Posted: February 26th, 2023
Philosophy of Science and Philosophy of Technology
Please use these questions and the included articles, please make it about 700 words long. Please paraphrase instead of using quotation marks.
a) How do Skolimowski and Bunge define technology?
b) What are their criteria in differentiating between science and technology?
c) Explain the main points of both authors when they try to find the relationship between science and technology.
d) What happens in design process?
The Structure of Thinking in Technology
Author(s): Henryk Skolimowski
Source: Technology and Culture, Vol. 7, No. 3 (Summer, 1966), pp. 371-383
Published by: The Johns Hopkins University Press on behalf of the Society for the History
of Technology
Stable URL:
http://www.jstor.org
/stable/3101935
Accessed: 29/12/2009 17:45
Your use of the JSTOR archive indicates your acceptance of JSTOR’s Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR’s Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=jhup.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Society for the History of Technology and The Johns Hopkins University Press are collaborating with JSTOR
to digitize, preserve and extend access to Technology and Culture.
http://www.jstor.org
http://www.jstor.org/stable/3101935?origin=JSTOR-pdf
http://www.jstor.org/page/info/about/policies/terms.jsp
http://www.jstor.org/action/showPublisher?publisherCode=jhup
The Structure of Thinking
in Technology
HENRYK SKOLIMOWSKI
Inquiry into the philosophy of technology, due to the infancy of the
subject, must start with some reflections on what technology itself is.
There is at present a tendency to identify technology with a demiurge
of our times, or perhaps even with a Moloch who will bring doom to
mankind, that is, mankind as dreamt of by philosophers, not by organi-
zation men. In this setting technology assumes a role similar to that
which was ascribed to history in the nineteenth century: the role of the
final cause which shapes the destiny of mankind and, more specifically,
which aims at the total subjugation of man to the machine or, in other
words, at turning the human being into a technological component.
It cannot be denied that reflections on technology in this fashion are
philosophical reflections and that consequently they belong to some
system of the philosophy of technology. At this point, however, a vital
distinction should be made between a philosophy of technology and a
technological philosophy. The former belongs to the realm of epistemo-
logical inquiry and attempts to situate technology within the scope of
human knowledge; the latter belongs to the realm of sociology, broad-
ly conceived, or social philosophy, and is concerned primarily with the
future of human society.
Those who prophesy that our civilization will be devoured by the
Moloch of technology are expanding a certain vision of the world, are
viewing the world through technological lenses, are attempting to
establish a new kind of monism, the technological monism, in which the
technological order is shown to be the prime mover and the ultimate
justification of other orders, moral, aesthetic, cognitive, social, and polit-
ical. The articulation of this technological philosophy is perhaps most
important from a social point of view-as a way of alerting us to the
dangers of technological tyranny. However, for the time being this
technological monism, or whatever name is given to this sociohistorical
prophecy, is but a prophecy. As important as it may be from a human
DR. SKOLIMOWSKI, a philosopher of science and technology, is at the School of
Philosophy of the University of Southern California.
371
372 Henryk Skolimowski
point of view, it cannot serve as a substitute for a philosophy of tech-
nology proper, that is, for a philosophy that aims at the investigation of
the nature and structure of technology, conceived as a branch of human
learning and analyzed for its cognitive content.
I shall not be concerned here with the transformation of society by
technology. It seems to me that the “monolithic technical world” is but
a graphic and perhaps fearsome expression, but not reality. For the time
being the evidence that technology pervades the totality of human rela-
tionships is rather slim. In the realm of art, for example, modern tech-
nology perpetuates at least some traditional human values. The unprece-
dented spread of superb reproductions of the great masters, the easy
availability of the finest recordings of music of the last five centuries, the
spectacular rise in the production and distribution of paperback books,
are all due to the advances of technology, and all serve, at least in part,
the cause of highbrow culture, not technological culture.
It may be that a comprehensive philosophy of technology should in-
clude the moral implications of technological progress. It may be, as
some philosophers insist, that, in spite of the semiautonomous develop-
ment of technology, a substantial part of modern technology is moved
by non-technological forces, that, for example, motor cars are produced
in order to make money, intercontinental missiles in order to kill
people. Consequently, a comprehensive treatment of the philosophy of
technology must examine the presuppositions lying at the foundation of
these technological “events” and must attempt to assess their implica-
tions for mankind at large. The weight of these problems cannot be
underestimated. However, they are outside the scope of my consider-
ations.
In this paper I shall be concerned with what I call the philosophy of
technology proper, that is, with the analysis of the epistemological status
of technology. Technology is a form of human knowledge. Epistemol-
ogy investigates the validity of all human knowledge, its conditions,
its nature. Therefore, it is the business of epistemology to investigate
the peculiarities of technology and its relation to other forms of human
knowledge. In particular, it is of crucial importance to analyze the
relationship of technology to science. I shall argue in the course of
this paper that: (1) it is erroneous to consider technology as being an
applied science, (2) that technology is not science, (3) that the differ-
ence between science and technology can be best grasped by examining
the idea of scientific progress and the idea of technological progress.
In the following sections I shall attempt to provide a basis for a
philosophy of technology rooted in the idea of technological progress.
Then I shall proceed to show that in various branches of technology
The Structure of Thinking in Technology
there can be distinguished specific thought patterns which can be seen
as explaining technological progress.
* * *
Many methodologists and philosophers of science insist that technol-
ogy is in principle a composition of various crafts. Regardless of how
sophisticated these crafts may have become, they are still crafts. It is
argued that technology is methodologically derivative from other
sciences, that it has no independent methodological status, and that what
makes it scientific is the application of various other sciences, natural
sciences in particular. Thus, the scientific part of technology can be de-
composed into particular sciences and accounted for as physics, optics,
chemistry, electromagnetics, etc. This view misconstrues the situation
because it does not take into account the idea of technological progress.
My thesis is that technological progress is the key to the understand-
ing of technology. Without the comprehension of technological prog-
ress, there is no comprehension of technology and there is no sound
philosophy of technology. Attempts that aim at reducing technology to
the applied sciences fail to perceive the specific problem situation in-
herent in technology. Although in many instances certain technological
advancements can indeed be accounted for in terms of physics or
chemistry, in other words, can be seen as based on pure science, it should
not be overlooked that the problem was originally not cognitive but
technical. With an eye to solving a technical problem, we undertake in-
quiries into what is called pure science. Our procedures are extremely
selective. Out of infinitely many possible channels of research only very
few are chosen. Problems thus are investigated not with an eye to in-
creasing knowledge but with an eye to a solution of a technical problem.
If it were not for the sake of solving some specific technological prob-
lems, many properties of physical bodies never would have been
examined, and many theories incorporated afterward into the body of
pure science never would have been formulated. Perhaps the most ob-
vious examples can be found in the sciences of electronics and of space
physics. The development of computors resulted in the replacement of
tubes by transistors. In developing transistors many properties and laws
governing the behavior of semiconductors have been formulated which
might never have been formulated otherwise. To take another example,
the problem of metal fatigue and many other phenomena concerning the
behavior of solids in space might never have been investigated, and
theories resulting from them might never have been established if it
were not for the sake of constructing supersonic planes and intercon-
373
374 Henryk Skolimowski
tinental rockets. To mention finally atomic physics, it was in the Man-
hattan Project where plutonium, an element not found in nature, had to
be developed in the process of producing the atom bomb. Thus, in one
sense science, that is pure science, is but a servant to technology, a char-
woman serving technological progress.
* * *
I shall now discuss the thesis that technology is not science. By this
statement I mean to say that the basic methodological factors that ac-
count for the growth of technology are quite different from the factors
that account for the growth of science. Consequently, the idea of tech-
nological progress as contrasted with scientific progress must be
examined more carefully.
I am in full agreement with Karl Popper that science, in order to exist,
must progress; the end of scientific progress is the end of science. This
progress results from the continuous improvement of scientific theories
and constant enlargement of the scientific store; more precisely it results
from a permanent overhaul of theories and incessant replacement of
worse theories by better ones; “better” means simpler, or more uni-
versal, or more detailed, or of greater explanatory power, or all these
things together. The objective underlying this endless succession of
theories is the increase of knowledge. The pursuit of knowledge
(which is another expression for the pursuit of truth) has been and still
is the most important aim of science. We critically scrutinize our
theories by devising tests of increasing ingenuity and severity in order to
learn how squarely they can face reality. Whatever operationists and
conventionalists of various denominations may say, science is about real-
ity. The acquisition of knowledge and the pursuit of truth are only pos-
sible if there is reality. Thus it is contained in the idea of scientific prog-
ress that we investigate reality and that we devise theories of increasing
depth in order to comprehend this reality.
What about technology? Is it another instrument for investigating
reality? Does it aim at the enlargement of knowledge and the acquisition
of truth? The answer is negative in both cases. Hence we come to
significant differences between science and technology. In science we
investigate the reality that is given; in technology we create a reality
according to our designs. In order to avoid confusion I should perhaps
say at once that these two kinds of reality are not of the same order. To
put it simply, in science we are concerned with reality in its basic mean-
ing; our investigations are recorded in treatises “on what there is.” In
technology we produce artifacts; we provide means for constructing
The Structure of Thinking in Technology
objects according to our specifications. In short, science concerns itself
with what is, technology with what is to be.
The growth of technology manifests itself precisely through its abil-
ity to produce more and more diversified objects1 with more and more
interesting features, in a more and more efficient way.
It is a peculiarity of technological progress that it provides the means
(in addition to producing new objects) for producing “better” objects
of the same kind. By “better” many different characteristics may be in-
tended, for example: (a) more durable, or (b) more reliable, or (c)
more sensitive (if the object’s sensitivity is its essential characteristic), or
(d) faster in performing its function (if its function has to do with
speed), or (e) a combination of the above. In addition to the just-men-
tioned five criteria, technological progress is achieved through shorten-
ing the time required for the production of the given object or through
reducing the cost of production. Consequently, two further criteria are
reduced expense or reduced time, or both, in producing an object of a
given kind.
It hardly could be denied that contemporary freeways and highways
mark a technological advancement in terms of durability when com-
pared with Roman or even nineteenth-century roads; that modern
bridges are far more reliable (in addition to other advantages) than
bridges of previous centuries; that photographic cameras installed in
artificial satellites are considerably more sensitive (in addition to being
more reliable and more durable) than those used in the pre-Sputnik age;
that the speed of jet airplanes makes them superior to the planes of the
brothers Wright. And no one can deny that if the same plane or bridge
or camera can be manufactured less expensively, or alternatively in
shorter time (at the same expense), then it will equally mean a tech-
nological advancement.
The criteria of technological progress cannot be replaced by or even
meaningfully translated into the criteria of scientific progress. And, con-
versely, the criteria of scientific progress cannot be expressed in terms
of the criteria of technological progress. If an enormous technological
improvement is made and at the same time no increase in pure science is
accomplished, it will nevertheless mark a step in technological progress.
On the other hand, it is of no consequence to pure science whether a
given discovery is utilized or not; what is of significance is how much
the discovery adds to our knowledge, how much it contributes to the
comprehension of the world.
It may be argued that in the pursuit of technological progress we
1 By the “technological object” I mean every artifact produced by man to serve
a function; it may be a supersonic airplane as well as a can-opener.
375
376 Henryk Skolimowski
often bring about scientific progress as well. It should be observed, on
the other hand, that scientific progress may and indeed does facilitate
technological progress. Discoveries in pure science, regardless of how
abstract they appear at first, sooner or later find their technological em-
bodiment. These two observations lead to a conclusion that perhaps
neither scientific nor technological progress can be achieved in its pure
form; that in advancing technology, we advance science; and in ad-
vancing science, we advance technology. This being the case, it should
not prevent us from analyzing these two kinds of progress separately,
particularly because scientific progress is often treated autonomously
and is regarded as the key to an explanation of the growth and nature
of science. If we are permitted to divorce scientific progress from
technological progress when examining the nature of science, we should
be equally permitted to divorce technological progress from scientific
progress when examining the nature of technology.
In this context it is rather striking that even such mature and eminent
philosophers of science like Popper have nothing better to say than to
equate technology with computation rules. Neither Popper nor, to my
knowledge, any other authority in the philosophy of science, has cared
to examine the idea of technological progress. Hence their remarks on
technology, whenever they find it convenient to mention it, are rather
harsh and far from adequate.
To summarize, scientific and technological progress are responsible
for what science and technology, respectively, attempt to accomplish.
Science aims at enlarging our knowledge through devising better and
better theories; technology aims at creating new artifacts through devis-
ing means of increasing effectiveness. Thus the aims and the means
are different in each case.
* * *
The kernel of scientific progress can be expressed simply as being the
pursuit of knowledge. The answer seems to be less straightforward with
regard to technological progress. However, in spite of the diversity of
criteria accounting for the advancement of technology, there seems to
be a unifying theme common to them all, or at any rate into which they
can be translated. This theme is the measure of effectiveness. Techno-
logical progress thus could be described as the pursuit of effectiveness
in producing objects of a given kind.
Now, the question is: Can this measure of effectiveness be studied in
general terms or, to put it differently, can we aim at a general theory of
efficient action and then incorporate it in the idea of technological
progress? And a second question: Is there only one, or are there many
The Structure of Thinking in Technology
different patterns leading to an increase of the measure of effectiveness
in different branches of technology?
In relation to the first question, it should be observed that, in addition
to specific formulas for efficient action constructed for limited scopes of
human activity (e.g., the science of management), there is indeed a
general theory of efficient action for all activities we choose to analyze.
This general theory of efficient action is called praxiology. This theory
has been worked out in detail by the Polish philosopher, Tadeusz Kotar-
binfski. Since the principles of praxiology are treated extensively in
Kotarbiniski’s treatise,2 I shall be very brief here.
Praxiology analyzes action from the point of view of efficiency.
Praxiology is a normative discipline; it establishes values, practical
values, and assesses our action in terms of these values. Practical values
should not be confused with other values, aesthetic or moral. Whether
we are aware of this or not, it is through constructing praxiological
models that we accomplish progress in technology. Progress means an
improvement of the measure of effectiveness in at least one aspect.
Usually the praxiological model assumes some losses in effectiveness in
order to attain more substantial gains. It is sometimes facinating to ana-
lyze how meticulous and impeccable is the calculus of gains and losses
in the praxiological model, which very often is constructed without an
awareness of its praxiological nature.
It seems to me that if the characterization of technological progress as
the pursuit of effectiveness is correct, the philosopher of technology
must include the study of praxiology and in particular the study of
praxiological models in his inquiry. Organization theory is simply inade-
quate for this purpose because of its limited scope. The advances of
modern technology take on a very complex form requiring integration
of a variety of heterogeneous factors as well as the establishment of a
hierarchy of levels. What finally matters is the increased measure of
effectiveness, but the road to this increase is multichanneled and multi-
leveled. Traditional organization theories are unable to handle this com-
plexity, but praxiology can.
* * *
Technological progress, analyzed in terms of measures of effective-
ness, led us to two questions. The first was whether technological
effectiveness can be treated in general terms-this prompted us to con-
sider praxiology. The second was whether we can distinguish specific
2Praxiology-An Introduction to the Science of Efficient Action (London,
1965). See also my article, “Praxiology-the Science of Accomplished Acting,”
Personalist, Summer 1965.
377
378 Henryk Skolimowski
patterns of thinking leading to the increase of effectiveness in different
branches of technology.
I shall devote the remaining part of this paper to the second question.
That is, I shall attempt to discern specific patterns of technological
thinking for some branches of technology. I do not propose to find such
patterns for technology as a whole. What I can offer are some sugges-
tions as to how one may approach the problem and discern these pat-
terns in less complex fields. If the procedure is right, it will lead to the
discovery of other patterns in other branches of technology.
Before I attempt to spell out some of the structures or patterns of
thinking in technology, I shall show what they are and how they work
in microbiology. The microbiologist makes daily observations of micro-
scopic sections which are quite simple from a certain point of view.
Now what is a microscopic section, for example, of a diphtheria culture?
It is, in the layman’s language, a specific configuration of certain forms
which possess characteristic structures. This is how far we can go in de-
scribing the phenomenon verbally. In other words, no amount of verbal
explanation will render it possible for the layman and generally for the
untrained person to recognize the diphtheria culture by mere description.
At first the layman and beginning students of microbiology are simply
unable to perceive what is there to be seen. After some period of train-
ing they do perceive and are in fair agreement as to what they see. The
ability to recognize certain microscopic structures is thus peculiar to
students of microbiology.
The art of observation is not universal but specific for a given field or
subject matter. Whenever observation plays a significant role in scien-
tific investigation, it is selective observation directed toward perceiving
some objects and their configurations and toward neglecting others.
Observation, however, is not only a perceptual process but also involves
some conceptual thinking. Certain types of observation are intrinsically
connected with thinking in terms of certain categories.
In general, it seems to me that specific branches of learning originate
and condition specific modes of thinking, develop and adhere to cate-
gories through which they can best express their content and by means
of which they can further progress. I shall illustrate this thesis by
examining some branches of technology, namely, surveying, civil engi-
neering, mechanical engineering, and architecture, with the understand-
ing that the last, architecture, is only in part a branch of technology.
I will start with surveying. The final products of surveying are maps,
plans, and profiles in elevation. In order to avoid complications in the
analysis, instead of considering a map that is a projection of a larger area
of land on a sphere, I shall examine a plan that consists of a projection of
The Structure of Thinking in Technology
a smaller piece of land on a plane as the referential surface. It is quite
obvious that we must measure all angles of the figure to be projected on
the plane, all its sides, and at least one azimuth. Now, the specific ques-
tions for this surveying operation, and indeed for all geodesy, are: Why
is this method applied, not any other? Why should we measure the sides
with a metal tapeline and not by steps or by eye? Why should we check
and adjust our instruments? A surveyor, who is quite capable of skil-
fully performing all the geodetical operations, might be less capable of
relating all these operations to one theme, one central element that
accounts for the specificity of surveying. It is one thing to follow a
procedure and another to be able to grasp and verbalize the essence of
this procedure or, in other words, to make measurements and to be
aware of the specific structures of thinking characteristic for surveying.
What, then, is specific for thinking in surveying? It is the accuracy of
the measurement. This can be seen while tracing the development of
surveying from its earliest stages as well as while following its recent
progress. In the final analysis, it is always the accuracy that lies at the
bottom of all other considerations. Sometimes it is expressed in an in-
direct and disguised form, for example, when we inquire which of two
or three methods is most economic or most efficient. However, even in
this case, the silent assumption is that the accuracy remains the same or,
at any rate, that the decrease of accuracy is negligible and the economic
gains-which sometimes may be of prime importance-are quite con-
siderable. It is thus the most conspicuous feature of geodesy that it aims
at a progressively higher accuracy of measurement; in an indirect form
this may mean a reduction of cost or time or work while preserving the
same accuracy. Thus, we may say in a succinct form: To think geodeti-
cally is to think in terms of accuracy.
Succinct forms have the virtue that they pin down one crucial ele-
ment of the analysis; they have the vice that (for the sake of brevity)
they neglect other elements and consequently present a simplification
of the phenomenon under investigation. So it-is with our succinct char-
acterization of geodetical thinking. It is by no means the only kind of
thinking the surveyor performs. It is not even the dominant thinking in
terms of the actual time devoted to it. But thinking in terms of accuracy
is the most instrumental for surveying. And that means that the practi-
tioner of surveying will be a better practitioner if he is aware of the
specificity of geodesy and if he applies consistently his knowledge in his
practice. And this also means that the researcher in geodesy will be a
better one if he consistently keeps in mind that geodesy aims at a pro-
gressively higher accuracy of measurement. Furthermore, the grasp of
the specificity of surveying will help the scholar who investigates the
379
380 Henryk Skolimowski
history of surveying. History of any branch of learning is twisted and
full of unexpected turns and blind alleys. Unless we discover the
“Ariadne’s thread” in its development, a history of any discipline will be
but a mosaic of unrelated or loosely related events, descriptions, theo-
ries. Thus, the discernment of patterns specific for a given branch of
learning is not only an activity that may give us the comfort and aes-
thetic satisfaction that accompanies neat classifications for the sake of
classification but may indeed be of a concrete value to the practitioner,
researcher, and historian. It is in these terms that I deem the analysis of
patterns of thinking important.
To return to technology, when we consider a typical civil engineer-
ing project-whether the construction of a house or a bridge-the deci-
sive element is the durability of the construction. Therefore, we may
say that thinking, specific for the civil engineer, is in terms of durability.
Durability is the starting point, or at any rate the ultimate element of
the analysis. The choice of materials and the methods of construction
must be related to the required durability.
Theoretical research in civil engineering is directed toward the dis-
covery of combinations of materials that will either increase the durabil-
ity (of the construction) or lower the costs at the same durability. Dur-
ing the execution of a project, some calculations may be made and the
accuracy of the calculations taken into account, but here they are of
subsidiary importance. The main issue is durability, although admittedly
the form of its manifestation may be very complex or disguised.
Perhaps this can be seen even more clearly when we review the his-
tory of civil engineering or, in other words, when we review the history
of architecture in its constructional aspect. If we omit the aesthetic and
utilitarian aspects, the history of architecture can be seen as the develop-
ment and perfection of those architectural forms and those combina-
tions of materials that increase durability. Although the progression of
more and more durable forms is often hidden under the guise of artistic
trends and movements, it is there and can be traced easily.
Turning now to architecture proper, architectural thinking is simul-
taneous thinking in terms of durability and aesthetics and utility, and
the two latter categories are perhaps more important than the first one.
When projecting a house, the civil engineer must consider new mate-
rials and their combinations as well as new constructional designs. When
designing the same house, the architect must consider the standards of
comfort, hygiene, and, generally speaking, the “livability” prevailing for
his times, as well as the aesthetic tastes of his epoch, its predilections and
aversions. Thus, thinking in terms of utility and artistic predilections
separates the architect from the civil engineer.
The Structure of Thinking in Technology
I shall now very briefly consider mechanical engineering. The key
element in this branch of engineering is efficiency (in the narrow sense
of the term when it refers to the efficiency of an engine, whether steam
or combustion). Thus, thinking, specific for mechanical engineering, is
in terms of efficiency (efficiency here is meant in the narrow sense speci-
fied above). In designing engines, the problem of efficiency has two
aspects: either we attempt to increase the absolute efficiency and raise it
as close as possible to 1, or we attempt to construct a “better” engine
while keeping the same efficiency (“better” can mean: safer, cheaper,
longer lasting, more resistant). Obviously, certain problems concerned
with the strength of materials have to be considered and solved, and
therefore thinking in terms of durability takes place here as well; it is,
however, of a derivative character. By saying it is derivative, I do not
mean to say that it has little significance or no significance at all but,
rather, that the starting point for an analysis of durability are problems
of efficiency. Problems of durability are not chosen at random but are
selected with an eye to the solution of the problems of efficiency.
In considering machine tools, the question of efficiency is not im-
mediately obvious but may be shown to be of crucial importance as
well. A number of other factors, such as the cost of construction, dura-
bility, and useful life, are analyzed at the same time. Finally, we either
attempt to raise efficiency while preserving the same cost, the same use-
ful life, and the same durability; or we attempt to reduce the cost while
preserving the same efficiency, the same useful life, and the same dur-
ability; or to prolong the useful life with the remaining data unchanged.
To summarize, to think in terms specific for a given discipline is to
think in those terms that (a) determine the lines of investigation within
this discipline; (b) account for the historical development of this disci-
pline; (c) explain the recent growth of the discipline.
Once again it should be emphasized that categories specific for various
branches of technology or, more generally, specific for various branches
of learning, are not those that end all but rather those that begin all.
They are the key to the analysis. They are the key to the idea of tech-
nological progress. Neither should it be surmised that categories I call
specific have anything to do with Kantian categories. Perhaps my termi-
nology is unfortunate. My point was simply to draw attention to certain
patterns of thinking which can be discerned as characteristic for various
branches of technology and elsewhere. The most important conceptual
elements in these patterns I call categories.
I should not be surprised if the “categorical” analysis as sketched
here will be viewed as insufficient for an exhaustive epistemological
description of technology. Perhaps it should be remembered that as yet
381
382 Henryk Skolimowski
no general philosophy of science-which after all has been developed for
some centuries-is viewed as sufficient. Can we then expect more from a
subject that is beginning to emerge than we expect from a related sub-
ject that has achieved a considerable maturity?
* * *
The analysis of the structure of thinking in technology is hampered
by the fact that nowadays the construction of bridges, highways, auto-
mobiles, or even domestic gadgets is inseparably linked with the con-
sideration of beauty and comfort which are basically “non-technical”
categories. Technical categories, such as accuracy and durability, are,
so to say, the technological constants. They are the yardstick of techno-
logical progress. Aesthetic satisfaction and comfort are to a certain de-
gree variables. They cannot be measured objectively for all epochs. The
more decisive their influence on the object designed, the more difficult
it is to recapture the structure of thinking peculiar to a given branch of
technology. Architecture again can serve as an example.
Luigi Nervi, Oscar Niemeyer, and Frank Lloyd Wright, among
others, are architects for whom the element of a construction (e.g., the
beam of a house) is often at the same time a component of an over-all
aesthetic pattern. These constructor-architects think at the same time in
terms of durability and in terms of aesthetic satisfaction; they find
aesthetic expression in functional, that is, purely constructional elements.
A similar situation occurs in other domains of technology. While de-
signing and constructing automobiles or lathes, can-openers or inter-
continental ballistic missiles, the purely technical aspects often are inter-
woven with aesthetic and utilitarian aspects. The technological phe-
nomenon no longer is identical with the technical phenomenon and can-
not be analyzed entirely in terms of the engineering sciences. The social
context, the economic structure of a society, the existing social mores
and aesthetic predilections-all have their imprint on the technological
phenomenon and, to a certain extent, determine its character.
* * *
In summary, I should like to observe that mistaken ideas about the
nature of technology reflect what technology was a century or two
centuries ago and not what it is today. In the twentieth century, and
particularly in our day, technology has emancipated itself into a semi-
autonomous cognitive domain. There are many links between science
and technology, but a system of interrelations should not be mistaken
for a complete dependence. A fruitful way of reconstructing the epis-
The Structure of Thinking in Technology 383
temological status of technology is through grasping the idea of tech-
nological progress. Technological progress is the pursuit of effectiveness
in producing objects of a given kind. The purely technical elements,
such as the accuracy or durability of our products, are often considered
in larger economic frameworks which complicate the basically techno-
logical typology and even impede the analysis in terms of purely tech-
nological categories. In addition, the standards of beauty and utility are
becoming intrinsic ingredients of technological products, and this
makes our analysis even more difficult. However, our task is to meet
these difficulties, not to avoid them. The point is that the structure of
technology is far more complex than the methodologist of science is
prepared to admit. It is only through recognizing this complexity, and
through granting to technology a methodological autonomy, that we
may be able to end the stagnation in a field which as yet has only a name
-the philosophy of technology.
- Article Contents
- Issue Table of Contents
p.371
p.372
p.373
p.374
p.375
p.376
p.377
p.378
p.379
p.380
p.381
p.382
p.383
Technology and Culture, Vol. 7, No. 3 (Summer, 1966), pp. 301-451
Front Matter
Towards a Philosophy of Technology
Prefatory Note [pp.301-302]
Technics and the Nature of Man [pp.303-317]
Technology as Skills [pp.318-328]
Technology as Applied Science [pp.329-347]
The Confusion between Science and Technology in the Standard Philosophies of Science [pp.348-366]
The Need for Corroboration: Comments on Agassi’s Paper [pp.367-370]
The Structure of Thinking in Technology [pp.371-383]
The Social Character of Technological Problems: Comments on Skolimowski’s Paper [pp.384-390]
Memorial
Lynn Thorndike (1882-1965) [pp.391-394]
The Cover Design
Dyeing Fabrics in Sixteenth-Century Venice [pp.395-397]
Communications
A Postscript to Reti’s Notes on Juanelo Turriano’s Water Mills [pp.398-401]
Technology, Traditionalism, and Military Establishments [pp.402-407]
The Inaccurate “Slide Rule” [pp.408-409]
Book Reviews
untitled [pp.410-411]
untitled [pp.412-413]
untitled [pp.413-415]
untitled [pp.415-418]
untitled [pp.418-420]
untitled [pp.420-421]
untitled [pp.421-424]
untitled [pp.424-426]
untitled [pp.426-428]
untitled [pp.428-429]
untitled [pp.429-432]
untitled [pp.432-434]
untitled [pp.434-436]
untitled [pp.436-438]
untitled [pp.438-440]
untitled [pp.440-441]
untitled [pp.441-443]
untitled [pp.443-444]
untitled [pp.444-445]
untitled [pp.445-446]
untitled [pp.446-447]
Notes and Announcements [pp.448-451]
Back Matter
Technology as Applied Science
Author(s): Mario Bunge
Source: Technology and Culture, Vol. 7, No. 3 (Summer, 1966), pp. 329-347
Published by: The Johns Hopkins University Press and the Society for the History of
Technology
Stable URL: https://www.jstor.org/stable/3101932
Accessed: 30-09-2019 20:14 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
The Johns Hopkins University Press, Society for the History of Technology are
collaborating with JSTOR to digitize, preserve and extend access to Technology and Culture
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
MARIO BUNGE
The application of the scientific method and of scientific theories to
the attainment of practical goals poses interesting philosophical prob-
lems, such as the nature of technological knowledge, the alleged validat-
ing power of action, the relation of technological rule to scientific law,
and the effects of technological forecast on human behavior. These
problems have been neglected by most philosophers, probably because
the peculiarities of modern technology, and particularly the differences
between it and pure science, are realized infrequently and cannot be
realized as long as technologies are mistaken for crafts and regarded as
theory-free. The present paper deals with those problems and is there-
fore an essay in the nearly non-existent philosophy of technology.
Science: Pure and Applied
The terms “technology” and “applied science” will be taken here as
synonymous, although neither is adequate: in fact, “‘technology” sug-
gests the study of practical arts rather than a scientific discipline, and
“applied science” suggests the application of scientific ideas rather than
that of the scientific method. Since “technique” is ambiguous and “epis-
technique” unborn, we shall adopt the current lack of respect for ety-
mology and go over to more serious matters.
The method and the theories of science can be applied either to in-
creasing our knowledge of the external and the internal reality or to
enhancing our welfare and power. If the goal is purely cognitive, pure
science is obtained; if primarily practical, applied science. Thus, where-
as cytology is a branch of pure science, cancer research is one of
applied research. The chief divisions of contemporary applied science
are the physical technologies (e.g., mechanical engineering), the bio-
logical technologies (e.g., pharmacology), the social technologies (e.g.,
operations research), and the thought technologies (e.g., computer sci-
DR. BUNGE, a theoretical physicist and philosopher of science, was formerly a pro-
fessor at the University of Buenos Aires; during the academic year 1965-66, he has
been visiting at the Institut fur theoretische Physik at the University of Freiburg,
and in 1967 he will be at Yale University. He is the author of Causality, The Myth
of Simplicity, and other books, including a forthcoming two-volume work entitled
Scientific Research.
329
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
330 Mario Bunge
ence). In many cases technology succeeds a craft: it solves some of the
latter’s problems by approaching them scientifically. In other cases, par-
ticularly those of the social and thought technologies, there is no ante-
cedent prescientific skill because the problems themselves are new. But
in every case a distinction must be made between artisanal knowledge
and scientific knowledge, as well as between pure research, applied re-
search, and the applications of either to action.
The division of science into pure and applied is often challenged on
the ground that all research is ultimately oriented toward satisfying
needs of some sort or other. But the line must be drawn if we want to
account for the differences in outlook and motivation between the in-
vestigator who searches for a new law of nature and the investigator
who applies known laws to the design of a useful gadget: whereas the
former wants to understand things better, the latter wishes to improve
our mastery over them. At other times the difference is acknowledged,
but it is claimed that applied science is the source of pure science rather
than the other way around. Clearly, though, there must be some knowl-
edge before it can be applied, unless it happens to be a skill or know-
how rather than conceptual knowledge.
What is true is that action-industry, government, warfare, education,
etc.-often poses problems that can be solved only by pure science. And
if such problems are worked out in the free and lofty spirit of pure
science, the solutions to them eventually may be applied to the attain-
ment of practical goals. In short, practice is one of the sources of scien-
tific problems, the other being sheer intellectual curiosity. But giving
birth is not rearing. A whole cycle must be performed before anything
comes out from practice: Practice -> Scientific Problem -> Scientific
Research (statement and checking of hypotheses) -> Rational Action.
Even so, this is far from being the sole way in which scientific research
and action mingle. Ever since theoretical mechanics began, in the
eighteenth century, to shape industrial machinery, scientific ideas have
been the main motor and technology their beneficiary. Since then, intel-
lectual curiosity has been the source of most, and certainly of all impor-
tant, scientific problems; technology has often followed in the wake of
pure research, with a decreasing time lag between the two.
This is not to debase applied science but to recall how rich its con-
ceptual background is. In applied science a theory is not only the sum-
mit of a research cycle and a guide to further research; it is also the basis
of a system of rules prescribing the course of optimal practical action.
On the other hand, in the arts and crafts theories are either absent or
instruments of action alone. In past epochs a man was regarded as prac-
tical if, in acting, he paid little or no attention to theory or if he relied
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
on worn-out theories and common knowledge. Nowadays a practical
man is one who acts in obedience to decisions taken in the light of the
best technological knowledge-not pure scientific knowledge, because
this is mostly remote from or even irrelevant to practice. And such a
technological knowledge, made up of theories, grounded rules, and data,
is in turn an outcome of the application of the method of science to
practical problems.
Since technology is as theory laden as pure science, and since this
either is overlooked or explicitly denied by most philosophers, we must
take a closer look at technological theories and their application.
Technological Theories: Substantive and Operative
A theory may have a bearing on action either because it provides
knowledge regarding the objects of action, for example, machines, or
because it is concerned with action itself, for example, with the decisions
that precede and steer the manufacture or use of machines. A theory of
flight is of the former kind, whereas a theory concerning the optimal
decisions regarding the distribution of aircraft over a territory is of the
latter kind. Both are technological theories but, whereas the theories of
the first kind are substantive, those of the second kind are, in a sense,
operative. Substantive technological theories are essentially applications,
to nearly real situations, of scientific theories; thus, a theory of flight is
essentially an application of fluid dynamics. Operative technological
theories, on the other hand, from the start are concerned with the
operations of men and man-machine complexes in nearly real situations;
thus, a theory of airways management does not deal with planes but
with certain operations of the personnel. Substantive technological
theories are always preceded by scientific theories, whereas operative
theories are born in applied research and may have little if anything to
do with substantive theories-this being why mathematicians and logi-
cians with no previous scientific training can make important contribu-
tions to them. A few examples will make the substantive-operative dis-
tinction clearer.
The relativistic theory of gravitation might be applied to the design
of generators of antigravity fields (i.e., local fields counteracting the
terrestrial gravitational field), which in turn might be used to facilitate
the launching of spaceships. But, of course, relativity theory is not par-
ticularly concerned with either field generators or astronautics; it just
provides some of the knowledge relevant to the design and manufacture
of antigravity generators. Paleontology is used by the applied geologist
engaged in oil prospecting, and the latter’s findings are a basis for mak-
ing decisions concerning drillings; but neither paleontology nor geol-
331
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
332 Mario Bunge
ogy is particularly concerned with the oil industry. Psychology can
be used by the industrial psychologist in the interests of production, but
it is not basically concerned with production. All three are examples of
the application of scientific (or semiscientific, as the case may be)
theories to problems that arise in action.
On the other hand the theories of value, decision, games, and opera-
tions research deal directly with valuation, decision-making, planning,
and doing; they even may be applied to scientific research regarded as a
kind of action, with the optimistic hope of optimizing its output. (These
theories could not tell how to replace talent but how best to exploit it.)
These are operative theories, and they make little if any use of the sub-
stantive knowledge provided by the physicial, biological, or social sci-
ences: ordinary knowledge, special but non-scientific knowledge (of,
e.g., inventory practices), and formal science are usually sufficient for
them. Just think of strategical kinematics applied to combat or of
queuing models: they are not applications of pure scientific theories but
theories on their own.
What these operative or non-substantive theories employ is not sub-
stantive scientific knowledge but the method of science. They may be
regarded, in fact, as scientific theories concerning action, in short, as
theories of action. These theories are technological in respect of aim,
which is practical rather than cognitive, but apart from this they do not
differ markedly from the theories of science. In fact, every good opera-
tive theory will have at least the following traits characteristic of scien-
tific theories: (1) they do not refer directly to chunks of reality but to
more or less idealized models of them (e.g., entirely rational and per-
fectly informed contenders or continuous demands and deliveries); (2)
as a consequence they employ theoretical concepts (e.g., “probabil-
ity”); (3) they can absorb empirical information and in turn can enrich
experience by providing predictions or retrodictions; and (4) conse-
quently they are empirically testable, though not as toughly as scientific
theories.
Looked at from a practical angle, technological theories are richer
than the theories of science in that, far from being limited to accounting
for what may or does, did or will happen regardless of what the deci-
sion-maker does, they are concerned with finding out what ought to be
done in order to bring about, prevent, or just change the pace of events
or their course in a preassigned way. In a conceptual sense, the theories
of technology are definitely poorer than those of pure science; they are
invariably less deep, and this because the practical man, for whom they
are intended, is chiefly interested in net effects that occur and are con-
trollable on the human scale; he wants to know how things within his
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
reach can be made to work for him, rather than how things of any kind
really are. Thus, the electronics expert need not worry about the diffi-
culties that plague the quantum electron theories; and the researcher in
utility theory, who is concerned with comparing people’s preferences,
need not burrow into the origins of preference patterns-a problem for
psychologists.
Consequently, whenever possible the applied researcher will attempt
to schematize his system as a black box; he will deal preferably with ex-
ternal variables (input and output), will regard all others as at best
handy intervening variabes with no ontological import, and will ignore
the adjoining levels. This is why his oversimplifications and mistakes are
not more often harmful-because his hypotheses are superficial. (Only
the exportation of this externalist approach to science may be harmful.)
Occasionally, though, the technologist will be forced to take up a
deeper, representational viewpoint. Thus, the molecular engineer who
designs new materials to order, that is, substances with prescribed mac-
roproperties, will have to use certain fragments of atomic and molecular
theory. But he will neglect all those microproperties that do not show
up appreciably at the macroscopic level; after all, he uses atomic and
molecular theories as tools-which has misled some philosophers into
thinking that scientific theories are nothing but tools.
The conceptual impoverishment undergone by scientific theory when
used as a means for practical ends can be frightful. Thus, an applied
physicist engaged in designing an optical instrument will use almost only
ray optics, that is, essentially what was known about light toward the
middle of the seventeenth century. He will take wave optics into ac-
count for the explanation in outline, not in detail, of some effects, most-
ly undesirable, such as the appearance of colors near the edge of a lens;
but he will seldom, if ever, apply any of the various wave theories of
light to the computation of such effects. He can afford to ignore these
theories in most of his professional practice because of two reasons.
First, the chief traits of the optical facts relevant to the manufacture of
optical instruments are adequately accounted for by ray optics; those
few facts that are not so explainable require only the hypotheses (but
not the whole theory) that light is made up of waves and that these
waves can superpose. Second, it is extremely difficult to solve the wave
equations of the deeper theories save in elementary cases, which are
mostly of a purely academic interest (i.e., which serve essentially the
purpose of illustrating and testing the theory). Just think of the enter-
prise of solving a wave equation with time-dependent boundary con-
ditions such as those representing the moving shutter of a camera. Wave
optics is scientifically important because it is nearly true; but for most
333
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
334 Mario Bunge
present-day technology it is less important than ray optics, and its
detailed application to practical problems in optical industry would be
quixotic. The same argument can be carried over to the rest of pure
science in relation to technology. The moral is that, if scientific research
had sheepishly applied itself to the immediate needs of production, we
would have no pure science, hence no applied science either.
Does Practice Validate Theory?
A theory, if true, can be employed successfully in applied research
(technological investigation) and in practice itself-as long as the theory
is relevant to either. (Fundamental theories are not so applicable because
they deal with problems much too remote from practical problems.
Just think of applying the quantum theory of scattering to car col-
lisions.) But the converse is not true, that is, the practical success or
failure of a scientific theory is no objective index of its truth value. In
fact, a theory can be both successful and false; or, conversely, it can be
a practical failure and nearly true. The efficiency of a false theory may
be due to either of the following reasons. First, a theory may contain
just a grain of truth, and this grain alone is employed in the theory’s
applications. In fact, a theory is a system of hypotheses, and it is enough
for a few of them to be true or nearly so in order to be able to entail
adequate consequences if the false ingredients are not used in the de-
duction or if they are practically innocuous. Thus, it is possible to
manufacture excellent steel by combining magical exorcisms with the
operations prescribed by the craft-as was done until the beginning of
the nineteenth century. And it is possible to improve the condition of
neurotics by means of shamanism, psychoanalysis, and other practices
as long as effective means, such as suggestion, conditioning, tranquil-
izers, and above all time are combined with them.
A second reason for the possible practical success of a false theory
may be that the accuracy requirements in applied science and in prac-
tice are far below those prevailing in pure research, so that a rough and
simple theory supplying quick correct estimates of orders of magnitude
very often will suffice in practice. Safety coefficients will mask the
finer details predicted by an accurate and deep theory anyway, and such
coefficients are characteristic of technological theory because this must
adapt itself to conditions that can vary within ample bounds. Think of
the variable loads a bridge can be subjected to or of the varying indi-
viduals that may consume a drug. The engineer and the physician are
interested in safe and wide intervals centered in typical values rather
than in exact values. A greater accuracy would be pointless since it is
not a question of testing. Moreover, such a greater accuracy could be
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
confusing because it would complicate things to such an extent that the
target-on which action is to be focused-would be lost in a mass of
detail. Extreme accuracy, a goal of scientific research, not only is point-
less or even encumbering in practice in most cases but can be an obstacle
to research itself in its early stages. For the two reasons given above-
use of only a part of the premises and low accuracy requirements-
infinitely many possible rival theories can yield “practically the same
results.” The technologist, and particularly the technician, are justified
in preferring the simplest of them: after all, they are interested pri-
marily in efficiency rather than in truth, in getting things done rather
than in gaining a deep understanding of them. For the same reason,
deep and accurate theories may be impractical; to use them would be
like killing bugs with nuclear bombs. It would be as preposterous-
though not nearly so dangerous-as advocating simplicity and efficiency
in pure science.
A third reason why most fundamental scientific theories are of no
practical avail is not related to the handiness and sturdiness required by
practice but has a deep ontological root. The practical transactions of
man occur mostly on his own level; and this level, like others, is
rooted to the lower levels but enjoys a certain autonomy with respect
to them, in the sense that not every change occurring in the lower
levels has appreciable effects on the higher ones. This is what enables
us to deal with most things on their own level, resorting at most to
the immediately adjacent levels. In short, levels are to some extent
stable: there is a certain amount of play between level and level, and
this is a root of both chance (randomness due to independence) and
freedom (self-motion in certain respects). One-level theories will suf-
fice, therefore, for many practical purposes. It is only when a knowl-
edge of the relations among the various levels is required in order to
implement a “remote-control” treatment, that many-level theories
must be tried. The most exciting achievements in this respect are those
of psychochemistry, the goal of which is, precisely, the control of
behavior by manipulating variables in the underlying biochemical level.
A fourth reason for the irrelevance of practice to the validation of
theories-even to operative theories dealing with action-is that, in
real situations, the relevant variables are seldom adequately known and
precisely controlled. Real situations are much too complex for this,
and effective action is much too strongly urged to permit a detailed
study-a study that would begin by isolating variables and tying some
of them into a theoretical model. The desideratum being maximal ef-
ficiency, and not at all truth, a number of practical measures will
usually be attempted at the same time: the strategist will counsel the
335
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
336 Mario Bunge
simultaneous use of weapons of several kinds, the physician will pre-
scribe a number of supposedly concurrent treatments, and the poli-
tician may combine promises and threats. If the outcome is satisfactory,
how will the practitioner know which of the rules was efficient, hence
which of the underlying hypotheses was true? If unsatisfactory, how
will he be able to weed out the inefficient rules and the false under-
lying hypotheses?
A careful discrimination and control of the relevant variables and a
critical evaluation of the hypotheses concerning the relations among
such variables is not done while killing, curing, or persuading people,
not even while making things, but in leisurely, planned, and critically
alert scientific theorizing and experimentation. Only while theorizing
or experimenting do we discriminate among variables and weigh their
relative importance, do we control them either by manipulation or by
measurement, and do we check our hypotheses and inferences. This is
why factual theories, whether scientific or technological, substantive
or operative, are empirically tested in the laboratory and not in the
battlefield, the consulting office, or the market place. (“Laboratory”
is understood here, in a wide sense, to include any situation which,
like the military maneuver, permits a reasonable control of the rele-
vant variables.) This is, also, why the efficiency of the rules employed
in the factory, the hospital, or the social institution, can be determined
only in artificially controlled circumstances.
In short, practice has no validating force; pure and applied research
alone can estimate the truth value of theories and the efficiency of
technological rules. What the technician and the practical man do,
by contrast to the scientist, is not to test theories but to use them with
non-cognitive aims. (The practitioner does not even test things, such
as tools or drugs, save in extreme cases: he just uses them, and their
properties and their efficiency again must be determined in the labora-
tory by the applied scientist.) The doctrine that practice is the touch-
stone of theory relies on a misunderstanding of both practice and
theory, on a confusion between practice and experiment and an asso-
ciated confusion between rule and theory. The question “Does it
work? “-pertinent as it is with regard to things and rules-is impertinent
in respect of theories.
Yet it might be argued that a man who knows how to do something
is thereby showing that he knows that something. Let us consider the
three possible versions of this idea. The first can be summed up in the
schema “If x knows how to do (or make) y, then x knows y.” To
ruin this thesis it is enough to recall that, for nearly one million years,
man has known how to make children without having the remotest
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
idea about the reproduction process. The second thesis is the converse
conditional, namely, “If x knows y, then x knows how to do (or make)
y.” Counterexamples: we know something about stars, yet we are
unable to make them, and we know part of the past, but we cannot
even spoil it. The two conditionals being false, the biconditional “x
knows y if and only if x knows how to do (or make) y” is false, too.
In short, it is false that knowledge is identical with knowing how to
do, or know-how. What is true is rather this: knowledge considerably
improves the chances of correct doing, and doing may lead to knowing
more (now that we have learned that knowledge pays), not because
action is knowledge, but because, in inquisitive minds, action may
trigger questioning.
It is only by distinguishing scientific knowledge from instrumental
knowledge, or know-how, that we can hope to account for the co-
existence of practical knowledge with theoretical ignorance and the
coexistence of theoretical knowledge with practical ignorance. Were
it not for this the following combinations hardly would have occurred
in history: (1) science without the corresponding technology (e.g.,
Greek physics); (2) arts and crafts without an underlying science (e.g.,
Roman engineering and contemporary intelligence testing). The dis-
tinction must be kept, also, in order to explain the cross-fertilizations
of science, technology, and the arts and crafts, as well as to explain
the gradual character of the cognitive process. If, in order to exhaust
the knowledge of a thing, it were sufficient to produce or reproduce
it, then certain technological achievements would put an end to the
respective chapters of applied research: the production of synthetic
rubber, plastic materials, and synthetic fibres would exhaust polymer
chemistry; the experimental induction of cancer should have stopped
cancer research; and the experimental production of neuroses and psy-
choses should have brought psychiatry to a halt. As a matter of fact,
we continue doing many things without understanding how, and we
know many processes (such as the fusion of helium out of hydrogen)
which we are not yet able to control for useful purposes (partly be-
cause we are too eager to attain the goal without a further develop-
ment of the means). At the same time it is true that the barriers be-
tween scientific and practical knowledge, pure and applied research,
are melting. But this does not eliminate their differences, and the
process is but the outcome of an increasingly scientific approach to
practical problems, that is, of a diffusion of the scientific method.
The identification of knowledge and practice stems not only from
a failure to analyze either but also from a legitimate wish to avoid
the two extremes of speculative theory and blind action. But the
337
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
338 Mario Bunge
testability of theories and the possibility of improving the rationality
of action are not best defended by blurring the differences between
theorizing and doing, or by asserting that action is the test of theory,
because both theses are false and no program is defensible if it rests
on plain falsity. The interaction between theory and practice and the
integration of the arts and crafts with technology and science are not
achieved by proclaiming their unity but by multiplying their con-
tacts and by helping the process whereby the crafts are given a tech-
nological basis and technology is entirely converted into applied sci-
ence. This involves the conversion of the rules of thumb peculiar to
the crafts into grounded rules, that is, rules based on laws. Let us
approach this problem next.
Scientific Law and Technological Rule
Just as pure science focuses on objective patterns or laws, action-
oriented research aims at establishing stable norms of successful hu-
man behavior, that is, rules. The study of rules-the grounded rules of
applied science-is therefore central to the philosophy of technology.
A rule prescribes a course of action; it indicates how one should
proceed in order to achieve a predetermined goal. More explicitly, a
rule is an instruction to perform a finite number of acts in a given
order and with a given aim. The skeleton of a rule can be symbolized
as a string of signs, such as 1-2-3- . . . -n, where every number stands
for a corresponding act; the last act, n, is the only thing that separates
the operator who has executed every operation, save n, from the goal.
In contrast to law formulas, which say what the shape of possible
events is, rules are norms. The field of law is assumed to be the whole
of reality, including rule-makers; the field of rule is but mankind; men,
not stars, can obey rules and violate them, invent and perfect them.
Law statements are descriptive and interpretive, whereas rules are
normative. Consequently, while law statements can be more or less
true, rules can be only more or less effective.
We may distinguish the following genera of rules: (1) rules of con-
duct (social, moral, and legal rules); (2) rules of prescientific work
(rules of thumb in the arts and crafts and in production); (3) rules
of sign (syntactical and semantical rules); (4) rules of science and
technology (grounded rules of research and action). Rules of conduct
make social life possible (and hard). The rules of prescientific work
dominate the region of practical knowledge which is not yet under
technological control. The rules of sign direct us how to handle sym-
bols-how to generate, transform, and interpret signs. And the rules of
science and technology are those norms that summarize the special
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
techniques of research in pure and applied science (e.g., random-
sampling techniques) and the special techniques of advanced modern
production (e.g., the technique of melting with infrared rays).
Many rules of conduct, work, and sign, are conventional, in the
sense that they are adopted with no definite reasons and might be
exchanged for alternative rules with little or no concomitant change
in the desired result. They are not altogether arbitrary, since their
formation and adoption should be explainable in terms of psycho-
logical and sociological laws, but they are not necessary either; the
differences among cultures are largely differences among systems of
rules of that kind. We are not interested in such groundless or con-
ventional rules but rather in founded rules, that is, in norms satisfy-
ing the following definition: A rule is grounded if and only if it is
based on a set of law formulas capable of accounting for its effective-
ness. The rule that commands taking off the hat when greeting a lady
is groundless in the sense that it is based on no scientific law but is
conventionally adopted. On the other hand, the rule that commands
greasing cars periodically is based on the law that lubricators decrease
the wearing out of parts by friction; this is neither a convention nor a
rule of thumb like those of cooking and politicking-it is a well-
grounded rule.
To decide that a rule is effective it is necessary, though insufficient,
to show that it has been successful in a high percentage of cases. But
these cases might be just coincidences, such as those that may have
consecrated the magic rituals that accompanied the huntings of primi-
tive man. Before adopting an empirically effective rule we ought to
know why it is effective; we ought to take it apart and reach an un-
derstanding of its modus operandi. This requirement of rule founda-
tion marks the transition between the prescientific arts and crafts and
contemporary technology. Now, the sole valid foundation of a rule
is a system of law formulas, because these alone can be expected to
correctly explain facts, for example, the fact that a given rule works.
This is not to say that the effectiveness of a rule depends on whether
it is founded or groundless but only that, in order to be able to judge
whether a rule has any chance of being effective, as well as in order
to improve the rule and eventually replace it by a more effective one,
we must disclose the underlying law statements, if any. We may take
a step ahead and claim that the blind application of rules of thumb has
never paid in the long run; the best policy is, first, to try to ground
our rules and, second, to try to transform some law formulas into
effective technological rules. The birth and development of modern
technology is the result of these two movements.
339
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
340 Mario Bunge
But it is easier to preach the foundation of rules than to say exactly
what the foundation of rules consists in. Let us try to make an inroad
into this unexplored territory-the core of the philosophy of tech-
nology. As usual when approaching a new subject, it will be con-
venient to begin by analyzing a typical case. Take the law statement
“Magnetism disappears above the Curie temperature (770? C for
iron).” For purposes of analysis it will be convenient to restate our
law as an explicit conditional: “If the temperature of a magnetized
body exceeds its Curie point, then it becomes demagnetized.” (This
is, of course, an oversimplification, as every other ordinary-language
rendering of a scientific law: the Curie point is not the temperature
at which all magnetism disappears but, rather, the point of conversion
of ferromagnetism into paramagnetism, or conversely. But this is a
refinement irrelevant to most technological purposes.) Our nomo-
logical statement provides the basis for the nomopragmatic statement
“‘If a magnetized body is heated above its Curie point, then it is de-
magnetized.” (The pragmatic predicate is, of course, “is heated.”)
This nomopragmatic statement is, in turn, the ground for two differ-
ent rules, namely, RI: “In order to demagnetize a body heat it above
its Curie point,” and R2: “To prevent demagnetizing a body do not
heat it above its Curie point.” Both rules have the same foundation,
that is, the same underlying nomopragmatic statement, which in turn
is supported by a law statement assumed to represent an objective
pattern. Moreover, the two rules are equiefficient, though not under
the same circumstances (changed goals, changed means).
Notice, first, that unlike a law statement a rule is neither true nor
false; as a compensation it can be effective or ineffective. Second, a
law is consistent with more than one rule. Third, the truth of a law
statement does not insure the efficiency of the associated rules; in fact,
the former refers to idealized situations which are not met with in
practice. Fourth, whereas given a law we may try out the correspond-
ing rules, given a rule we are unable to trace the laws presupposed by
it; in fact, a rule of the form “In order to attain the goal G employ
the means M” is consistent with the laws “If M, then G,” “M and G,”
“M or G,” and infinitely many others.
The above has important consequences for the methodology of rules
and the interrelations between pure and applied science. We see there
is no single road from practice to knowledge, from success to truth;
success warrants no inference from rule to law but poses the problem
of explaining the apparent efficiency of the rule. In other words, the
roads from success to truth are infinitely many and consequently
theoretically useless or nearly so, that is, no bunch of effective rules
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
suggests a true theory. On the other hand, the roads from truth to
success are limited in number, hence feasible. This is one of the rea-
sons why practical success, whether of a medical treatment or of a
government measure, is not a truth criterion for the underlying hy-
potheses. This is also why technology-in contrast to the prescien-
tific arts and crafts-does not start with rules and end up with theories
but proceeds the other way around. This is, in brief, why technology
is applied science whereas science is not purified technology.
Scientists and technologists work out rules on the basis of theories
containing law statements and auxiliary assumptions, and technicians
apply such rules jointly with groundless (prescientific) rules. In either
case, specific hypotheses accompany the application of rules, namely,
hypotheses to the effect that the case under consideration is one where
the rule is in point because such and such variables-related by the
rule-are in fact present. In science such hypotheses can be tested;
this is true of both pure and applied research. But in the practice of
technology there may not be time to test them in any way other
than by applying the rules around which such hypotheses cluster-
and this is a poor test indeed, because the failure may be blamed either
on the hypotheses or on the rule or on the uncertain conditions of
application.
Scientific Prediction and Technological Forecast
For technology knowledge is chiefly a means to be applied to the
achievement of certain practical ends. The goal of technology is suc-
cessful action rather than pure knowledge, and accordingly the whole
attitude of the technologist while applying his technological knowl-
edge is active in the sense that, far from being an inquisitive onlooker
or a diligent burrower, he is an active participant in events. This dif-
ference of attitude between the technologist in action and the re-
searcher-whether pure or applied-introduces certain differences be-
tween technological forecast and scientific prediction.
In the first place, whereas scientific prediction says what will or
may happen if certain circumstances obtain, technological forecast
suggests how to influence circumstances so that certain events may
be brought about, or prevented, that would not normally happen;
it is one thing to predict the trajectory of a comet, quite another to
plan and foresee the orbit of an artificial satellite. The latter presup-
poses a choice among possible goals, and such a choice presupposes a
certain forecasting of possibilities and their evaluation in the light of
a set of desiderata. In fact, the technologist will make his forecast on
his (or his employer’s) estimate of what the future should be like if
341
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
342 Mario Bunge
certain desiderata are to be fulfilled; contrary to the pure scientist, the
technologist is hardly interested in what would happen anyway; and
what for the scientist is just the final state of a process becomes for
the technologist a valuable (or disvaluable) end to be achieved (or
to be avoided). A typical scientific prediction has the form “‘If x
occurs at time t, then y will occur at time t’ with probability p.” By
contrast, a typical technological forecast is of the form “If y is to be
achieved at time t’ with probability p, then x should be done at time
t.” Given the goal, the technologist indicates the adequate means, and
his forecast states a means-end relationship rather than a relation be-
tween an initial state and a final state. Furthermore, such means are
implemented by a specified set of actions, among them the technolo-
gist’s own actions.
This leads us to a second peculiarity of technological forecast: where-
as the scientist’s success depends on his ability to separate his object
from himelf (particularly so when his object happens to be a psycho-
logical subject)-that is, on his capacity of detachment-the technolo-
gist’s ability consists in placing himself within the system concerned-
at the head of it. This does not involve subjectivity, since after all the
technologist draws on the objective knowledge provided by science;
but it does involve partiality, a parti pris unknown to the pure research-
er. The engineer is part of a man-machine complex, the industrial psy-
chologist is part of an organization, and both are bound to devise and
implement the optimal means for achieving desiderata which are not
usually chosen by themselves; they are decision-makers, not policy-
makers.
The forecast of an event or process that is not under our control will
not alter the event or process itself. Thus, for example, no matter how
accurately an astronomer predicts the collision of two stars, the event
will occur in due course. But if an applied geologist can forecast a
landslide, then some of its consequences can be prevented. Moreover,
by designing and supervising the appropriate defense works the engi-
neer may prevent the landslide itself; he may devise the sequence of
actions that will refute the original forecast. Similarly, an industrial
concern may forecast sales for the near future on the (shaky) assump-
tion that a given state of the economy, say prosperity, will continue
during that lapse. But if this assumption is falsified by a recession, and
the enterprise had accumulated a large stock which it must get rid of,
then instead of making a new sales forecast (as a pure scientist would
be inclined to do), the management will try to force the original fore-
cast to come true by increasing advertisement, lowering sale prices, and
so on. As in the case of vital processes, a diversity of means will alter-
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
natively or jointly be tried to attain a fixed goal. In order to achieve this
goal any number of initial hypotheses may have to be sacrificed: in the
case of the landslide, the assumption that no external forces would inter-
fere with the process and, in the case of the sales, that prosperity would
continue. Consequently, whether the initial forecast is forcefully falsi-
fled (as in the case of the landslide) or forcefully confirmed (as in the
case of the sales forecast), this fact cannot count as a test of the truth
of the hypotheses involved; it will count only as an efficiency test of
the rules that have been applied. The pure scientist, on the other hand,
need not worry about altering the means for achieving a preset goal,
because pure science has no goals external to it.
Technological forecast, in sum, cannot be used for controlling things
or men by changing the course of events perhaps to the point of stop-
ping them altogether, or for forcing the predicted course even if un-
predictable events should interfere with it. This is true of the forecasts
made in engineering, medicine, economics, applied sociology, political
science, and other technologies: the sole formulation of a forecast
(prognosis, lax prediction, or prediction proper), if made known to
the decision-makers, can be seized upon by them to steer the course of
events, thus bringing about results different from those originally fore-
casted. This change, triggered by the issuance of the forecast, may con-
tribute either to the latter’s confirmation (self-fulfilling forecast) or to
its refutation (self-defeating forecast). This trait of technological fore-
cast stems from no logical property of it; it is a pattern of social action
involving the knowledge of forecasts and consequently is conspicuous
in modern society. Therefore, rather than analyzing the logic of caus-
ally effective forecast, we should start by distinguishing three levels in
it: (1) the conceptual level, on which the prediction p stands; (2) the
psychological level-the knowledge of p and the reactions triggered by
this knowledge; and (3) the social level-the actions actually performed
on the basis of the knowedge of p and in the service of extra-scientific
goals. This third level is peculiar to technological forecast.
This feature of technological forecast sets civilized man apart from
every other system. A non-predicting system, be it a jukebox or a frog,
when fed with information it can digest will process it and convert it
into action at some later time. But such a system does not purposely
produce most of the information, and it does not issue projections
capable of altering its own future behavior. A predictor-a rational
man, a team of technologists, or a sufficiently evolved automaton-can
behave in an entirely different way. When fed with relevant informa-
tion It at time t, it can process this information with the help of the
knowledge (or the instructions) available to it, eventually issuing a
343
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
344 Mario Bunge
prediction Pt’, at a later time t. This prediction is fed back into the
system and compared with the preset goal that controls the whole
process (without either causing it or supplying it with energy). If the
two are reasonably close, the system takes a decision that eventually
leads it to act so as to take advantage of the course of events. If, on the
other hand, the prediction differs significantly from the goal, this dif-
ference will again trigger the theoretical mechanism, which will elab-
orate a new strategy: a new prediction, Pt”, will eventually be issued
at time t”, a forecast including a reference to the system’s own partici-
pation in the events. The new prediction is fed back into the system
and, if it still disagrees with the goal, a new correction cycle is triggered,
and so on until the difference between the prediction and the goal
becomes negligible, in which case the system’s predicting mechanism
comes to rest. Henceforth the system will gather new information re-
garding the present state of affairs and will act so as to conform to the
strategy it has elaborated. This strategy may have required not only
new information regarding the external world (including the attitudes
and capabilities of the people concerned) but also new hypotheses or
even theories which had not been present in the instruction chart
originally received by the predictor. If the latter fails to realize it or to
obtain and utilize such additional knowledge, his or its action is bound
to be ineffective. Moral: the more brains the better.
Technological Forecast and Expert Prognosis
The preceding account of technological forecast is based on the
assumption that it relies on some theory, or rather theories, whether
substantive or operative. This assumption may be found wanting by
anyone knowing that the forecasts issued by experts in medicine, fi-
nance, or politics are often successful and yet involve no great deal of
theorizing. True, most often expert prognosis relies on inductive (em-
pirical) generalizations of the form “A and B occur jointly with the
observed frequency f,” or even just “A and B occur jointly in most
cases,” or “Usually, whenever A then B.” The observation that a given
individual, say a human subject or an economic state of affairs, has the
property A is then used to forecast that it has, or will acquire, the
property B. In daily life such prognoses are all we do, and the same
applies to most expert prognoses. Occasionally such prognoses made
with either ordinary knowledge or specialized but non-scientific knowl-
edge are more successful than predictions made with full-fledged but
false or rough theories; in many fields, however, the frequency of hits
is not better than the one obtained by flipping a coin. The point, though,
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science
is that expert forecast using no scientific theory is not a scientific ac-
tivity-just by definition of “scientific prediction.”
Yet it would be wrong to think that experts make no use of special-
ized knowledge whenever they do not employ scientific theories; they
always judge on the basis of some such knowledge. Only, expert knowl-
edge is not always explicit and articulate and, for this reason, it is not
readily controllable: it does not learn readily from failures, and it is
hard to test. For the progress of science, the failure of a scientific pre-
diction is by far preferable to the success of an expert prognosis, because
the scientific failure can be fed back into the theory responsible for it,
thereby giving us a chance to improve it, whereas in the case of expert
knowledge there is no theory to feed the failure into. It is only for im-
mediate practical purposes that expert prognoses made with shallow but
well-confirmed generalizations are preferable to risky scientific pre-
dictions.
Another difference between expert prognosis and technological fore-
cast proper would seem to be this: the former relies more heavily on
intuition than does scientific prediction. Yet the difference is one of
degree rather than of kind. Diagnosis and forecast, whether in pure
science, in applied science, or in the arts and crafts, involve intuitions of
a number of kinds: the quick identification of a thing, event, or sign;
the clear but not necessarily deep grasp of the meaning and/or the
mutual relations of a set of signs (text, table, diagram, etc.); the ability
to interpret symbols; the ability to form space models; skill in realizing
analogies; creative imagination; catalytic inference, that is, quick passage
from some premises to other formulas by skipping intermediate steps;
power of synthesis or synoptic grasp; common sense (or rather con-
trolled craziness), and sound judgment. These abilities intertwine with
specialized knowledge, whether scientific or not, and are reinforced
with practice. Without them theories could neither be invented nor
applied-but, of course, they are not suprarational powers. Intuition is
all right as long as it is controlled by reason and experiment; only the
replacement of theorizing and experimenting by intuition must be
feared.
A related danger is that of pseudoscientific projection tools, so com-
mon in applied psychology and sociology. A number of techniques
have been devised to forecast the performance of personnel, students,
and even psychologists themselves. A few tests, the objective ones, are
somewhat reliable; this holds for intelligence and skill tests. But most
tests, particularly the subjective ones (the “global evaluation” of per-
sonality by means of interviews, the thematic apperception tests, the
Rorschach, etc.) are in the best of cases inefficient and in the worst of
345
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
346 Mario Bunge
cases misleading. When they have been subjected to the test of pre-
diction-that is, when their results have been checked with the actual
performance of the subjects-they have failed. The failure of most in-
dividual psychological tests, and particularly of the subjective ones, is
not a failure of psychological testing in general; what is responsible for
such failures is either the total absence or the falsity of the underlying
psychological theories. Testing for human abilities without first estab-
lishing laws relating objective indexes of abilities or personality traits
is as thoughtless as asking a tribesman to test an aircraft. As long as no
theoretical foundations of psychological tests are secured, their employ-
ment as predictive instruments is not better than crystal gazing or coin-
flipping: they are practically inefficient and, even if they succeeded,
they would not contribute to psychological theory, because they are
unrelated to theory. The limited success of psychological testing has
led many to despair of the possibility of finding a scientific approach
to human behavior, but the right inference is that such an attempt has
been tried only after a large number of alleged tests invaded the market.
What is wrong with most of “applied” (educational, industrial, etc.)
psychology is that it does not consist in the application of scientific
psychology at all. The moral is that practical wants-such as personnel
training and selection-should not be allowed to force the construction
of “technologies” without an underlying science.
Technological forecast should be maximally reliable. This condition
excludes from technological practice-not, however, from technological
research-insufficiently tested theories. In other words, technology will
ultimately prefer the old theory that has rendered distinguished service
in a limited domain and with a known inaccuracy to the bold new
theory that promises unheard-of forecasts but is probably more complex
and therefore partly less well tested. It would be irresponsible for an
expert to apply a new idea in practice without having tested it under
controlled conditions. (Yet this is still done in pharmacy: recall the
affair of the mutagenic drugs in the early 1960’s.) Practice, and even
technology, is bound to be more conservative than science. Conse-
quently, the effects of a close association of pure research with applied
research, and of the latter with production, are not all of them bene-
ficial; while it is true that technology challenges science with new
problems and supplies it with new equipment for data-gathering and
data-processing, it is no less true that technology, by its very insistence
on reliability, standardization (routinization), and speed, at the expense
of depth, range, accuracy, and serendipity, can slow down the advance-
ment of science.
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
Technology as Applied Science 347
Other Problems
We have looked into a few problems of the philosophy of technol-
ogy. Many other challenging problems have been left out, for example,
the logic of technological rules; the test of technological theories; the
patterns of technological invention; the reason textile, aircraft, and
other industries are still largely based on crafts; and the power of tech-
nology to bring together previously separate fields (cases of cybernet-
ics, nuclear engineering, computer science, space science, and bioengi-
neering). These and many other problems are waiting to be discovered
and worked out by philosophers attentive to their own times. Why
should the waiting time be so long?
This content downloaded from 169.235.64.141 on Mon, 30 Sep 2019 20:14:07 UTC
All use subject to https://about.jstor.org/terms
- Contents
- Issue Table of Contents
image 1
image 2
image 3
image 4
image 5
image 6
image 7
image 8
image 9
image 10
image 11
image 12
image 13
image 14
image 15
image 16
image 17
image 18
image 19
Technology and Culture, Vol. 7, No. 3, Summer, 1966
Front Matter
Towards a Philosophy of Technology
Prefatory Note [pp.301-302]
Technics and the Nature of Man [pp.303-317]
Technology as Skills [pp.318-328]
Technology as Applied Science [pp.329-347]
The Confusion between Science and Technology in the Standard Philosophies of Science [pp.348-366]
The Need for Corroboration: Comments on Agassi’s Paper [pp.367-370]
The Structure of Thinking in Technology [pp.371-383]
The Social Character of Technological Problems: Comments on Skolimowski’s Paper [pp.384-390]
Memorial
Lynn Thorndike (1882-1965) [pp.391-394]
The Cover Design
Dyeing Fabrics in Sixteenth-Century Venice [pp.395-397]
Communications
A Postscript to Reti’s Notes on Juanelo Turriano’s Water Mills [pp.398-401]
Technology, Traditionalism, and Military Establishments [pp.402-407]
The Inaccurate “Slide Rule” [pp.408-409]
Book Reviews
untitled [pp.410-411]
untitled [pp.412-413]
untitled [pp.413-415]
untitled [pp.415-418]
untitled [pp.418-420]
untitled [pp.420-421]
untitled [pp.421-424]
untitled [pp.424-426]
untitled [pp.426-428]
untitled [pp.428-429]
untitled [pp.429-432]
untitled [pp.432-434]
untitled [pp.434-436]
untitled [pp.436-438]
untitled [pp.438-440]
untitled [pp.440-441]
untitled [pp.441-443]
untitled [pp.443-444]
untitled [pp.444-445]
untitled [pp.445-446]
untitled [pp.446-447]
Notes and Announcements [pp.448-451]
Back Matter
Expert paper writers are just a few clicks away
Place an order in 3 easy steps. Takes less than 5 mins.