Dear Klaus and Chuck,
This is a short reply to your notes. Because the thread has evolved and developed over several days, I am appending snip quotes of the key issues from past postings. The quotes are longer than this reply.
Thanks, Klaus, for your note [1, below]. The methods that designers use in thinking, solving problems, and creating products and services involve a specific set of general human skills, capacities, and activities. The role of communication and interaction with stakeholders of all kinds is paramount. The notion of an abstract “thinking” doesn’t really describe what designers do.
While Keith’s post refers to philosophers and thinking, this thinking is also highly conversational. Plato’s great work involved reports of the conversations and dialogs of Socrates. As it is, Plato did not likely record what Socrates said in exactly the way the Socrates said it — but Plato did write a conversational record of his memories, or at least an account of his evolving and changed memories of conversation.
Aristotle did not write most of Aristotle’s writings, either. He spoke them, and these accounts most likely evolved in conversations recorded by students.
The founding pragmatists — C. S. Peirce, John Dewey, George Herbert Mead, and William James — emphasised the conversational and interactive nature of thought. While Peirce made the major contributions to logic and developed the issue of abduction, Mead’s work reflects and supports your notes here [1, below] on the role of conversation and situated thought in effective human action.
With respect to Chuck’s post [2, below], I must respectfully disagree on several points.
As I see it, there is a real need for anyone who engages in action on behalf of other human beings to understand the issues involved in logic. While design thinking on behalf of others requires conversation, anyone who acts on behalf of others does so better with a few of the basics. Designers, physicians, lawyers, jurists, economists, bankers politicians managers all do better work for those whom they serve when they think more clearly.
When it comes to designing products and services, there is sometimes a need for deduction. Designers create new products or services, or — more commonly — they redesign existing products or services. They do so against a background of what human beings already know or believe. Some of what we know involves physical, chemical, biological, or technical facts — or facts of other kinds. Some of what we know involves well established principles of human behaviour. This includes principles that function at the individual level, the group level, in societies, or in cultures.
In my third post in this thread [4, below], I gave examples of the massive failure rate of products and services, and of the companies that produce them. Many of these failures could have been prevented by the simple use of deductive logic from well established principles and facts. Others could have been ameliorated. Some innovative products and services fail that should properly succeed — by applying both induction and deduction to the multiple contexts of creation and use, we can sometimes predict possible successes or failures well enough to test them.
It is problematic to argue that abduction and induction are sufficient to designers and design researchers. John Vickers (2016) addresses these issues in an excellent article on the problem of induction:
http://plato.stanford.edu/entries/induction-problem/
See, also, Frans Huber’s article on induction:
http://www.iep.utm.edu/conf-ind/
The universe has not changed that dramatically since David Hume first discussed this issue — we have managed to understand it more completely, but designers have no special trick that permits them to leap-frog this problem. Albert Einstein (1970: 53) attributed a decisive step in developing the theory of relativity to the critical reasoning that reading David Hume helped him to achieve. I argue that a deeper understanding of the relations between abduction, induction, and deduction is as helpful in developing a theory of design or of design thinking as it was in developing the theory of relativity.
Designers who solve problems do best when they use multiple methods to reach conclusions, and they reach conclusions best when they test and examine both premisses and outcomes. To do this, abduction, induction, and deduction are all necessary.
Yours,
Ken
Ken Friedman, PhD, DSc (hc), FDRS | Editor-in-Chief | 设计 She Ji. The Journal of Design, Economics, and Innovation | Published by Tongji University in Cooperation with Elsevier | URL: http://www.journals.elsevier.com/she-ji-the-journal-of-design-economics-and-innovation/
Chair Professor of Design Innovation Studies | College of Design and Innovation | Tongji University | Shanghai, China ||| University Distinguished Professor | Centre for Design Innovation | Swinburne University of Technology | Melbourne, Australia
--
References
Einstein, Albert. 1970 (1949). “Autobiographical Notes.” Albert Einstein: Philosopher-Scientist. P. A. Schilpp, trans. and ed. La Salle, Illinois: Open Court, pp. 1-95.
Huber, Franz. 2016. “Confirmation and Induction.” Internet Encyclopedia of Philosophy. URL:
http://www.iep.utm.edu/conf-ind/
Vickers, John. 2016. “The Problem of Induction.” The Stanford Encyclopedia of Philosophy (Spring 2016 Edition), Edward N. Zalta (ed.). URL:
http://plato.stanford.edu/archives/spr2016/entries/induction-problem/
—
Klaus Krippendorff wrote:
[1]
—snip—
... human thinking is always derivative of experience obtained in conversations with fellow human beings they have grown up with. this is particularly so because designers are always working with others, with clients where problems are negotiated, in design teams where solutions are proposed, elaborated or replaced by better ways. even when they have internalized these conversations, are able to talk with themselves and come to their own proposals, they cannot help but having to communicate their designs to clients, producers, users, etc. = stakeholders that need to be enrolled in the designers' project or a designer fails as a professional.
unless designers articulate how they came to what they are suggesting, what needs to be done to realize a design, and who could benefit from a design, their thinking is an entirely private affair, accessible only to the thinking individual but irrelevant to understanding design practices.
—snip—
--
Charles Burnette wrote:
[2]
—snip—
This discussion fails to illuminate how abduction might occur in design thinking, to indicate why it is appropriate to designing. or to explain how it might be used to establish an outcome that can be empirically tested. As with “heuristics”, “abduction" is treated as an umbrella word without immediate practical definition or application.
The paper “Building and Using A Theory of Design Thinking” at
http://www.independent.academia.edu/charlesburnette
demonstrates how abductive thinking can be applied to produce testable outcomes. Although interpretive correlation is an inductive process. I see no reason for deduction to be called upon. Interpretation and the practical testing of outcomes and assumptions are highly desirable.
"Situated thought” involving the context, background, and circumstances of interest or concern, the identification of “Needs and Desires" within this situated thought, "Intentional stances" that brings purposeful interpretation to each situation, “Structured Thought" based on scientifically valid findings and practical utility, and an "Iterative Process" of evaluation, assimilation and adaptation are seen as necessary components of an abductive process.
We need to focus research on the mechanisms of how designers actually think rather than attempt to force design thinking into normative patterns and methods basic to other disciplines.
—snip—
--
Adam Girard wrote:
[3]
—snip—
... understanding logic is important to making design arguments, and it would be great if more was written to make this explicit in the relationship between science and design.
There are a few resources that may be helpful in addition to those listed by Ken.
In the design context, this was addressed in:
Dorst, K. (2011). The core of “design thinking”and its application. *Design Studies*, *32*(6), 521–532.
Additional resources that I found useful include:
Mans, D., & Preyer, G. (1999). On Contemporary Developments in the Theory of Argumentation. In *Reasoning and argumentation* (Vol. 13). Protosociology: An International Journal of Interdisciplinary Research.
Walton, D. (2001). Abductive, presumptive and plausible arguments. *Informal
Logic*, *21*(2). Retrieved from
http://windsor.scholarsportal.info/ojs/leddy/index.php/informal_logic/article/view/2241
—snip—
—
Ken Friedman wrote:
[4]
—snip—
... Deduction requires necessary premisses. An abductive inference does not provide the foundation for necessary premisses.
What is missing here is a series of steps that show the results of an abductive inference to be valid or necessary. This requires testing. In science, it requires significant testing. One reason for the high prestige given to deductive inference in physics is the extensive testing of premisses. It has taken a century of careful, repeated, and varied tests to find evidence for gravitation waves, the last major deductive prediction of Einstein’s general theory of relativity.
Your description of how designers sometimes work is reasonable. Your description of the relationship between abduction and valid hypothetic-deductive inference is only partial. If you are describing a general heuristic approach to design outcomes that *seem* to work, this is one way that things happen.
The outcome of the normal heuristic design process in developing products and services is far from reliable. Much like the process of evolution in nature, heuristic processes work effectively at a high price in failed developments and extinct lines. The evidence of new product failure is clear. In one study, Mansfield et al. (1971: 57) concluded that once new product ideas move beyond the proposal stage, 57% achieve technical objectives, 31% enter full-scale marketing, and only 12% earn a profit . According to others, over 80% of all new products fail when they are launched, and another 10% fail within five years (Lukas 1998, McMath 1998).
Clearly, the world would benefit from better ways to work. At the same time, many designed artefacts form their own tests in the real world — the question is how many tests the world can afford, and how much waste we can still manage. We live in a world where entrepreneurs, companies, inventors — and designers — want to produce things, try them in markets, and test them at scale. Thus we live in a world where 80% of all new products fail on launch, 10% more fail within five years, and only a small fraction of new companies last more than five years.
The reason that I point to these statistics is that they are one outcome of the normal process of design and innovation as we have practiced it for the past several centuries. One goal of design research is to achieve better outcomes with less waste. While design is not a science in the sense that physics is, design is a professional practice that can benefit from better thinking. It can, to some degree, benefit from a richer foundation in science, in much the way that medical practice does.
As with medical practice, design often involves diagnosis to solve problems. Not all design involves solving problems — design sometimes involves invention or pure creation. Even invention or pure creation may involve research or science. For example, skepticism and rigorous testing should be our approach to any product or service based on principles that claim to defeat the laws of physics. Designers invent things that do not yet exist in the world. The iPhone, Google, and the hoverboard are cases in point. The perpetual motion machine is not.
Serious research and richer research training will hopefully help us to reduce waste and create a better world. This is definitely the case when designers work with complex socio-technical systems (see, f.ex., Norman and Stappers 2016). To a great degree, much of what designers do will involve trials in a real world governed by imperfections, hurried managers, deadlines, and inadequate resources. The goal of research in the university system is to pursue responsible, valid answers without respect to the commercial pressures that designers face in the workplace.
My reason for posting my notes on abduction, induction, and deduction was to clarify a research issue. Again, heuristics work — or seem to — because that’s often all we have time for in design practice. It’s also the reason for the massive failure rates we face in design practice. I posted a note to clarify an issue in an ideal sense. Understanding the nature of abduction, induction, and deduction won’t help you to overcome the difficulties and challenges of design practice when you face the constraints of industry and business — but this kind of understanding will help you to think better and more effectively, so understanding these issues thoroughly will give you a modest edge.
The better designers still have their failures. Products and services fail for a great many reasons. What I suspect, however, is that some designers get their failure rates down from 80% plus 10% to 60% and 5%. Some do even better. Over a lifetime of designing products and services, that adds up to a real difference.
Achieving this difference is one reason for design research. Compare the failure rates in medical practice in 1890 or 1915 against our failure rates today. You can also measure these differences in actuarial rates and human longevity, and you can measure them in the diseases that no longer exist. I’d like to think that we can achieve similar improvements in design practice — I doubt that we can make equally massive gains, but we can make useful gains.
To do this, designers need some fundamental thinking skills and research skills, much as physicians do when they take their pre-med sequence. One set of thinking skills involves understanding the difference between rigorous, valid logic and workable heuristics.
Heuristics work. Everyone uses them, just as we all use rules of thumb and we all use abduction. We must nevertheless recognise the differences between different modes of logic, what they mean, and how to use them.
One cannot transform an abductive conclusion to a deductive premise and then expect the deductive conclusion to be *necessarily* valid.
...
References
Lukas, Paul. 1998. “The Ghastliest Product Launches.” Fortune 16 March 1998, p. 44.
Mansfield, Edwin, J. Rapaport, J. Schnee, S. Wagner, and M. Hamburger. 1971. Research and Innovation in Modern Corporations. New York: Norton.
McMath, Robert. 1998. What Were They Thinking? Marketing Lessons I’ve Learned from Over 80,000 New Product Innovations and Idiocies. New York: Times Business.
Norman, Donald A. and Pieter Jan Stappers. 2016. "DesignX: Complex Sociotechnical Systems.” She Ji: The Journal of Design, Economics, and Innovation. [Uncorrected Proof.] doi:10.1016/j.sheji.2016.01.002
http://www.sciencedirect.com/science/article/pii/S240587261530037X
—snip—
--
Ken Friedman wrote:
[5]
—snip—
In my post yesterday, I seem to have omitted James Hawthorne’s (2014) article from the Stanford Encyclopaedia of Philosophy on inductive logic. This article also contains links to other useful resources, including encyclopedia entries and teaching resources.
http://plato.stanford.edu/entries/logic-inductive/
—snip--
--
Ken Friedman wrote:
[6]
Several times in the past few days, colleagues have written to ask for material on abduction, induction, and deduction — and the differences among them. Because this issue comes up frequently in research seminars and doctoral courses, I have gathered my notes from different replies to share them with the list.
There is a great deal of confusion on the issue of abduction. Put simply, it is a way to describe hypothesis formation. We often hear people describe induction as “inference to best explanation,” but this is not entirely accurate. While inductive inference is an inference to a *likely* best explanation, an inductive inference does not tell you whether the explanation is true, valid or correct.
Many of the false convictions that are now being overturned through DNA and other forms of scientific evidence were originally obtained by police officers or prosecutors who relied on abduction inference rather than hard evidence. One reads many versions of this story:
A crime takes place. Person X — or someone who looks like Person X — is seen in the area around the time of the crime. Despite the absence of hard evidence, if there are no other suspects for the crime, the *likely* best explanation is that Person X committed the crime. It is true that a crime has been committed. And it is therefore true that someone must be guilty. The problem with many convictions is that closing the case requires ascribing guilt.
Abduction also plays a role in many of the great tragedies of history. When much of the world believed in witches and witchcraft, people were killed as witches through accusations based on abduction. Now that we no longer believe in witchcraft, we recognise the accusations of witchcraft as false. Those accusations were often based on abductive inference.
Abduction is good when used properly.
Abduction had a role in many scientific breakthroughs and technological developments.
Scientists test abductive inferences. Inventions work — or they don’t.
It is incorrect to say that designers somehow achieve results through abduction while scientists use induction and deduction. Designers more often fail through abduction than they succeed. And scientists more often find that hypotheses fail than succeed. The point is to find what works, then find out why it works, without mistaking intuition or inspiration for valid conclusions.
The short version of the argument is that scientists and scholars use induction and deduction to achieve reliable scientific or scholarly results, where designers use abduction to achieve reliable outcomes. This is not so. Abduction is not a logic of justification, and reliability lies within the logic of justification.
It is also wrong to suggest that scientists and scholars do not use abduction. They do. Everyone uses abduction to some degree — in daily life as well as in research. Abduction involves hypothesis formation, but an hypothesis is only an idea without evidence to determine whether the idea has substance. Abductive inference it not reliable. It is indicative. Just as one must test hypotheses, one must test abduction to determine whether an abductive inference is correct.
Mautner’s Dictionary of Philosophy (2000: 1) explains abduction well:
“abduction n. 1 (in Aristotle) a syllogistic inference from a major premiss which is certain, and a merely probable minor premiss, to a merely probable conclusion (Prior Analytics 2,25 69a 2o ff.).
“2 (in C. S. Peirce) reasoning of this form: (a) facts of type B have been observed; (b) a true statement of the form If A then B can explain B. Therefore, probably A. Peirce called this pattern abduction, believing that he used the term in the Aristotelian sense. He held that abduction is the standard form of setting up scientific hypotheses, and can count as the third kind of inference, together with induction and deduction.Since then, it has been stressed that what makes A probable is that it is the best explanation we can think of. Scientifically useful abduction is, then, INFERENCE TO THE BEST EXPLANATION. The general form of such an inferenceis: (1) D is a collection of data; (2) H (a hypothesis) would, if true, explain D ; (3) no other hypothesis can explain D as well as H does (4) Therefore, H is probably true. Of course, abductive reasoning is common also in everyday life, whenever we try to find answers to questions why something is the case. Syn. retroduction.”
In design, many people refer to the great logician and scientists C. S. Peirce without actually reading Peirce. For Peirce, abduction belongs within the logic of discovery. One requires more than abduction for the logic of justification.
To see what Peirce (1998) has to say, I suggest the excellent collection from Indiana University Press.
Igor Douven (2011) wrote an excellent article on abduction in the Stanford Encyclopaedia of Philosophy.
http://plato.stanford.edu/entries/abduction/#DedIndAbd
He also wrote a supplementary article on C. S. Peirce’s views
http://plato.stanford.edu/entries/abduction/peirce.html
Robin Smith’s (2015) discussion on induction and deduction appear in the article on Aristotle’s logic
http://plato.stanford.edu/entries/aristotle-logic/#IndDed
The Internet Encyclopaedia of Philosophy also has a good article on induction and deduction
http://www.iep.utm.edu/ded-ind/
Mautner (2000: 145) describes deduction this way:
“deduction n. The premisses in a deduction do not have to be general or necessary. But in that respect, older concepts of deduction differed: 1 a deduction is a valid inference from necessary premisses. This is a traditional concept of deduction. Descartes defined it as an operation by which we have insight into something which follows necessarily from other things that are known with certainty “Rule III in Regles pour la direction de l'esprit (Rules for the direction of the mind)”. 2 a deduction is a valid inference from more general premisses to a less general i.e. a more specific conclusion. It is contrasted with induction which is an inference from particular instances to a general conclusion. This is the classical Aristotelian concept. 3. (in older jurisprudence) a deduction establishes the legal in contradistinction to the factual grounds for action in a court of law. This is the sense used metaphorically by Kant in the Critique of Pure Reason. His ‘transcendental deduction’ of the categories is the justification of the application of the categories to objects, i.e. the account of why the categories necessarily apply to all objects of experience. 4 in the modern sense, a valid deduction or a valid deductive inference is one in which the conclusion is a necessary consequence of the premisses so that the conclusion cannot be false if all the premisses arc true. In contrast the conclusion of a sound INDUCTION is supported by the premisses and may be very probable given the premisses, but it can be false even if all the premisses are true.”
Here is Mautner's (2000: 273) discussion of induction:
“induction n. inference from a finite number of particular cases to a further case or to a general conclusion. For instance if a number of ravens have been observed all of which are black and if no raven has been encountered that is not black the inferences to the conclusion that the next observed raven will be black or to the general conclusion that all ravens are black are inductive inferences. Many inductive inferences seem plausible some indeed seem extremely plausible but the truth of all the premisses can never guarantee the truth of the conclusion since the conclusion goes beyond what is given in the premisses. In this respect they are contrasted with deductive inferences in which the truth of all the premisses guarantees the truth of the conclusion. Aristotle introduced the concept of induction in the Posterior Analytics. It has been claimed, however, that there ‘induction’ does not mean a process of reasoning but the examination of instances that results in a common feature coming to view.”
To summarise, valid deductive conclusions drawn from correct premisses lead to correct conclusions. It is nevertheless possible to have *logically valid* but *incorrect* conclusions. Lewis Carroll’s playful logic games offer delightfully silly examples of logically correct statements that are nevertheless false of silly.
One may reach inductive conclusions, but there is no way to render induction universally reliable in the same way that deduction is universally reliable.
Abduction is useful, but an abductive inference may useful without being correct. One valid use of abduction is that it may be a step in ruling conclusions out rather than in reaching them.
Designers who solve problems do best when they use multiple methods to reach conclusions, and they reach conclusions best when they test and examine both premisses and outcomes.
...
References
Douven, Igor. 2011. “Abduction." The Stanford Encyclopedia of Philosophy (Spring 2011 Edition). Edward N. Zalta, ed. URL:
http://plato.stanford.edu/archives/spr2011/entries/abduction/
Mautner, Thomas. 2000. The Penguin Dictionary of Philosophy. London: Penguin Books.
Peirce, C. S. 1998. The Essential Peirce. Selected Philosophical Writings. Volume 2. 1893-1913. Edited by the Peirce Edition Project. Bloomington, Indiana: Indiana University Press.
Smith, Robin. 2015. "Aristotle's Logic." The Stanford Encyclopedia of Philosophy (Summer 2015 Edition). Edward N. Zalta, ed. URL:
http://plato.stanford.edu/archives/sum2015/entries/aristotle-logic/
—
-----------------------------------------------------------------
PhD-Design mailing list <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------
|
