Changing Existing Situations Into Preferred Ones
Simon's definition of design, the chapter he rewrote between editions, and what a science of the artificial would actually have to contain.
Date
Category
Writer
The first edition was a small book: xii plus 123 pages, four chapters, published by MIT Press in 1969. Three of the chapters were the Karl Taylor Compton Lectures, which Herbert A. Simon delivered at MIT in the spring of 1968. The fourth, "The Architecture of Complexity", was a 1962 essay bolted on at the back. A contemporary review held in the Herbert Simon collection at Carnegie Mellon describes the project as "the design of an intellectual structure aimed at accommodating those empirical phenomena that are 'artificial' rather than 'natural'". That is a thin volume to have founded a field on.
The sentence everyone knows from it is one line in one chapter. I wanted to establish exactly where it sits, what Simon says immediately afterwards, and which edition I was actually quoting. The answer turned out to matter more than I expected.
The sentence, and the sentence after it
In the third edition, chapter 5 is "The Science of Design: Creating the Artificial", running from page 111 to page 138. The definition appears near its opening:
"Everyone designs who devises courses of action aimed at changing existing situations into preferred ones."
What almost never gets quoted is the sentence that follows, which is where the claim gets its force: "The intellectual activity that produces material artifacts is no different fundamentally from the one that prescribes remedies for a sick patient or the one that devises a new sales plan for a company or a social welfare policy for a state."
Read on its own, the first sentence sounds like a generous invitation. Read with the second, it is a much harder proposition: that making a chair, writing a prescription and drafting welfare policy are instances of one intellectual activity, and that this activity can be studied as such. Simon states the consequence directly: "Design, so construed, is the core of all professional training; it is the principal mark that distinguishes the professions from the sciences." And: "Schools of engineering, as well as schools of architecture, business, education, law, and medicine, are all centrally concerned with the process of design."
The chapter is not a celebration of design. It is a complaint about what had happened to it: "In view of the key role of design in professional activity, it is ironic that in this century the natural sciences almost drove the sciences of the artificial from professional school curricula", followed by the line that should be printed on the wall of every design school that has recently added a research degree: "Engineering schools gradually became schools of physics and mathematics; medical schools became schools of biological science; business schools became schools of finite mathematics."
The famous last line of the chapter is also usually mangled. It is not a free-standing aphorism. Simon writes that "The proper study of mankind has been said to be man", argues that people may be relatively simple and that most of the complexity of their behaviour is drawn from their environment, and only then concludes: "The proper study of mankind is the science of design, not only as the professional component of a technical education but as a core discipline for every liberally educated person." The claim about every liberally educated person is the whole point, and it is the half that gets cut.
Which edition says what
This is where I had been sloppy, and where most citations I have seen are sloppy too. In the preface to the second edition, reproduced in the third, Simon states the provenance of each chapter plainly: "The Compton lectures comprise chapters 1, 3, and 5, and the Gaither lectures, chapters 2, 4, and 6."
So the design chapter descends from the 1968 Compton lectures and was in the book from the start. The chapters on economic rationality, on memory and on social planning came in later, from the Gaither lectures. In the preface to the third edition Simon says he added a new chapter on complexity, moving "The Architecture of Complexity" to chapter 8, and adds: "Most of the remaining changes in this third edition aim at updating the text. In particular, I have taken account of important advances that have been made since 1981 in cognitive psychology (chapters 3 and 4) and the science of design (chapters 5 and 6)."
That last clause is the one that should make anyone careful. Simon explicitly revised the design chapter for the third edition. Citing the famous sentence as "Simon 1969, p. 55" therefore asserts something about a text I have not seen, in a chapter its author says he later updated.
Edition | Chapters | What it contains, on the record |
|---|---|---|
First, MIT Press, 1969 | Four, xii plus 123 pages | The three Compton lectures of spring 1968, plus "The Architecture of Complexity" (1962) as a fourth chapter. By Simon's own account of which lectures became which chapters, the design lecture is chapter 3 here. |
Second, MIT Press, 1981 | Not verified by me | Adds the Gaither lectures as chapters 2, 4 and 6, which by arithmetic implies seven chapters. I have not held a copy to confirm the count or the pagination. |
Third, MIT Press, 1996 | Eight, xiv plus 231 pages | Adds a new chapter 7, "Alternative Views of Complexity", pushing "The Architecture of Complexity" to chapter 8, and updates chapters 3, 4, 5 and 6 for work done since 1981. The design chapter is chapter 5, pages 111 to 138. |
Satisficing, and where it came from
Simon introduces the word in the design chapter itself: "Since there did not seem to be any word in English for decision methods that look for good or satisfactory solutions instead of optimal ones, some years ago I introduced the term 'satisficing' to refer to such procedures." He is precise about why it matters for design rather than for economics: "No one will settle for good or better if he can have best. But that is not the way the problem usually poses itself in actual design situations."
The "some years ago" points back to A Behavioral Model of Rational Choice, published in the Quarterly Journal of Economics volume 69, number 1, February 1955, pages 99 to 118. Its programme is stated on the first page: "The task is to replace the global rationality of economic man with a kind of rational behavior that is compatible with the access to information and the computational capacities that are actually possessed by organisms."
Worth noting, because the 1955 paper is routinely cited as the source of the phrase: I could not find "bounded rationality" anywhere in it. What Simon writes about there is approximating procedures and simplified approximations. The phrase came later. In The Sciences of the Artificial the idea is carried by chapter 2, "Economic Rationality: Adaptive Artifice", where the argument is that because real-world optimisation is impossible, the real economic actor is a satisficer who accepts alternatives that are good enough, operating against an aspiration level that moves.
For a designer this is not a footnote about economics. It is the epistemological licence for the whole practice. If optimisation were available, design would be a computation and there would be nothing to argue about. Satisficing says the alternative set is not enumerable and the evaluation is not complete, so the decision procedure is: set a standard, search until something meets it, stop. That is a description of what happens in a studio, and it is also an admission that the stopping point is set by the designer, not by the problem.
What a science of the artificial would actually require
Simon does not leave the proposal at the level of a slogan. The chapter ends with a section headed "Summary: Topics in the Theory of Design", which is a curriculum. The chapter's own section headings, and the shape of the syllabus he sketches, look like this:
Three things follow from taking that list seriously, and they are what a science of the artificial would actually require.
A vocabulary of phenomena, not of outcomes
Every item on Simon's list names a mechanism rather than a result. Not "good layout" but "representation of design problems". A science needs objects it can study across cases. Design discourse mostly names outcomes and styles, which are unique by construction and therefore cannot accumulate.
Results that survive being wrong
Utility theory, decision theory and heuristic search all produce claims that can fail. A field with no reportable failures is not accumulating. This is where design writing is thinnest: negative results are almost never published, and the ones that are tend to be published by adjacent fields.
Method described well enough to be repeated by someone who dislikes you
Simon's items 2, 4 and 5 are all procedures. A procedure is the unit a science trades in, because it travels without its author. Most design methodology travels with its author attached, which is why it reads as advocacy rather than as evidence.
Who took the programme up
The answer is: not, in the main, art-school design education. Engineering design and information systems did.
The clearest institutional marker is the journal Design Science, launched by Cambridge University Press on 24 July 2015. Its founding editorial, Design Science: Why, What and How by Panos Y. Papalambros, is article e1 of volume 1. It defines the field in almost Simonian terms: "Design science studies the creation of artifacts and their embedding in our physical, psychological, economic, social and virtual environments." It cites Simon's book in both its 1969 and 1996 forms, and it is candid that the field is still being assembled rather than established, calling for the consolidation of phenomena, research areas and terminology. Forty-six years after the Compton lectures, a journal was founded to do the consolidating that Simon's curriculum presupposed.
That is a real uptake, and it is almost entirely in engineering, management and computing departments. The parts of design education that train practitioners largely went the other way, and for a stated reason.
Who declined it, and why they were not simply wrong
The counter-tradition runs through Donald Schön. In Leonard Waks's account in the International Journal of Technology and Design Education volume 11, 2001, pages 37 to 51, Schön "rejected 'technical rationality' entirely. He saw is as the defective epistemology of practice generating the flawed normative professional education and research program of the research universities". Waks puts the contrast crisply: "For Dewey the paradigm site of education is the scientific laboratory; for Schon it is the design studio."
Schön's positive proposal is an inversion of Simon's. Where Simon wants design formalised so that professional schools can teach it as a science, Schön wants professional education restructured around the studio, the "practicum", and a coach, on the hypothesis that "all professions are 'designlike' in some relevant respects". Both men think design is the core of professional education. They disagree completely about what follows.
Nigel Cross made the distinction that clears up most of the confusion. His paper "Designerly Ways of Knowing: Design Discipline Versus Design Science", Design Issues volume 17, number 3, 2001, pages 49 to 55, separates three things usually run together: scientific design, design science, and a science of design. The Open University's record of the paper summarises his position as arguing that "there are forms of knowledge peculiar to the awareness and ability of a designer, independent of the professional domain of design practice". A science of design studies designers. Design science tries to make designing scientific. They are not the same project, and Simon is usually cited for the second when the defensible version is the first.
What I am not sure of
I could not obtain a first edition. Everything I have quoted from chapter 5 comes from the third edition of 1996, where the chapter runs from page 111 to page 138. The widely circulated citation "Simon 1969, p. 55" is consistent with a 123-page book whose design chapter is third of four, but I have not verified the page and I would not print it as fact. Nor can I say whether the definition's wording is identical across editions, given that Simon says he updated that chapter.
The thing I had wrong for years is smaller and more embarrassing. I quoted the definition as a democratic gesture, a way of saying that everybody is a designer. Simon's next sentence makes clear he meant something closer to the opposite: that design is a specific intellectual activity with a formal structure, teachable as a curriculum, and that professional schools had lost it. The sentence is not an invitation. It is the opening of an argument that design should be harder to enter, not easier.
Sources
Herbert A. Simon, The Sciences of the Artificial, third edition, MIT Press, Cambridge, Massachusetts, 1996, xiv plus 231 pages. ISBN 0262193744 and 0262691914. Chapter 5, "The Science of Design: Creating the Artificial", pages 111 to 138. punyamishra.com/wp-content/uploads/2017/09/simon-scienceoftheartificial-ch5.pdf
Table of contents of the third edition, with chapter start pages and the prefaces to the second and third editions. Library scan. toc.uni.li/FLMF018081.pdf
Prefaces to the second and third editions, used for the Compton and Gaither lecture attributions and the list of chapters updated in 1996. silo.pub/the-sciences-of-the-artificial-3rd-edition-r-2364756.html
Bibliographic record for the third edition, MIT Press 1996, xiv plus 231 pages, with the full chapter list. archive.org/details/sciencesofartifi0000simo
Contemporary review of the first edition, Herbert A. Simon Collection, Carnegie Mellon University Libraries, giving xii plus 123 pages, four chapters, three from the Compton lectures plus the 1962 essay. iiif.library.cmu.edu/file/Simon_box00077_fld06175_bdl0001_doc0001/Simon_box00077_fld06175_bdl0001_doc0001.pdf
Open Library record for the 1969 MIT Press edition, giving xii, 123 pages, Karl Taylor Compton lectures 1968. openlibrary.org/books/OL5680110M/The_sciences_of_the_artificial
Herbert A. Simon, "A Behavioral Model of Rational Choice", The Quarterly Journal of Economics, volume 69, number 1, February 1955, pages 99 to 118. cooperative-individualism.org/simon-herbert_a-behavioral-model-of-rational-choice-1955-feb.pdf
Panos Y. Papalambros, "Design Science: Why, What and How", Design Science, volume 1, 2015, article e1. doi 10.1017/dsj.2015.1. cambridge.org/core/journals/design-science/article/design-science-why-what-and-how/A575D750B05AE57AAF9E7CA07E44FEF4/core-reader
Leonard J. Waks, "Donald Schön's Philosophy of Design and Design Education", International Journal of Technology and Design Education, volume 11, 2001, pages 37 to 51. matt-koehler.com/OtherPages/Courses/CEP_901B_SP03/Readings/Design/schon.pdf
Nigel Cross, "Designerly Ways of Knowing: Design Discipline Versus Design Science", Design Issues, volume 17, number 3, 2001, pages 49 to 55. doi 10.1162/074793601750357196. Record and abstract at oro.open.ac.uk/3281
Related reading
Latest Articles.
© Moc Mallen Studio
Thoughts, ideas, and perspectives on design, simplicity, and creative process.
