FOREWORD

Peirce belongs much more to the 20th Century than to the 19th. The French audience must be informed of this fact.
Christiane Chauviré, Peirce et la signification (PUF, 1995)

  Conventional wisdom recommends never having too many irons in the fire. But it is the risk we take in this book and it ought to be explained. The gathering, under a single topic, of concerns from Software Engineering on one side, and Semiotics on the other, may surprise the reader who is a specialist in either of these areas. Furthermore the linkage can be questioned because, despite a few exceptions, the computer scientist does not see himself as a semiotician and the reverse is also true. However, they have in common their interest in the issue of Design, whether in the philosophical sense of ordinary conception, or in the sense of technical conception in the engineering sciences, or in the theoretical sense of conception in logic. More than thirty years ago Herbert Simon in the Sciences of the Artificial (MIT Press, 1969) called for such a Science of Design. But the discipline still seems not to have retained all the interest it deserves.

  We began with the practice and teaching of software design in the domain of computer systems engineering. The first chapter presents some observations of practical nature about the work of design engineers, work focused on the task at hand in a corporate environment. This work is also characterized by a strong dependence on available technologies as well as their evolutions. It is commonplace to point out that the rate of technological change is fast and deemed unpredictable. We show however there are discernable trends that explain some radical changes in ways of thinking about the objectives, methods, and working tools themselves. Within these trends two points deserve to be particularly noticed. The first is the increasingly collaborative nature of these engineering activities in teams where the hierarchy as well as the division of functions fades. The new generic function of “developer” has gradually replaced the old specialized jobs of project manager, system analyst, application analyst, and programmer.

  At the same time when technological change multiplies the complexity of their work, corporate computer people are faced with another kind of cooperation with their environment: the improperly named “users” of computer systems. The usual image of the ordering person or sponsor has turned into that of a partner, from the business requirements to the evaluation of the adequacy of the software to the requirements through all the middle stages of design. Thus, many technical skills that were part of the reserved area of the computer engineer have been transferred to the users themselves, such as creating a database for example. The image of the computer engineer face to face with a machine and its programs is transformed gradually into that of a psychoanalyst, listening simultaneously by several couches. Overall, the issue of communication has erupted in computer systems, but in a way that goes far beyond the technical applications of the Internet or local networks. The other aspect concerns the emerging working tools as well as ways of doing things. Instead of small working communities steeped in their own corporate culture often shaped by the priorities of hardware manufacturers, there are now standards and standardization of fact whose vocation is immediately global. Real institutions such as the Object Management Group or the World Wide Web Consortium have invested themselves the power to establish the rules on the subject, standards that go well beyond the usual technological aspects. Beside the issue of communication, information is involved in the shared production of information and knowledge as evidenced by the movement and community of free software.

  The proliferation of phenomena can be bewildering: does it call for some scientific account and if so which one? Our response is deliberately positive, provided we place it on the right level. We consider the activities of the computer engineer as a living laboratory in which such a science, not yet qualified, is in progress: a laboratory whose test tubes and microscopes are curiously named XML, JAVA, HTTP, UML, etc. Such a science that remains to be built must be based on theories that themselves remain to be constructed. Yet the semblance of novelty of the phenomena should not conceal the fact that they could be grounded on rules already known, at least in part, in other places, and possibly under different headings. This belief became increasingly essential in the development of our teaching experience in Software Engineering Design, be it said that the ability to be taught is a criterion for the scientific nature of a discipline. We spare the reader the long list of undoubted practical connections between the discipline of Design and Semiotics. We shall mention only one example here. It concerns an experience of setting up a computer application aimed at customer profiling in a large bank. The base material consisted of scattered draft notes, made by employees during individual meetings with clients whom they were responsible for monitoring. The systematization of the procedure for the establishment of a computer system led to turning what was seen by all as a fact denoted by the term appointment into a concept called mission. Simple name change, one would say, but a change that changed everything in rendering it possible to assign properties to the fact itself. After this operation, the view of their own work by the employees had been deeply transformed. From the perspective of design techniques we had at the time, and because lack of a better words, such metamorphoses are illustrated by means of an approximate Chinese proverb: “A small association can become a large entity provided God gives it life”, signifying at the same time some impenetrable character of the problem. Studying Peirce years later, we discovered a precise formulation of the transformation, which he called hypostatic abstraction. One example proves nothing, as is well known. However, when the number of examples of the same kind comes to grow repeatedly, when they are attested to from several viewpoints and it is possible to verify them, we are forced to admit that behind them there is something more general. It is just at this point that the engineer must become a man of science. Such a science, which is the missing link between the formation of concepts and information through signs is, in our opinion, Logic. But the question of what kind of logic it can be is actually the subject of this book.

  Such are the reasons that led us to undertake the study, if not exhaustive, at least thorough, of scientific semiotics of which Peirce is the founder, a work that we believe unmatched to this day. These reasons should be given and, of course, it will be for the reader to judge their relevance. Throughout this study we discovered the deep present-day pertinence of his thought to the point that the assessment of Christiane Chauviré quoted in epigraph may stand witness to an error of one century. The rediscovery of Peirce’s work in the second half of the 20th Century by philosophers has not been been followed in the same manner by logicians. Yet the current pertinence of his logical system, which he called Semiotics, makes a serious bid for the development of logic at the dawn of the 21st Century. It is the idea of this topicality that we wanted to emphasize in this book. Peirce is indeed less a precursor in logic than a thinker for tomorrow. Although the purpose is centered on the issue of the sign and its relationship to information with which we are first concerned, we mention here and there the obvious connections with linguistics and more broadly with cognitive sciences in the whole.

  The second chapter is devoted to the presentation of the logical definition of the sign and its essential philosophical underpinnings. These are the least badly known aspects of Peirce, but they could appear totally obsolete in the eyes of many contemporary logicians. Indeed it is a logic that we can describe as “natural”, based on proposition and predicate. But its interaction with the philosophical part of Peirce’s work confers upon this logic a special meaning. It will hardly be grasped if it is looked at, as it usually is, through the generally accepted ideas in the field and evaluated through the logical positivist approach that dominated the 20th century. Peirce is not a logician of objects but a logician of the way things behave, that is to say of the method. He is less a logician aiming at exact structures, although he does make classifications determined by relations of order, than a logician of the forms of processes. For example, his analysis of the proposition is of interest in its ability to illuminate the way propositions can be joined to form an argument, that is to say to account for the validity of arguments and their degree of argumentative force. If he does not neglect the logic of classes, it is second to his logic of relations because he raises the primacy of the relationship on its related objects. You will not find in Peirce’s work any formal language except a few signs of his own which he called “notations”. On the contrary all his work is based on natural-language definitions, following medieval (even ancient) logicians’ usage. Yet he had a very clear idea of the role of mathematics in this logic and never confused the two disciplines which he saw as mutually dependent on each other, that is to say, in relationship. It could be argued that Peirce is an “experimental logician”, a qualifier that is sure to sound like an oxymoron to the ears of contemporary logic, but the reason of which lies in his philosophical conception of truth, known as pragmatism. Indeed, his system — because it is really a systematic thinking — determines itself in the articulation of a logic that goes from Aristotle to the scholastics with the new logic that is required by the tremendous development of the methods of the sciences of nature and life throughout the 19th Century. While deductive reasoning had been widely studied as a syllogistic form since ancient times, the scientific revolution of the 17th Century (Galileo, Kepler, Newton) opened a new field, that of induction. The need to link the two forms together will lead Peirce to invent a third one, the logic of hypothesis or abduction.

  In a way this original logic assumes a reading conceptually independent from theories that have succeeded it in the field if we really want to evaluate the contribution. Unfortunately, the report that we are often given takes such an approach rarely and leads not only to putting Peirce’s work in a philosophical framework that was not his own, but also mostly to ignoring its main interest: the logic of the movement of signs he called “semiosis”. This is the question to which we have devoted the third chapter, a point on which his conception of the relationship between mathematics and logic is essential. We emphasize three important aspects. The first concerns the place assigned to the category of Virtuality, a precondition of semiotic processes which he called the foundation or “ground” of the sign. The second concerns the nature of these processes which cannot correctly be grasped without resorting to his “doctrine” of continuity, and without which the well-known thesis of the infinite sequence of interpretants of the sign seems a mere description, at best a poetic one. One more time the logic of relations is involved here, as well as Peirce’s logical and philosophical conception of time in order to determine the nature of the relation between the continuous and discrete.

  The third aspect is the method, provided that the method by which we think is also the method of reproduction of the signs. This method which will be condensed in the pragmatist maxim permits one to highlight that, if the analysis is a key moment of scientific activity, it needs its complement, the synthesis moment. As always with Peirce, the activity itself is controlled by the logical nature of the dynamic relationship that mediates these two moments. A funny formulation — although very serious — is given with felicity by the author of a famous television cartoon, J. Rouxel in The Shadocks: “It’s much more interesting to look at where we’re not going, for the reason that wherever we go, there will be time enough to look when we have gotten there.” The first step is to eliminate the places where we do not go, it is the analysis of the conditions of possibility of something. The second step is the determination of the places that we propose to move into and rests clearly on the first step. The third step is the future meeting, “when we have gotten there”, as well as the experimental test. It is only upon this third step that some truths will emerge as to the results of the method, a degree of approximation to which what was logically anticipated has been more or less achieved. What is possible — something as vague as it is broad — is analytically obtained by its difference from what is not possible. This explains why Peirce often uses the method of reduction to absurdity in his demonstrations. This procedure may appear outdated too, in that it fails to state some positive truths. However, it allows for saving the essentially axiomatic assumption of the existence of a relationship of access to possible worlds, a relationship highlighted by contemporary modal logics. Peirce indeed poses conversely the prerequisite that “we have no conceptions of the unknowable.”

  The fourth chapter takes up the question of the sign at the other end or extremity, if we may say so, the issue of the “contents of consciousness” as they are called today; not the foundations and the characterization of the sign relation, but the phenomenology of signs mixing and variety, those signs as they present themselves to the mind which Peirce called “phanerons”. But this other end, one of those places where one anticipated going, is deeply rooted for Peirce in the first end. We develop the argument that these new places are given in an inverted, or reflected, form because of the folded nature of the signs when they are actualized in semiosis. This place, methodologically aimed at by any research into the logic of semiotic facts, is the place where the phenomenon of design is engraved as well as the phenomenon of information. It is also the place assigned to meta-signs that are encountered in abundance in computerized information technologies. Obviously, the question of the mode of connection of both temporal extremities arises. We propose a hypothesis on this point, which to our knowledge has not been explicitly raised in the Peirce research community. It is relative to the status and organization of the second division of signs developed by their author after 1904; and especially its relation to the first classification. It then appears that the first classification should be included into the second, the latter providing a development of the previous for experimentation needs. If this hypothesis were confirmed it would be clear that the Peircean “Second Way of Dividing Signs” is the only truly operational tool for the scientific study of particular semiotics which Peirce put in order in the special sciences, including the study of interpretive practices in context.

  The final chapter undertakes such an experimental work in the case of a particular variety of mostly iconic signs, namely diagrams, in current use in Design. In response to questions from the first chapter, we propose a first step towards what could be in the future an expanded science of design (a theoretical as well as an engineering one). This implies that the theory of the singularity of the diagram — due to its alleged specific visual properties — be abandoned. We think that this thesis is in fact at the source of difficulties for a scientific account of the activity of design. Here Peirce’s semiotics offer a way to consider the problem in terms of reasoning and diagrammatic processes, which, furthermore, are observable beings in the diagrams produced by the engineers. In such a problematic, the form of abduction, understood as a method which encompasses the other two — deduction and induction — appears to be the way to be explored.