SAMI PAAVOLA

ABDUCTION AS A LOGIC AND METHODOLOGY OF

DISCOVERY: THE IMPORTANCE OF STRATEGIES

ABSTRACT. There are various “classical” arguments against abduction as a logic

of discovery, especially that (1) abduction is too weak a mode of inference to be
of any use, and (2) in basic formulation of abduction the hypothesis is already
presupposed to be known, so it is not the way hypotheses are discovered in the
first place. In this paper I argue, by bringing forth the idea of strategies, that
these counter-arguments are weaker than may appear. The concept of strategies
suggests, inter alia, that many inferential moves are taken into account at the same
time. This is especially important in abductive reasoning, which is basically a
very weak mode of inference. The importance of strategic thinking can already be
seen in Charles S. Peirce’s early treatments of the topic, and N.R. Hanson’s later
writings on abduction although they did not use the concept of “strategies.” On the
whole, I am arguing that the focus should be more on methodological processes,
and not only on validity considerations, which have dominated the discussion
about abduction.

KEY WORDS: abduction, logic of discovery, methodology, strategies

1. BASIC CRITICISM AGAINST ABDUCTION

(AS A LOGIC OF DISCOVERY)

Abductive inference arouses increasing interest and methodological

discussion in various fields. In the philosophical context abduction
has, however, very often been subjected to severe criticism (see
e.g., Frankfurt, 1958; Nickles, 1980b; Kapitan, 1992). There are, so
to speak, various “classical” counter-arguments against abduction,
especially if abduction is presented as a logic of discovery. A basis
for these counter-arguments is the widely held view that discovery
is something that cannot be treated by conceptual or philosophical
means.
One basic way of formulating abduction is the one made already
by Charles S. Peirce (see Peirce, CP 5.189; Hanson, 1972, p. 86):
The surprising fact, C, is observed;

Foundation of Science 9: 267–283, 2004.

© 2004 Kluwer Academic Publishers. Printed in the Netherlands.
268 SAMI PAAVOLA

But if H [an explanatory hypothesis] were true, C would

be a matter of course,
Hence, there is reason to suspect that H is true.

So abduction can be understood as a mode of inference where

explanations are searched for (anomalous or suprising) phenomena.
But this kind of an inference can be criticized because it is too
permissive to be of much use. It seems to permit inferences to
all sorts of wild hypotheses (Frankfurt, 1958, p. 596; Nickles,
1980a, p. 24; Kapitan, 1992, p. 6). Peter Achinstein has presented
various, often cited, counter-examples in line with this criticism.
For example, the hypothesis that I will be paid one million dollars
if this paper is published, would explain (if it were true) why I
am writing this paper. But still, there is no reason to think that
I am about to come a millionaire (unfortunately so!). Or another
Achinstein example: Let us suppose as an observed fact that I am
happy about some news I have just received. Then a hypothesis
could be proposed that I have just received the news that I have won
the Nobel Prize in literature (because it is reasonable to suppose
that anyone who hears the news about the Nobel Prize winning is
happy). But here again, the fact that I am happy, should give no
reason to believe that I have won the Nobel Prize. So, from the mere
fact that some hypothesis H, if it was true, would explain the data,
does not usually follow that there is reason to think that that H is true
(Achinstein, 1970, p. 92, 1971, p. 118, 1987, p. 413). It is to be noted
that Achinstein does not take into account that the fact observed
should somehow be ‘surprising’ (Is it surprising that I am writing
this paper? Is it surprising that I am happy? In what sense are these
facts supposed to be surprising?). But still, Achinstein’s argument
appears to be basically adequate. The problem with abduction is
that its basic formula seems to allow these kinds of inferences to all
sorts of wild and crazy hypotheses.
Another basic criticism against abduction is that it cannot be
a logic of discovery because the hypothesis is already included
(or supposed to be known) in the premises (see above the second
premise) (e.g., Frankfurt, 1958, p. 594; Nickles, 1980a, p. 23;
Hoffmann, 1999, pp. 278–279). So it seems that the new idea is
not a result of abductive inference, and abduction can be at most
a logic for preliminary evaluation or appraisal of a hypothesis that
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 269

is already discovered by some other means (Schon, 1959, pp. 501–

502; Kapitan, 1992, p. 2). In this sense abduction is often placed in
that phase of activity which is carried out after original discovery but
before final justification (e.g., Nickles, 1980a, pp. 18–22; Laudan,
1980, p. 174). This means – if the idea is followed – that the old
distinction between the context of discovery and the context of justi-
fication (see Reichenbach, 1938; Nickles, 1980a) is not enough, and
a third region is needed. The basic idea is that this third area can be
analysed by logical and conceptual means, although it is not a matter
of justification in proper sense yet. But the context of discovery in
genuine sense is still, according to this idea, something inexplicable,
or possibly a subject for empirical sciences but not for conceptual
analysis.

2. THE MEANING OF STRATEGIES IN ABDUCTION

I think that abduction can still be defended as a very promising

candidate for a logic of discovery if the meaning of strategies is
taken into account. Jaakko Hintikka has emphasized a distinction
between two sorts of rules in reasoning and logic (or in games in
general): the definitory rules and the strategic rules. Hintikka main-
tains that for the theory of logic and reasoning, especially at the
level of introductory textbooks and courses, the study of excellence
of reasoning is often forgotten, and the emphasis is on the avoidance
of mistakes in reasoning (e.g., Hintikka, 1999). According to him,
students are not taught how to reason well but to maintain their
logical virtue (i.e., to avoid logical fallacies and to learn what is and
what is not admissible and valid). The focus has been on definitory
rules of logic, and strategic rules have largely been neglected. The
definitory rules tell what are valid rules in particular system of logic.
By analogy: the definitory rules of chess tell what one is allowed to
do in chess (how chessmen may be moved etc.). But by knowing
only the definitory rules of chess one cannot say that one plays
chess well. Excellence in chess requires that one master strategic
rules extremely well. According to Hintikka, this same idea applies
to logic. No one is good in logic and reasoning by knowing only the
definitory rules of logic, but by mastering well the strategic rules.
Strategies have, however, been a quite neglected topic in philos-
ophy of science. There are some exceptions. In the interrogative
270 SAMI PAAVOLA

approach to inquiry, the meaning of strategies has been emphasized

(Hintikka, 1985, 1989; Jung, 1996). But usually the merits of infer-
ence are assessed by investigating whether the truth of the premises
guarantees or makes probable the truth of the conclusion. And this
has also been the basic way of evaluating abduction.
Hintikka has emphasized strategic aspects also in relationship to
abductive inference: “the validity of an abductive inference is to
be judged by strategic principles rather than by definitory (move-
by-move) rules” (Hintikka, 1998, p. 513). Hintikka himself does
not, however, treat abduction as a separate mode of inference in his
interrogative model of inquiry. In Hintikka’s model the problem of
abduction is subsumed under a more general problem concerning
the nature of ampliative reasoning in general. In Hintikka’s model,
abduction is closely related to a question-answer step in the process
of inquiry (ibid., pp. 519–523).
I suggest that abduction as a separate mode of inference can also
be defended by taking strategies into account much more seriously
than before (see also Magnani, 1999, pp. 235–236). But what of
strategies mean in the area of reasoning? Strategy is in itself a very
complicated concept. Strategies are related to goal-directed activity,
where the ability to anticipate things, and to assess or choose
between different possibilities, are important (see e.g., Hintikka,
1989, 1999). One central point in strategic rules is that they cannot
normally be judged only in relationship to particular moves, but the
whole strategic situation must be taken into account (see Hintikka,
1998, p. 513). This means that in strategies more than one step or
move can and must be taken into account at the same time. I am
not maintaining that this characterization does full justice to the
meaning of strategies in (abductive) inference, but it is surely one
essential point. And in abduction, strategies are especially important
because it is basically such a weak mode of inference. The force of
abductive inference is much strengthened if one takes into account
that the hypotheses are to be searched for in relationship to various
phenomena and background information and not just in order to
explain one, surprising phenomenon.
So if I am a researcher looking for a good explanatory hypothesis
for some anomalous phenomenon, I can (and must) try to constrain
and guide my search by taking into account that my explanation
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 271

must explain or at least be consistent with, most other clues and

information that I have available concerning the subject matter. And
I try to anticipate that my explanation has some chance of survival
in subsequent tests and assessments. Usually I must also take into
account that the proposed explanation should not be totally uncon-
vincing, or if it seems to be that, I should have a good further
explanation for why this explanation still deserves attention. So, I
should have an explanation for my explanation (see Thagard and
Shelley, 1997). These are strategic principles where more than one
move can and should be taken into account simultaneously.
In Achinstein’s counterexamples, one important reason why
those wild hypotheses are not reasonable (it is to be noted that the
term “reasonable” is just the right one here) is that they do not fit
well with other relevant information. Normally no one would pay
one million dollars for any academic paper (unfortunately again!),
which rules out this hypothesis as strategically bad. And there is
no reason to think that I have won the Nobel Prize in literature
– only because I am happy – if this hypothesis does not fit at all
with other information concerning who I am and what I have done.
Strategically, it would be bad reasoning to suggest such implausible
hypotheses, if there were no good further reasons or backing for
these. A strategically good hypothesis takes also into account that
there is an explanation for my explanation (or at least explanation
why there cannot be any further explanation). Why someone would
pay me one million dollars for this paper? Why would someone
even suggest a Nobel Prize for me? If there are no good answers for
these questions, there is no point in even suggesting these kinds of
hypotheses at the first place. Achinstein himself apparently notices
the situation: normally we take the relevant background informa-
tion into account, and we require further evidence for some odd
hypothesis in order to take it seriously (see Achinstein, 1970, p. 92,
1971, p. 118, 1987, p. 416). But what Achinstein is missing is the
idea that strategies are involved here.
The strategic viewpoint does not, however, rule out the possibility
of suggesting implausible hypotheses altogether. It is, for example,
possible to imagine such a course of events, where the one million
dollars hypothesis (or the Nobel Prize hypothesis) is the correct
one, because of some extraordinary circumstances. Especially if
272 SAMI PAAVOLA

more conventional or plausible hypotheses have not worked out, it

might be a good strategy to try more implausible hypotheses. (In
Achinstein’s one-million dollars example, if the situation were that,
for some reasons, it would be very surprising that I am giving this
presentation, and other explanations would not seem to work out,
it would be a good candidate hypothesis to think that someone is
paying me a lot of money for this.) What I am arguing is that in
strategies, the reasoner tries to anticipate the counter-arguments, and
to take into account all the relevant information, and this rules out
very “wild” hypotheses, except, when there is no other available, or
alternatively, when these are presented simply as “wild guesses”.
I will now give further consideration to the criticism that abduc-
tion cannot be the logic of discovery because the hypothesis or the
idea is already presupposed in the premises. To begin with, I am
not maintaining that abductive inference is an automatic means for
making discoveries.
I agree with N.R. Hanson, who emphasized that abductive infer-
ence is of a way of analysing conceptual issues in discoveries
rather than “a manual” or an algorithmic device for making these
discoveries (Hanson, 1961, pp. 21–22). If abduction is to be
considered as the logic of discovery, the whole methodological
process must be taken into account; one must not just concentrate on
the form of the argument. In this sense, it is for example important to
think how these “surprising facts” (see the formulation of abduction
above) operate as clues in the search for explanations or hypotheses.
From the point of view of the inquirer, the difficult part in abductive
search might be to find fruitful premises. But this does not mean that
abduction cannot analyse properly the logical form of discoveries.
Strategies are also involved here. Although the hypothesis is in
the premises, abduction can still be the logic of discovery because,
as Peirce wrote (Peirce, CP 5.181):
It is true that the different elements of the hypothesis were in our minds before;
but it is the idea of putting together what we had never before dreamed of putting
together which flashes the new suggestion before our contemplation.

How I interpret this idea from the strategic viewpoint is that a

hypothesis suggested can in itself be something old and even well
known (cf. Anderson, 1987, pp. 47–48). But the way in which this
hypothesis is seen to fit with this particular problem in question
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 273

and with other relevant information (besides the one anomalous

phenomenon) is crucial. And it is also essential that there is a
further explanation or clarification for this particular hypothesis. For
example, the idea of evolution was not new when Charles Darwin
proposed it. It was widely admitted that it would have been a good
explanation for various phenomena, but the problem was that there
were no plausible explanations for how evolution operates more
specifically (and in fact there seemed to be lots of evidence which
was against the evolutionary hypothesis). As Darwin later recol-
lected his discovery and its relationship to the known facts (Darwin
in his autobiography: Barlow, 1958, pp. 118–119):
It was evident that such facts as these [various important observations that he
had made on the famous voyage of the Beagle], as well as many others, could
be explained on the supposition that species gradually become modified; and
the subject haunted me. But it was equally evident that neither the action of the
surrounding conditions, nor the will of the organisms (especially in the case of
plants), could account for the innumerable cases in which organisms of every
kind are beautifully adapted to their habits of life. I had always been much
struck by such adaptations, and until these could be explained it seemed to me
almost useless to endeavour to prove by indirect evidence that species have been
modified.

And likewise many details in Darwin’s theory were not new, but
the real discovery and insight was that Darwin showed that these
ideas really work in this particular context. This is also how Howard
Gruber has described Darwin’s famous Malthusian insight (Gruber,
1981, p. 42):
. . . his notebooks show that he had or almost had the same idea a number of times
before, during the fifteen months of deliberate effort leading up to the moment in
question. So the historic moment was in a sense a re-cognition of what he already
knew or almost knew.

So my point is, that the basic formula of abduction can still be an

essential part in the logic of discovery (even though the hypothesis
is in the premises) if the difficult part in discovery is the recognition
that the hypothesis really is a viable way of solving this particular
problem and that the hypothesis works more generally (and not only
in relationship to one, particular anomalous phenomenon).
I am not maintaining that the importance of strategies in
discovery means that in (scientific) discovery the researchers always
274 SAMI PAAVOLA

have their hypotheses already at their disposal and that the only
problem would be to strategically see, somehow, a connection
between moves in games of reasoning. Discovery means that some-
thing new is brought (or abducted!) to the particular situation. But
strategies must be taken into account when “aha-experiences” or
insights are involved. An aha-experience means that the hypothesis
(or the solution) in question fits with those constraints and clues
that are involved in the problem situation in question, i.e., the
insight seems to take into account many counter-arguments and
moves in advance. It is not enough that the hypothesis explains only
some detached, anomalous phenomena (e.g., for Darwin the idea of
evolution and the Malthusian principle were important discoveries
only when these ideas could be integrated to the larger argument
concerning species, and not as separate and unconnected explana-
tions). It almost seems that the basis for the aha-experience is a
situation where, first, various constraints and hints characterize the
situation and then some solution seems to fit with these constraints.
And this outlining of constraints and hints is, I submit, closely
related to strategic thinking, at least in the sense I use ’strategies’
here. A good insight is also a good one strategically.
I think that strategies are also involved when it is said that
abductive inference starts from anomalous or somewhat surprising
phenomena. It might be asked, why it is so often emphasized
that abductive inference starts from surprising phenomena (but cf.
Hoffmann, 1999, p. 281)? It does not seem to affect to the validity
of inference if it starts from surprising or from non-surprising
phenomenon. I think that this is also a strategic rule, for the
following reasons: In difficult problems or in cases where something
new is required, it is a good strategy (or a worthwhile one) to start
from anomalous facts or from little details, and try with them to find
a solution or a hypothesis. This is at least a strategy that detectives
(or detective novels) recommend (the connection between abduc-
tion and the reasoning that detectives use is often noticed, see e.g.,
Eco and Sebeok, 1988; Niiniluoto, 1999b). This is also how Francis
Darwin described how (his father) Charles Darwin worked (Darwin,
1892, pp. 94–95):

There was one quality of mind which seemed to be of special and extreme
advantage in leading him to make discoveries. It was the power of never letting
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 275

exceptions pass unnoticed. Everybody notices a fact as an exception when it is

striking or frequent, but he had a special instinct for arresting an exception. A
point apparently slight and unconnected with his present work is passed over by
many a man almost unconsciously with some half-considered explanation, which
is in fact no explanation. It was just these things that he seized on to make a start
from.

3. STRATEGIES IN PEIRCE’S AND HANSON’S TREATMENT OF

ABDUCTION: METHODEUTIC FOR ABDUCTION

Charles Peirce’s seminal writings concerning abduction already

contained many strategic insights, although he did not label them
as such (Hintikka, 1998, pp. 512–516). Peirce made a distinction
between three areas of logic: (Speculative) Grammar, Critic, and
Methodeutic. Peirce characterized this “trivium” in various ways
during his long career, so it is not possible to interpret this distinc-
tion in exact terms or unequivocally (see e.g., Peirce, CP 1.559,
CP 1.444, CP 2.93). Especially the third area, Methodeutic, leaves
room for various interpretations (in many writings he called it e.g.,
“Speculative Rhetoric”) (Liszka, 1996). In Grammar the nature and
meanings of signs are studied; in Critic arguments are classified and
the validity and the force of arguments are studied (see Peirce, CP
1.191). Methodeutic, however, “studies the methods that ought to be
pursued in the investigation, in the exposition, and in the application
of truth” (ibid.), and “the principles of the production of valuable
courses of research and exposition” (Peirce, EP2, p. 272).
In order to develop further the model of abductive inference, all
these three Peircean areas of logic (Grammar, Critic, Methodeutic)
are salient. According to Peirce, Critic is important in abduction
although it is a very “weak” mode of inference (Peirce, CP 5.188):
. . . abduction, although it is very little hampered by logical rules, nevertheless is
logical inference, asserting its conclusion only problematically or conjecturally, it
is true, but nevertheless having a perfectly definite logical form.

The area of Grammar should also be developed further in order to

better understand the nature of abductive inference. For example, the
special role of iconic and perceptual elements in abductive infer-
ence is, I think, part of Grammar (see Peirce, CP 2.96; Shelley,
1996; Thagard and Shelley, 1997). But what is important from the
276 SAMI PAAVOLA

point of view of this paper is that, in abductive inference, the role

of Methodeutic can be seen as especially significant because, in
Peirce’s words, “Of the different classes of arguments, abductions
are the only ones in which after they have been admitted to be just,
it still remains to inquire whether they are advantageous” (Peirce,
HP, p. 1035).
I maintain that according to the Peircean distinction of these three
areas of logic, strategies would belong to Methodeutic. Methodeutic
studies the process of inquiry and the way in which inquiry is carried
on, and strategies are part of it. As Peirce in his time emphasized
Methodeutic in reasoning, Hintikka, in recent decades has empha-
sized the need for strategic rules and study of fruitful methods of
logical reasoning.
There are also other concepts and ideas in Peirce’s writings
(besides Methodeutic) where the issue of strategies comes to the
fore. Hintikka has suggested that in Peirce’s concept of habit “there
lurks . . . a strategic rule trying to get out” (Hintikka, 1998, p. 515).
Peirce also emphasized the notion of “the economy of research,”
which is very closely related to strategic principles (Peirce, CP
7.220). According to Peirce various sorts of “economical” (broadly
interpreted) factors should guide the choice of the hypotheses, for
example, caution, which takes into account what will happen if
the hypothesis suggested does not work out. One example of this
economical caution is the game of twenty questions where the idea
is to guess what object someone is thinking by making good ques-
tions (ibid.). Only such questions are allowed as can be answered by
Yes or No (this limitation makes the game intriguing; there would
be no game if you were allowed to ask the object thought of right
away). The success in this game is based on skilful questions that
break up the search area most efficiently. I think it is quite clear
that this is the same as saying that the success is based on good
strategies.
Peirce himself often maintained that we humans must have some
sort of an instinct that guides our guesses (see e.g., Peirce, CP
2.753; CP 5.172–4). Peirce’s argument was that a human could not
arrive at his or her theories by pure chance. There are an infinite
number of theories that could be suggested if there is nothing that
helps us. According to Peirce our abductive guesses are not totally
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 277

haphazard or blind because we have an instinct for making these

guesses. But this explanation raises several questions. Is there any
evidence that we humans have this kind of a guessing instinct?
How would it operate? There is some plausibility to the Peircean
idea that evolution has moulded us to guess such hypotheses about
physical and psychical life, which have been important for our
survival. But how could this guessing instinct help humans to
discover modern scientific theories which are often counter-intuitive
and against a common sense way of understanding things? Peirce
himself admitted that this instinct explanation is not very plausible
when one considers the genesis of very complicated theories by
which we “penetrate further and further from the surface of nature”
(Peirce, CP 7.508; see also Peirce, CP 7.606). It is also problematic
to combine this instinct explanation with the idea that abduction is a
third mode of inference. It seems that if the crucial element in abduc-
tion is the guessing instinct there is not much room for abductive
reasoning. There are various suggestions how instinct and inference
can be combined in a Peircean scheme (e.g., Fann, 1970; Anderson,
1987) but still it is a quite problematic how this combining could
satisfactorily be done.
Nicholas Rescher has suggested that Peirce’s “somewhat myster-
ious” capacity of instinct should be replaced by a methodology of
inquiry (Rescher, 1995, pp. 321–323; Hoffmann, 1999, p. 297). The
idea is that scientific inquiry and discovery is not blind because
methods and methodology give rational principles that guide the
processes of inquiry. Methods themselves, according to Rescher,
have emerged through evolutionary trial-and-error process. I am
not trying to evaluate Rescher’s solution comprehensively, but I
think that Rescher is making a very important point here. Scien-
tists need not start from scratch because methods and methodologies
that have proven to be successful are guiding inquiry. But it seems
that Rescher’s solution cannot be the whole story. If methods and
methodology are supposedly reached by trial-and-error, are we not
facing again the same problem that was supposed to be solved (this
is one version of the classical Meno paradox)? Are these methods
found by pure chance in evolutionary way? Is this possible? And
if so, is there some kind of an agreement what these successful
methods nowadays are? If they guide scientific inquiry in contrast
278 SAMI PAAVOLA

to instincts, should we not know them explicitly? And even if we

could characterize these methodologies in general terms, is there
any guarantee that they work in particular cases?
What I have tried to argue has some similarities to Rescher’s
solution. There is no need to abandon the idea that we humans might
have some abilities that can be called “instincts” and which help us
when we are trying to discover something new. But these abductive
instincts cannot be the fundamental basis for abductive inference. In
order to understand abductive reasoning better, the focus should be
more on methodological processes. But unlike Rescher, I don’t think
that the most important feature of methodology is that it is moulded
by evolutionary process.
The focus on strategies means that inquiry is seen as a kind of a
problem solving process where the inquirer uses the best inquiry
strategies possible. The use of strategies explains why inquiry is
not purely blind, even when something new is discovered. The
inquirer must try to take into account all the information that is
relevant to his or her subject area. Strategically, and from the
point of view of the inquirer, it would be bad reasoning to suggest
very “wild” hypotheses, or hypotheses based on pure chance; it
is wise to take into account existing knowledge. It is true that
new (especially revolutionary) discoveries often mean that some
parts of this existing knowledge must be abandoned. But even then
the inquirer must be able to combine the new ideas with existing
knowledge or constraints, (or to be able to show that the existing
knowledge is in some ways inadequate). These constraints can be
negative, in a sense that they inhibit new ideas, but they can also
be positive in suggesting methods, theories, information and so on
which must be taken into account and which might give clues how
to solve the problems in question. This idea is similar to the role of
“normal science” or “paradigms” in Thomas Kuhn’s famous model
of scientific growth (Kuhn, 1970). Paradigms are not automatic
ways of solving problems, but they give good suggestions for how
to conduct inquiry. This is the reason why inquiry does not always
have to start from scratch. Kuhn does not extend the idea to revolu-
tionary situations where paradigms themselves change. But even in
revolutionary situations, existing knowledge constrains and guides
the search for new information and discoveries.
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 279

I think that Norwood Russell Hanson’s old ideas concerning

abduction should also be seen from a strategic angle. Hanson was
already a proponent of the idea of the logic of discovery in the 1950s
and 1960s when these subjects were not popular in philosophy of
science (Hanson, 1961, 1972). Hanson’s formulations have been
criticized as inadequate for the logic of discovery (see e.g., Nickles,
1980b). My suggestion is that this criticism has been concerned
more with the validity of abduction (with “Critic” in Peircean
terms), and Hanson himself was more interested in abduction as
a part of methodological processes of inquiry (i.e., Methodeutic).
Hanson explicitly criticized the “logic of the finished research
reports;” he put the emphasis on processes of discovery and main-
tained that these processes can be analysed by conceptual means
(Hanson, 1961, pp. 20–22).
In his writings concerning abduction, Hanson distinguished three
ingredients in “the logic of discovery” (Hanson, 1965, pp. 47–65).
It is reasoning which
1) proceeds retroductively, from an anomaly to
2) the delineation of a kind of explanatory H which
3) fits into an organized pattern of concepts (ibid., p. 50).
I think that these points can be seen as strategic principles although
Hanson himself did not develop these ideas explicitly in this way.
The starting point in anomalies (point 1) can be seen as a strategic
principle (see the end of chapter 2. above). Hanson also pointed
out that abduction does not mean that some particular hypothesis is
found straight away. An important phase in the process of discovery
might be that the type of the solution is delineated before the solution
is acquired (point 2) (Hanson, 1961; Niiniluoto, 1999a, pp. S440–
S441). This is strategic thinking: the constraints and hints that
help in hypothesis finding are taken into account. And the goal in
abductive inference (at least in most cases) is to find an overall
pattern into which all evidence and clues fit (point 3), and this phase
especially requires that various inferential moves be put together
skilfully and by taking various clues and constraints into account
(a paradigmatic case is detective stories, but this is in itself a very
general model).
280 SAMI PAAVOLA

4. CONCLUSION: A LOGIC AND METHODOLOGY OF DISCOVERY

Various approaches in the philosophy of science nowadays

emphasize the need to find a model for inquiry that is not strictly
logical, at least in the “old” sense (i.e., in the sense that formal logic
has been usually understood in the last century) but one that is not
purely relativistic or historicistic either (see e.g., Pera, 1994; Jung,
1996; Thagard, 2000; Aliseda, 2001). This need arises especially
from an aim to understand processes of discovery and knowledge
formation and not just finished products (Sintonen, 1996; Aliseda,
1997). Philosophy of science and logic have for long concen-
trated on analysing the structure of finished products, e.g., what
is an explanation, and not processes of inquiry, e.g., how good
explanations are searched for and found (Aliseda, 2001). There
have been some notable exceptions, e.g., N.R. Hanson, but usually
these processes were literally defined to be something that cannot
be analysed by conceptual means, or by philosophical models.
Thomas Nickles has made a distinction between “heuristic
appraisal” and “epistemic appraisal” in methodology (Nickles,
1989). Epistemic appraisal is the standard way of doing things in the
philosophy of science. It is a retrospective, justificatory assessment
of scientific results. Heuristic appraisal means that the promising-
ness, or fertility of scientific proposals or problems is being
assessed. It can be maintained that this is at least as important, if
not more important than epistemic appraisal. Nowadays heuristical
aspects are often emphasized when rationality of discovery is
searched for (Jung, 1996; Magnani, 1999, pp. 235–236; Aliseda,
2001).
The “friends of discovery” have for long emphasized that the
concept of rationality must be broadened if the context of discovery
is to be taken into account (Nickles, 1980b). There is a new
opportunity for this broadening when new conceptual means are
developed, i.e., conceptual means that try to capture processes of
inquiry and discovery (see e.g., Hintikka, 1985, 1998; Sintonen,
1996; Jung, 1996; Aliseda, 1997, 2001). The basic point is that even
though old models of formal and symbolic logic are not adequate,
it is possible to develop new formal models and tools that are more
appropriate for this (see also Thagard, 2000). In this sense there is
no need to leave “the context of discovery” only to psychologists
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 281

or to empirical scientists, for there are also conceptual and logical

issues involved here.
The point in this paper has been that the logical apparatus should
be broadened at least in two dimensions from the “standard” way.
Deductive logic is not enough, and the model of abductive infer-
ence is especially needed in order to understand the processes of
discovery. But abduction in itself is not enough. Besides validity
considerations (which are in themselves important in abduction)
there is the art of using (abductive) reasoning. This concerns the area
of “Methodeutic” (in Peircean terms); an area especially important
in abductive reasoning. It seems that this is in some sense a return
to a pre-Fregean sense of logic, where the boundaries between
logic and methodology are quite vague (Aliseda, 2001); bound-
aries between logic and psychology must also be drawn anew (see
Thagard, 2000). In any case, this shift to focus more on processes of
inquiry necessitates a re-consideration of abductive strategies.

REFERENCES

Achinstein, P.: 1970, Inference to Scientific Laws. In R.H. Stuwer (ed.),

Minnesota Studies in the Philosophy of Science, Vol. 5, 87–111.
Achinstein, P.: 1971, Law and Explanation. An Essay in The Philosophy of
Science. Oxford: Clarendon Press.
Achinstein, P.: 1987, Scientific Discovery and Maxwell’s Kinetic Theory, Philos-
ophy of Science 54: 409–434.
Aliseda, A.: 1997, Seeking Explanations: Abduction in Logic, Philosophy of
Science and Artificial Intelligence. Institute for Logic, Language, and Compu-
tation. Dissertation Series, University of Amsterdam, the Netherlands.
Aliseda, A.: 2001, A Place for Logic in Scientific Methodology: A Second Oppor-
tunity? A paper presented at the Model-Based Reasoning: Scientific Discovery,
Technological Innovation, Values Conference. Pavia, Italy, May 17–19.
Anderson, D.R.: 1987, Creativity and the Philosophy of C.S. Peirce. Dordrecht:
Martinus Nijhoff Publishers.
Barlow, N. (ed.): 1958, The Autobiography of Charles Darwin, 1809–1882, with
original omissions restored. St James’s Place, London: Collins.
Darwin, F.: 1892, Reminiscences of My Father’s Everyday Life. In F. Darwin
(ed.), Charles Darwin: His Life Told in an Autobiographical Chapter, and in
a Selected Series of His Published Letters. London: John Murray, Albemarle
Street 66–103.
Eco, U. and T.A. Sebeok (eds.): 1988, The Sign of Three. Dupin, Holmes, Peirce.
USA: Indiana University Press.
282 SAMI PAAVOLA

Fann, K.T.: 1970, Peirce’s Theory of Abduction. The Hague: Martinus Nijhoff.
Frankfurt, H.G.: 1958, Peirce’s Account of Inquiry. The Journal of Philosophy
55(14): 588–592.
Gruber, H.E.: 1981, On the Relation Between ‘Aha Experiences’ and the
Construction of Ideas, History of Science 19: 41–59.
Hanson, N.R.: 1961, Is There a Logic of Scientific Discovery. In H. Feigl and G.
Maxwell (eds.), Current Issues in the Philosophy of Science. New York: Holt,
Rinehart and Winston, Inc.
Hanson, N.R.: 1965, Notes Toward a Logic of Discovery. In R.J. Bernstein (ed.),
Perspectives on Peirce. New Haven/London: Yale University Press, 42–65.
Hanson, N.R.: 1972, Patterns of Discovery. Cambridge: University Press, first
printed 1958.
Hintikka, J.: 1985, True and False Logic of Scientific Discovery, Communication
and Cognition 18(1/2): 3–14.
Hintikka, J.: 1989, The Role of Logic in Argumentation, Monist 72: 3–24.
Hintikka, J.: 1998, What is Abduction? The Fundamental Problem of Contem-
porary Epistemology, Transactions of the Charles S. Peirce Society 34(3):
503–533.
Hintikka, J.: 1999, Is Logic the Key to All Good Reasoning? In J. Hintikka (ed.),
Inquiry as Inquiry: A Logic of Scientific Discovery, Jaakko Hintikka Selected
Papers, Volume 5. Dordrecht: Kluwer Academic Publishers.
Hoffmann, M.: 1999, Problems with Peirce’s Concept of Abduction, Foundations
of Science 4: 271–305.
Jung, S.: 1996, The Logic of Discovery. An Interrogative Approach to Scientific
Inquiry, American University Studies, Series V, Philosophy, Vol. 168. New
York: Peter Lang.
Kapitan, T.: 1992, Peirce and the Autonomy of Abductive Reasoning, Erkenntnis
37: 1–26.
Kuhn, T.S.: 1970, The Structure of Scientific Revolutions, 2nd edn., enlarged.
Chicago: University of Chicago Press.
Laudan, L.: 1980, Why Was the Logic of Discovery Abandoned. In T Nickles
(ed.), Scientific Discovery, Logic, and Rationality. Dordrecht: D. Reidel
Publishing Company, 173–183.
Liszka, J.J.: 1996, A General Introduction to the Semeiotic of Charles Sanders
Peirce. Bloomington and Indianapolis: Indiana University Press.
Magnani, L.: 1999. Model-Based Creative Abduction. In L. Magnani, N.J.
Nersessian and P. Thagard (eds.), Model-Based Reasoning in Scientific
Discovery. New York: Kluwer Academic/Plenum Publishers.
Nickles, T.: 1980a, Introductory Essay: Scientific Discovery and the Future of
Philosophy of Science. In T. Nickles (ed.), Scientific Discovery, Logic, and
Rationality. Dordrecht: D. Reidel Publishing Company.
Nickles, T. (ed.): 1980b, Scientific Discovery, Logic, and Rationality. Dordrecht:
D. Reidel Publishing Company.
Nickles, T.: 1989, Heuristical Appraisal: A Proposal, Social Epistemology 3(3):
175–188.
 ABDUCTION AND THE IMPORTANCE OF STRATEGIES 283

Niiniluoto, I.: 1999a, Defending Abduction. Philosophy of Science 66: S436–

S451.
Niiniluoto, I.: 1999b, Abduction and Geometrical Analysis. Notes on Charles S.
Peirce and Edgar Allan Poe. In L. Magnani, N.J. Nersessian and P. Thagard
(eds.), Model-Based Reasoning in Scientific Discovery. New York: Kluwer
Academic/Plenum Publishers.
Peirce, C., S. (CP): 1931–1958, Collected Papers of Charles Sanders Peirce, vols.
1–6, C. Hartshorne and P. Weiss (eds.); vols. 7–8, A.W. Burks (ed.). Cambridge,
MA: Harvard University Press.
Peirce, C., S. (HP): 1985, Historical Perspectives on Peirces Logic of Science. A
History of Science, 2 vols., Carolyn Eisele (ed.). Berlin: Mouton Publishers.
Peirce, C., S. (EP2): 1992–1998, The Essential Peirce. Selected Philosophical
Writings, vol. 2 1893–1913, the Peirce Edition Project (ed.). Bloomington and
Indianapolis: Indiana University Press.
Pera, M.: 1994, The Discourses of Science. Chicago: The University of Chicago
Press.
Reichenbach, H.: 1938, Experience and Prediction. An Analysis of the Founda-
tions and the Structure of Knowledge. Chicago, IL: The University of Chicago
Press.
Rescher, N.: 1995, Peirce on Abduction, Plausibility, and the Efficiency of
Scientific Inquiry. In N. Rescher (ed.), Essays in the History of Philosophy.
Avebury: Aldershot.
Schon, D.: 1959, Comment on Mr. Hanson’s “The Logic of Discovery”, The
Journal of Philosophy 56(11): 500–503.
Shelley, C.: 1996, Visual Abductive Reasoning in Archeology, Philosophy of
Science 63: 278–301.
Sintonen, M.: 1996, Structuralism and the Interrogative Model of Inquiry. In W.
Balzer and C.U. Moulines (eds.), Structuralist Theory of Science. Focal Issues,
New Results. Berlin/New York: Walter de Gruyter.
Thagard, P.: 2000, Coherence in Thought and Action. A Bradford Book.
Cambridge, MA: MIT Press.
Thagard, P. and C. Shelley: 1997, Abductive Reasoning: Logic, Visual Thinking,
and Coherence. In M.L. Dalla Chiara, K. Doets, D. Mundici and J. van Bentham
(eds.), Logic and Scientific Methods. Dordrecht: Kluwer, 413–427, on-line:
http://cogsci.uwaterloo.ca/Articles/Pages/%7FAbductive.html.

Department of Psychology
Centre for Research on Networked Learning and
Knowledge Building
University of Helsinki
Finland
E-mail: sami.paavola@helsinki.fi