Design and Science (Part 2):
An Objectives-and-Methods Approach

by Craig Rusbult, Ph.D.

    Introduction
    This page builds on the foundation of Design and Science (Part 1), which describes the similarities and differences, in objectives and methods (in purpose and process), that are the basis for the Objectives-and-Methods approach (O-and-M or OM) discussed in this page.  Design and Science (Part 1) ends with a preview of the page you're now reading:

    These ideas, and others, will be explored more deeply in a follow-up page I'm writing: Design and Science (Part 2).  During this exploration, a few of the many interesting questions are:
    1) How should we define science and engineering?  Should we focus on functions (such as objectives and methods) or careers (by defining science as whatever a scientist does, and engineering as whatever an engineer does)?
    2) Is "applied science" (however this is defined) more similar to science or engineering?
    3) In each field, how should we view the contributions of experimentalists (who produce observations) and theoreticians (who interpret observations)?
    4) Do some people think the goal of science is not (or should not be, or cannot be) a search for truth?

    Currently, in the rest of this page the focus is Question 1:  How should we define science and design?
    Eventually, 2 and 3 will be discussed here, too.
    Question 4, re: science as a search for truth, is carefully examined in a page that asks, Should Scientific Method be EKS-Rated?  {note:  EKS has replaced X, in an effort to fool the filtering programs, in a page that's about "Wild controversies and hot debates!  Are some views of science dangerous for students?  Can too much of a good thing be harmful?  Do scientists seek the truth?  Do they claim proof?  Do they create reality?  How can we avoid running off (or being carried away) to silly extremes?"  [quoted from a description in the Grand Tour of my pages about learning, thinking, and teaching]}

    Defining (and comparing) Design and Science

    My "objectives and methods" model is based on comparing science and design, in order to understand their essential similarities and differences:

    In science our overall long-term objective is to search for the truth, to develop theories that are accurate representations of reality.  During this search our most useful [methodological] tool is a reality check that compares theory-based predictions with observations.
    In conventional design the main objective is to develop an improved product or strategy, and our most useful tools [methods] are quality checks that compare goals with predictions, and goals with observations.  Although in design it may be useful to get feedback about theories by comparing predictions and observations, this is not the central focus of action, as it is in science.

    In the remainder of the page, this theory about science-and-design (their relationships and their similarities and differences) is not examined with the sophistication and precision that it deserves, that eventually it will have when I'm able to invest more time in developing it.  Despite the loose ends, however, I think you'll find the following discussion to be interesting and stimulating, so here it is:

    The two main criteria for evaluating a theory are plausibility and utility: plausibility asks "Is it likely to be true, to accurately describe reality?", while utility asks "Is it useful?"

    The perceived plausibility of my Objectives-and-Methods (OM) approach will depend on our definitions:
    If science is defined as "what scientists do" and design is "what designers do," then some aspects of OM will appear unsatisfactory because "a scientist sometimes does design, and a designer sometimes does science."  Similarly, other definitions may lead to other perceived inadequacies in OM.  In other words, to the extent that science and design are defined in ways that differ from the OM-definitions, the OM-model may appear to be indadequate.
    But if science and design are defined in terms of objectives and methods, as in OM, the questions become more focused and we can ask, "In what ways are the O-and-M descriptions in OM satisfactory, and in what ways are they inadequate?"  Does OM allow — or even better, does it encourage and facilitate — a description of science and design that is similar to the views held by scientists and designers, by philosophers who think about "what scientists and designers should do," by historians who study "what scientists and designers actually do," and by others?
    What will be the outcome of our evaluations (about the plausibility and utility of how O-and-M are described in OM) and our debates about definitions?  I'm not sure.  I think OM is reasonably plausible, providing a reasonably accurate (although incomplete) description of the basic thinking methods that are used in science and design.  But others (and perhaps myself at a later time) may see things differently.

    When I'm considering OM's perceived plausibility (as judged by the community of scientists, designers, and scholars) my current response is that "I'm not sure."  By contrast, I'm very confident about the potential utility of OM.  In my opinion, OM provides a useful analytical framework that could stimulate plenty of productive thinking and could help improve our understanding of science and design.
    One reason for this contrast is that, in the early stages of its development, the utility of a theory does not necessarily depend on its accuracy.  If our goal is a search for truth (and I think this is the proper goal), we should always acknowledge the ways in which a theory seems inaccurate or incomplete, and eventually an incorrect theory should be either abandoned or modifed-and-improved.  And always, from the beginning, its limitations should be acknowledged.  But in the "pursuit" stage of development, when we are exploring a theory's implications and potential applications, and testing its accuracy, a false theory can be useful.
    Although this claim may sound strange when you first hear it, some rational justifications for it are outlined in the conclusion of a section about the criteria used by scientists when evaluating a model that is an "intentionally simplified artifical representation" of a more complex real system:

    Wimsatt (1987) discusses some ways that a false model can be scientifically useful.  Even if a model is wrong, it may inspire the design of interesting experiments.  It may stimulate new ways of thinking that lead to the critical examination and revision (or rejection) of another theory.  It may stimulate a search for empirical patterns in data.  Or it may serve as a starting point; by continually refining and revising a false model, perhaps a better model can be developed.  /  Many of Wimsatt's descriptions of utility involve a model that is false due to an incomplete description of components for entities, actions, or interactions.  When the erroneous predictions of an incomplete model are analyzed, this can provide information about the effects of components that have been omitted or oversimplified.  For example, to study how "damping force" affects pendulum motion,...;  or consider the Castle-Hardy-Weinberg Model for population genetics, which assumes an idealized system that never occurs in nature; deviations from the model's predictions indicate...  {note: I've added emphasis and have omitted the details (indicated by "...") that aren't necessary for this page.}  This excerpt is from Section 2A — which discusses "Simplicity" in terms of Logical Systematicity, Simplified Models, Coping with Complexity, Tensions between Conflicting Criteria , and (in the excerpt) False but Useful — in my Detailed Overview of Scientific Method.

    In my opinion, OM is a fairly accurate "first approximation model" that can be refined into a model that is even more accurate.  But I don't think any model of "science and design" — including OM and its logical extensions (which are implied whenever I say "OM") — will ever be totally accurate and complete. (*)  Although I think OM is basically sound, my expectations are realistic, so when I find ways in which OM is inadequate, and when others disagree with the views I've constructed, I won't necessarily be disappointed.  These flaws and debates will be fine with me, as long as the thinking and discussion that's being stimulated by OM is interesting and useful.  My ultimate goal is improved understanding, not a defense of OM.  { If we find a situation where OM seems to be accurate, this can be useful.  And if we find a situation where it seems in accurate, this can also be useful. }
    * Certainly, OM is incomplete.  Obviously, the actual complexity of science (as characterized in my comprehensive model of Integrated Scientific Method (*) which is summarized in a brief-yet-thorough Overview of Scientific Method) is much greater than the oversimplified model of science in OM.  But the main themes of science — comparing predictions with observations, with the goal of searching for truth about nature — are featured in OM, and this is all that I claim for it.  The features of design are similarly simplified, but again I think the essence of design is in the model.  /  * In the model's name, "Integrated" indicates that the model is based on integrative analysis in which the goal is not just to characterize the individual activities that are part of the process of scientific thinking, but also (by characterizing their mutually interactive relationships) to show the dynamics of their integration into a coherent overall process.

    Is it rational to ask if OM is true or false?  During evaluation in natural science, we are comparing a theory (that we have constructed) with nature (which we have not constructed).  But during an evaluation of OM, often we are comparing one theory (that is humanly constructed) with other theories (that also have been humanly constructed), so reality checks — which are the essence of natural science — must be interpreted in a more sophisticated way, with appropriate humility.  We should recognize the logical limitations that are inherent when we compare humanly constructed theories, when we compare an OM-model with other models that humans have constructed about science and design, about their definitions, characteristics, and relationships.  When we're thinking about OM it is much less appropriate — compared with a situation where a scientific theory about a natural phenomenon is being evaluated — to ask whether OM is true.
    Here is a summary/review, comparing the two contexts:  1) When we evaluate the plausibility of a scientific theory, we're asking whether it accurately describes what happens in nature, and we don't control the laws of nature.  2) But when we evaluate a model of science-and-design, one of our main questions is whether the model is an accurate representation of our own views about science-and-design, which we (as individuals and in communities) have constructed and do control.  In both 1 and 2, we should do "reality checks" in an effort to determine the extent to which a theory/model matches reality, but in the contexts of 1 and 2 the meaning of the "reality" differs in important ways.

    For example, imagine that someone criticizes OM because engineers (who are designers?) sometimes compare observations with predictions (but in OM this is "doing science") to test the adequacy of a theory they are using.  Is this criticism trivial, based only on a disagreement about whether the word "design" should be defined in terms of careers or methods?  In this case, we are just using different labels for similar views.  Or do the different labels reflect significant differences in the ways we are thinking about design?
    In this example, let's consider one type of thinking: the mental process of comparing predictions with observations.  Is it inconsistent to call this process "design" when it occurs in the overall context of design, but to call the same process "science" when it occurs in the overall context of science?  Or is it inconsistent, during a long-term process of design in which the overall goal is to improve a product, to say that a designer is shifting back and forth between short-term processes of design (when the immediate goal is to seek improvement) and science (when the immediate goal is to seek understanding)?
    Should we try to choose between these two perspectives?  Or should we, at least temporarily on a trial basis, try to develop both perspectives?  It might be productive to carefully study, using different perspectives, the situations in which (according to OM) a scientist does design, or a designer does science.  I think this type of study would help us improve our understanding of design and science, and would help us construct models that are more complete and accurate.  { Of course, asking "Can a designer do science?" is only one of the many interesting questions that could be asked in multi-perspective studies. }
    some comments about perspectives:  There are differences, among scholars, in our attitudes toward analysis, in our tendencies to use categories or avoid them, to choose criteria for categorizing (and to be lumpers or splitters), to categorize in terms of multiple dimensions, and to see the mutually interactive relationships that form the integrative aspect of integrative analysis.  Instead of simply arguing about the relative merits of each type of preference, maybe we should try to learn more from each other.

    Putting Things in Context: A Practical Perspective
    The main purpose of OM (and the models of design and science on which OM is based) is not scholarly discussion, it is educational application.  I'm hoping that OM, as part of an educational approach that uses Integrated Design Method and Integrated Scientific Method in creative ways, will help students understand design and science, and will help them learn how to think more effectively and enjoy it more.  Some ideas for doing this are outlined in the "Design and Science in Education" parts of An Introduction to Design and are developed more fully in other pages.

    As mentioned earlier, at this time in this page I won't be giving the many important questions the attention they deserve.  Instead, I'll re-emphasize the main theme, which is fairly simple:
    I don't expect everyone to agree with everything in OM, but I think our agreements and disagreements (about where OM is and isn't adequate) will be interesting and useful, will stimulate creative and critical thinking, and will improve our understanding of science and design.