientist? I am sure Ziman would not agree. Science cannot be usefully defined as anything a scientist does professionally. That is not to say that someone trained in science, in this case probably physics or mathematics, could not do the job. As mentioned, Edison employed a physicist to do this kind of work when he was developing the first electric power utility in New York. In that case the physicist employed the known laws of electricity to estimate current, voltage, power, etc in the system. If you are asked to calculate the strength of a bar, or how much it would lengthen under a certain tensile force you would be using the results of science but would you be doing science when you perform those calculations? You might apply Hook's Law, which states that the strain is proportional to stress, which is a law derived in science (albeit a simplification of what actually happens). This work would be important if applied to bridge, public safety would be involved, but you can be sure no scientist would get much recognition from other scientists for doing it. These calculations, however intricate they may be, will simply be the application of known rules and not a part of science, not even what Kuhn calls normal science. Science is a means of advancing our knowledge, giving us a new or extended understanding of the subjects concerned. Technology may involve the application of that knowledge. What we tend to confuse with science is the systematic method usually employed in such calculations. Scientists do not necessarily work systematically in their essential work which is the generation of ideas. It is only in conducting experiments and taking measurements that they may be systematic. Systematic procedure is not the essence of science. Yet this is often the notion behind an engineer's claim to be doing science or following the scientific method.
There are, nevertheless, situations in which engineers do do science. Indeed, anyone can do science in the sense that you don't have to trained in science to do science, although it helps to have some training and experience, both in the doing and in getting acceptance from other scientists. If an engineer sets out to study the creep of concrete, i.e. the slow deformation which occurs under load, a phenomenon pioneering designers with pre-tensioned concrete found they had to cope with, the engineer may engage in a form of science, namely an investigation of how creep depends on various variables. Some ideas on creep may be arrived at on observing the phenomenon in engineering structures, and the engineer may set out to test these ideas through experiment. In many cases this could be relatively straightforward work. Loads would be applied to test columns of concrete and the change in dimensions observed over time. Either linear or non-linear mathematical relationships would be fitted to the data. The relationships would only be as complex as needed to provide a reasonable fit. The intellectual challenge would not be anything like that facing Watson and Crick with DNA, but the investigator would nevertheless get some satisfaction from it, and the results would, and should be, subject to scrutiny and discussion by other investigators of this or similar phenomena. There is nothing wrong in calling this science. It would be even more within the scope of science if the investigators went on to provide ideas on how creep occurs, what goes on in the concrete as it deforms, and if work is done to try to confirm these ideas. However, all this is different from applying known relationships on creep to a structure, simply doing the calculations to estimate the amount of creep that will occur. That could be an application of the results of science.
Petroski does, however, bring science into engineering design in another, illuminating, way. Engineering design shares certain characteristics of science in that engineering design is always an hypothesis, sometimes strikingly so. The structure, when built, provides a test of that hypothesis. The hypothesis includes not only that the structure will stand, but that it will serve its intended purpose for the intended time, including staying within budget in construction and maintenance, and not produce any unexpected undesirable effect on society and environment. Robert Maillart's work might be considered science in this respect. It would be worth discussing whether this kind of engineering hypothesizing amounts to science.
The Brooklyn Bridge, completed in 1883, only had to be improved and strengthened in 1948 to carry current traffic. Certainly not all engineering structures end up satisfying their intended or implied hypothesis. A suspension bridge completed in 1966 across the Severn in England, had to have traffic levels across it restricted only fifteen or so years later to half the intended amount and a great deal of money spent on repairing it (8).
Postman's concerns - Scientism
In the previous section the main issue was the nature of engineering work, to what extent the practice of engineering is science, or scientific. There is another aspect of the attitude towards science and the use of science by scientists and non-scientists, including engineers, which concerns Neil Postman. He calls this "scientism".
Neil Postman is obviously not concerned with what science is or is not, but rather with the use and misuse of science as authority in human affairs. This is something engineers are also prone to, that is claiming their use of science as reason for more authority for their work, be it technical work or their ideas on social policy.
Engineers tend to pride themselves on their "scientific" training. As we have noted this really means that they have learned to be systematic. Furthermore, most of their technical knowledge has been systematized, i.e. formulated as rules and equations. In any given situation there is, or appears to be, a systematic procedure for the selection and application of the rules. Just as Hook's and Ohm's laws appear to be immutable, so all the systematic procedures taught to the engineer appear to be immutable law's of nature. However, just as there are many different ways to design a bridge, there are many different ways to proceed in many engineering situations. While only one way may seem to be acceptable at the time, that may only be a result of custom and current consensus, not a law of nature.
Engineers can also believe that their training gives them special skills in solving problems, any problem, because they have been fed a steady diet of problems to solve during their training, and much of their practice involves providing solutions to certain needs. An early engineering professor expressed this confidence, even arrogance, as so many do:
The use of the term "law" in this statement is part of the problem. Ordinary laws, enacted by a legislature, usually are statements of what members of a society must do, or must not do, in certain situations, along with penalties for anyone performing a prohibited action. A law in natural science is a description of what is expected to happen in a particular aspect of nature under certain circumstances. Usually the law is expressed mathematically and appears to have all the precision of common mathematical statements. For instance, Hooke's Law states that the extension (or strain) produced in an object by a force is proportional to the load (or stress) applied. This law is, in fact, only an approximation which is useful in many circumstances, but only if the strain in question is small. The same can be said for such well established laws, as Newton's Law of Gravity. More important in the present context, it is really a condensation of human experience of nature and human thought about nature. It is not a rule nature has to follow, a rule God has decided must be followed by materials (At least, we have no way of knowing whether that is so, or not!), though no doubt this was indeed the feeling many investigators of natural phenomenon had, including Hooke and Newton.
From the discovery that natural inanimate phenomenon can usually be described neatly by "laws", came the notion that the same must apply to the animate part of nature, including insects, birds, and animals, and there has been the same quest for simple relationships in psychology and sociology, i.e the individual and social behaviour of human beings. Many social scientists have felt, at least until well into this century, that the natural sciences should be emulated in their fields, but it has been, and probably still is, a particular temptation for engineers, as well as theorists on business management, to think that human economic and social behaviour can be described in terms of simple statements, i.e. laws. Most of what is taught in engineering is a distillation, an extract (often simplified), of a larger subject. This is true of what we teach in the mechanics of materials, materials science, thermodynamics, fluid mechanics, electromagnetic field theory, control theory and robotics. This is particularly true of the little exposure engineers get to the "softer" subjects, relating to business practice and social relations. As engineers we are very keen to extract simple ideas and rules from these subjects. In doing so we tend to ignore the uncertainty and controversy involved, and easily fall into the trap of assuming that these ideas are infallible guides for action. Engineers are not alone in this simplistic view of science and social science.
Postman points to an unreasonable faith in science among the general public and points to the way that faith can be exploited. He also castigates social science for its attempts to be like the physical sciences, and hence gain authority. Perhaps one reason for the desire of engineers to be considered scientific in their work is also, in part, a wish to gain authority. In part this may also be due to our ignorance of what real science is. We have hoodwinked ourselves, just as much as the media and the general public have been hoodwinked.