Text of the presentation by Antonio Dias de Figueiredo at the Workshop on Philosophy and Engineering, Royal Academy of Engineering, London, November 10-12, 2008. This text complements the slides with the same title available at SlideShare.

1. 2008 Workshop on Philosophy and Engineering, The Royal Academy of Engineering,
London, November 10-12, 2008
Toward an Epistemology of Engineering
Antonio Dias de Figueiredo
Centre for Informatics and Systems, University of Coimbra
Coimbra, Portugal
adf@dei.uc.upt
The design dimension sees engineering as the art of design. It
Keywords values systems thinking much more than the analytical thinking
Constructivism, design, critical discussion, engineering, that characterizes traditional science. Its practice is founded on
epistemology, methodology, ontology, philosophy of holistic, contextual, and integrated visions of the world, rather
knowledge, positivism, transdisciplinarity, wicked problems. than on partial visions. Typical values of this dimension include
exploring alternatives and compromising. In this dimension,
1. INTRODUCTION which resorts frequently to non-scientific forms of thinking, the
Although engineering is considered, today, as clearly distinct key decisions are often based on incomplete knowledge and
from science, the predominance of the components of basic intuition, as well as on personal and collective experiences.
science in the education of engineers implicitly contributes to The fourth mode views engineering as the art of getting things
convey the idea that engineering is, in essence, little more than done, valuing the ability to change the world and overcoming
the mere application of the exact and natural sciences to the complexity with flexibility and perseverance. It corresponds to
reality of practice. To help challenge this vision and contribute the art of the homo faber, in its purest expression, and to the
to a reflection on the epistemology of engineering, we propose ability to tuck up one’s sleeves and get down to the nitty-gritty.
a model where engineering is seen as developing in four In this dimension, the completed job, which stands before the
dimensions linked in a transdisciplinary relationship. We then world, leads to higher recognition.
articulate this model with the four key questions of the
philosophy of knowledge [1][2] to clarify the nature of this 3. A TRANSDISCIPLINARY APPROACH
relationship and illuminate some distinctive attributes of If we look at the aggregation of the four dimensions as an
engineering knowledge. exercise in transdiscipliarity, as defined by Gibbons et al. [3],
we may see engineering as resulting from the mutual
2. FOUR DIMENSIONS interpenetration of the epistemologies of the four dimensions in
In the discussion of engineering knowledge it is helpful to think the context of disturbances that shake up the corresponding
of engineering as comprising four major dimensions (Fig. 1): systems of knowledge production. This agrees with the
the dimensions of the basic sciences, of the social sciences, of understanding of transdisciplinarity as the continuous linking
design, and of practical accomplishment. This lets us think of and re-linking, in specific clusterings and configurations, of
the engineer as a professional who combines, in variable knowledge that is brought together on a temporary basis in
proportions, the qualities of a scientist, a sociologist, a designer, specific contexts of application, which makes it strongly
and a doer. oriented to, and driven by, problem-solving [3].
If we now take the epistemological traditions of each one of the
four dimensions, we are led to acknowledge a likely positivist
contribution from the epistemologies of the basic sciences.
Identically positivist dominance can generally be recognized in
the epistemological dimension of the social sciences, although
the adoption of constructivist approaches in this dimension is
gaining ground. Design brings to our epistemological cluster
the most challenging contribution, as we will briefly discuss in
the next section. Finally, although the epistemology of practical
realization tends to be less contemplated in the literature, its
constructivist nature is strongly supported by the tradition of
pragmatist philosophers, the works of Schön [4] and his
followers, and the contributions by Mintzberg [5], Ciborra [6]
Fig. 1 – The four dimensions of engineering. and many others to the theorization of crafting and bricolage.
The dimension inspired by the basic sciences views engineering
as the application of the natural and exact sciences, stressing the 4. EPISTEMOLOGY OF DESIGN
values of logics and rigour, and seeing knowledge as produced The epistemology of design has been suffering a dramatic
through analysis and experimentation. Research is the preferred evolution since the positivist scientization of design introduced
modus operandi of this dimension, where the discovery of first by the ‘modern movement of design’, in the early 1920s. It then
principles is seen as the activity leading to higher recognition. witnessed the backlash of the 1970s, against the science-
The social dimension of engineering sees engineers not just as inspired design methodologies and the claim that the
technologists, but also as social experts, in their ability to epistemology of science was in disarray and had little to offer to
recognize the eminently social nature of the world they act upon an epistemology of design, that there were forms of knowledge
and the social complexity of the teams they belong to. The peculiar to the awareness and ability of the designer, and that
creation of social and economic value and the belief in the we should rather concentrate on the ‘designerly’ ways of
satisfaction of end users emerge as central values in this knowing, thinking and acting [7]. More recently, the troubled
dimension of engineering. relationship between science and design seems to have started
94
Electronic copy available at: http://ssrn.com/abstract=1314224

2. to head toward reconciliation, with the recognition that the axiological question (which includes the ethical question),
epistemology of design is, indeed, different, and has much to inquires about the worth and value of engineering knowledge.
contribute to a renewed epistemology of science [8].
The talk answers these questions in the context of the proposed
This view, which expresses the transdisciplinary linking and re- model. It also stresses the key distinctive features of
linking fields that are closely related, incorporates the ability to engineering knowledge that emerge from the strong presence of
take into account ‘wicked problems’. ‘Wicked problems’ are a design dimension. This includes the importance attached to
problems that, because of their complexity and close abductive reasoning and the acceptance of courses of action that
interdependence with social and organizational factors, cannot seize upon chance information, adopt capricious ideas, and
be formulated [9]. To deal with wicked problems, which are provoke creative leaps that seem to go against traditional
becoming increasingly common in engineering, in spite of the scientific rigour [8]. In this respect, Popper’s concept of
fact that they cannot be handled through traditional scientific ‘critical discussion’ [13] will be used to illustrate how the
approaches, the process of solving a problem becomes identical epistemology of engineering can derive final and verifiable
with the process of understanding its nature, so that problem rigour from such apparently unsystematic, imprecise, and even
understanding and problem resolution are concomitant, with the random, intermediate steps [8].
information needed to understand the problem depending on the
designer’s ideas for solving it. 6. REFERENCES
Important contributions to this debate have been developing [1] Guba, E. G. & Lincoln, Y. S. (1994). Competing
recently in the information systems field, where the evolution of paradigms in qualitative research. In K. D. Denzin & Y.
systems design has been described as incorporating four S. Lincoln. Handbook of Qualitative Research. Thousand
categories: design as functional analysis, design as problem- Oaks, CA: Sage Publications.
solving, design as problem-setting, and design as emergent [2] Lincoln, Y. S., & Guba, E. G. (2000). Paradigmatic
evolutionary learning [10]. controversies, contradictions, and emerging confluences.
Design as functional analysis assumes requirements to be fully In N. K. Denzin & Y. S. Lincoln, eds. Handbook of
available at the outset, so that the designer just needs to analyze Qualiative Research, 2nd edition, Thousand Oaks, CA:
the problem and deductively proceed to the solution, following Sage Publications
a path closely inspired by the traditional basic sciences [8]. [3] Gibbons, M., Limoges, C., Nowotny, H., Schwartzman,
Design as problem-solving resolves complex, namely S., Scott, P. & Trow, M. 1994 The New Production of
organizational, problems by simplifying them to a level where Knowledge: The Dynamics of Science and Research in
they can still satisfy a minimal set of criteria leading to their Contemporary Societies. Sage Publications.
rational solution [8]. This category of design is inspired by [4] Schön, D. 1983 The Reflective Practitioner: How
Herbert Simon’s concept of “bounded rationality” [11], which Professionals Think in Action. Basic Books.
reflects an epistemological standing closer to some popular
visions of the social sciences. [5] Mintzberg, H. 1987 Crafting strategy. Harvard Business
Review. 66-75.
Design as problem-setting views design as a systemic activity
requiring the discovery and possible negotiation of unstated [6] Ciborra, C. U. 1998. Crisis and foundations: an inquiry
goals, implications, and criteria before a problem can be into the nature and limits of models and methods in the
formulated and, subsequently, solved [8]. By accepting the information systems discipline. The Journal of Strategic
framing of problems in terms of their context, before they can Information Systems, 7(1), 5-16.
be solved, this vision of design takes a phenomenological [7] Cross, N. 2001 Designerly ways of knowing: design
approach that expresses a constructivist epistemology. discipline versus design science. Design Issues. 17(3).
Design as emergent, evolutionary, learning sees design as the Massachusetts Institute of Technology.
convergence of problem and solution in an emergent process of [8] Figueiredo, A. D., and Cunha, P. R. 2006 Action research
learning about a situation and then planning short-term partial and design in information systems: two faces of a single
goals that emerge as the process progresses [10][12]. Aspects of coin. In Kock, N. (ed.) Information Systems Action
the solution are thus explored in conjunction with aspects of Research: An Applied View of Emerging Concepts and
problem understanding: not only the problem is unclear at the Methods. Springer.
start of the process, but the goals of the design are also ill-
[9] Rittel, H., & Webber, M. 1973 Dilemmas in a general
defined [10]. Design, emerging, in this case, in multiple circular
theory of planning. Policy Sciences. 4, 155-69.
references, linking problem formulation and problem solution,
explicitly emphasizes the constructivist nature of this approach. [10] Gasson, S. (2004) Organizational ‘problem-solving’ and
theories of social cognition (working paper).
5. EPISTEMOLOGY OF ENGINEERING http://www.cis.drexel.edu/faculty/gasson/Research/Proble
m-Solving.html.
Taking as a reference the proposed four-dimensional model and
the epistemology of design briefly discussed in the previous [11] Simon, H. A. 1973 The structure of ill-structured
section, the remainder of the talk analyses the epistemology of problems. Artificial Intelligence. 4, 181-201.
engineering in light of the four key questions of the philosophy [12] Suchman, L. 1987. Plans and Situated Action. Cambridge
of knowledge [1][2]: the ontological, the epistemological, the University Press.
methodological, and the axiological questions. For the case of
engineering, the ontological question inquires about what [13] Popper, K. (1994). Models, instruments and truth: the
reality can engineering know, the epistemological question status of the rationality principle in the social sciences. In
looks into what is engineering knowledge, the methodological The Myth of the Framework: In Defense of Science and
question asks how can engineering knowledge be built, and the Rationality. London: Routledge, 154-184.
95
Electronic copy available at: http://ssrn.com/abstract=1314224