The Relation between Formal and Informal Logic*

RALPH H. JOHNSON

Department of Philosophy
University of Windsor
Windsor, Ontario
N9B 3P4 Canada

ABSTRACT: The issue of the relationship between formal and informal logic depends
strongly on how one understands these two designations. While there is very little dis-
agreement about the nature of formal logic, the same is not true regarding informal logic,
which is understood in various (often incompatible) ways by various thinkers. After reviewing
some of the more prominent conceptions of informal logic, I will present my own, defend
it and then show how informal logic, so understood, is complementary to formal logic.

KEY WORDS: Argument, argumentation, formal logic, informal logic, implication, formal,
theory of argument

I. INTRODUCTION

My purpose in this paper is to differentiate between formal and informal

logic. I argue that they are to be differentiated both by their domains and
by their approaches.
I begin with some preliminary observations.
Note 1: I am going to assume that it is clear what is meant by formal
logic, or rather formal deductive logic (FDL), and that the main problem
in determining the relationship between it and informal logic arises because
of the other half of the equation still lacks a clear identity. Walton and
Brinton (1997) write:
Informal logic has yet to come together as a clearly defined discipline, one organized
around some well-defined and agreed upon systematic techniques that have a definite
structure and can be decisively applied by users. (p. 9)

Indeed, some (Hintikka (1989), perhaps Woods (1980), perhaps Massey

(1981)) will say that there is no such thing as informal logic. Even among
those who think there is such an inquiry, there is no clear consensus about
what it is. In (1997), Blair and I discuss 7 different senses (some of which
I will mention shortly). Thus the term has much less a fixed meaning/appli-
cation than does its counterpart. In clarifying the relationship between the
two, I assume that the larger part of the task is to clarify the nature of
informal logic, and accordingly most of my remarks in this paper are aimed
in that direction.
Note 2: Given the imprecision of the term ‘informal logic,’ any position

Argumentation 13: 265–274, 1999.

 1999 Kluwer Academic Publishers. Printed in the Netherlands.
266 RALPH H. JOHNSON

on its nature and its relationship to formal logic must be seen as stipula-
tive, as a recommendation about how we ought to see their respective
mandates.
Note 3: It seems obvious that ‘informal logic’ must in no small measure
derive its sense in contradistinction to ‘formal logic.’ To understand
informal logic, then, we must understand in what sense it is not formal. In
proceeding this way, I do not assume that there is any inherent opposition
between formal logic and informal logic. Indeed, properly understood,
informal logic and formal logic have complementary tasks.
It is true, historically, that many once believed that the two are opposed.
This comes out, for example, in the critique of formal logic that we find
in the seminal work of Perelman (1958) and Toulmin (1958). It is also
present in our (1980) article, as well as Scriven’s paper: ‘The emergence
of informal logic marks the end of the reign of formal logic’ (1980, p. 147).
It also is suggested by the very name ‘informal logic’ which hints at some
negation of formal logic. But very few any longer think that the two are
opposed.1
Note 4: Informal logic is often closely associated with critical thinking.2
Many authors treat them as virtually synonymous. Thus Copi writes:
‘Although I have been teaching undergraduate logic courses in logic since
1939, it was only about a half a dozen years ago that I first taught a
course in the subject now known variously as “informal logic” or “critical
thinking” ’ (1986, p. vii).
Yet informal logic, for all of its closeness to critical thinking, is impor-
tantly different from it. Critical thinking refers to a particular educational
ideal and practice which can be enriched by informal logic (and much else
besides).3 In my view, informal logic is best positioned somewhere between
the concern for enlightened practice and educational reform that drives
the critical thinking movement, and the concern for theory that lies behind
the development of FDL.
I shall begin by taking note of the range of usage of ‘informal logic’ and
then focus on those uses I take to be central. Then borrowing a useful inter-
vention from Barth and Krabbe (1982), I sort through those uses to see how
this intervention helps clarify the nature of informal logic. My own state-
ment about the nature of informal logic follows, and that in turn leads to
the last section in which I address the question of the relationship between
formal and informal logic.

II. SOME CONCEPTIONS OF INFORMAL LOGIC

I am going to run through an inventory of responses to the question: ‘Just

what is informal logic?’ beginning with what many consider a locus clas-
sicus – Ryle’s (1954) statement, according to which informal logic refers
to the logic of substantive concepts (like time and pleasure) whose logic
 THE RELATION BETWEEN FORMAL AND INFORMAL LOGIC 267

is ‘informal,’ as opposed to the logic of concepts like conjunction and

disjunction, whose logic is formal. In this usage, ‘informal logic’ becomes
roughly equivalent to what many would call conceptual analysis, or to
what Wittgenstein (1951, #664) would call the ‘depth grammar’ of an
expression.
A second conception of informal logic is that it is or is related to the
study of the informal fallacies (Carney and Sheer, 1964; Kahane, 1971;
Woods, 1980).
A third conception of informal logic is that it is formal logic without
the formalism. This idea is perhaps illustrated by the way that Copi
produced his Informal Logic (1986) text; that is, what Copi does is delete
the formal logic section from his (1982) Introduction to Logic text.
A fourth conception is that informal logic has the task of mediating
between formal logic and reasoning in natural language. Such a view can
be found in Goldman (1986) and Woods (1995). Thus Goldman writes
(1986, p. 82):
It is widely assumed that logic deals with principles of good reasoning. Logic is often
characterized as the art of reasoning. Unfortunately, such a billing is a bit of a sham. It
isn’t that logic courses are not useful for good reasoning; it’s just that there are no well-
established principles of good reasoning (good cognitive state transitions), and no satis-
factory theory of how good reasoning is related to formal logic. In short, there is not a
really well-established discipline of informal logic.

In (1995), John Woods shows the formidable obstacles formal logic faces
in coming to grips with arguments in natural language. His argument
implies that informal logic is in even worse shape. In fact, Woods suggests
there cannot be such a thing as informal logic, understood as the mediator
between natural language and the propostional calculus (PC).
A fifth conception of informal logic – one that finds favour with authors
such as McPeck (1981), Siegel (1988), and Weinstein (1994) – is that
informal logic is applied epistemology. Informal logic is the application
of epistemological findings to the evaluation of arguments. In this vein,
both Scriven and Finocchiaro take informal logic to be essentially theory.
Finocchiaro (1984) takes it to be the theory of reasoning. Scriven (1993)
sees it as ‘the metatheory of critical thinking.’
From this inventory, it is apparent that there is quite a range of opinion
concerning what informal logic is. It is further apparent that there is some
incompatibility in these views. If Ryle is right, then it is hard to see how
the others could be correct. If McPeck, Siegel and Weinstein are right, it
is hard to see how Goldman could be right. Such divergence might be taken
as an indication that Hintikka (1989) is correct when he suggests that there
is no such thing as informal logic. Rather than attempt to adjudicate here
among these rival conceptions, I prefer to direct attention to the views of
two of the leading theorists closely identified with informal logic – Trudy
Govier and Douglas Walton.
268 RALPH H. JOHNSON

III. GOVIER AND WALTON ON INFORMAL LOGIC

A. Govier
In Problems in Argument Analysis and Evaluation, Govier represents
informal logic as the art of argument evaluation, a task which she insists
is nonformal in character:
Informal logicians, contesting the drive for complete rigor and purity among formalists,
contend that the appraisal of natural arguments requires something other than translation
into a technical formal language and application of formal rules to test validity. However,
it is only in the stricter sense of ‘formal’ that informal logicians are committed to the
view that a theory of natural argument will not be a formal theory. (p. 14)

(Here Govier adumbrates the distinction I introduce below.) For Govier,

informal logic is the logic of real arguments:
To speak of informal logic is not to contradict oneself but to acknowledge what should
be obvious: that the understanding of natural arguments requires substantive knowledge
and insight not captured in the rules of axiomatized systems. The informal fallacies,
historically a central topic for informal logic, involve mistakes in reasoning which are
relatively common, but neither formal nor formally characterizable in any useful way.
The fact that an account of an informal fallacy makes it out to be just that does not show
that it is imprecise or lacking in rigor. (p. 204)

Thus for Govier informal logic is the logic which helps evaluate what she
calls natural arguments – a process requiring substantive knowledge and
insights not provided by formal logic. Noteworthy as well is Govier’s con-
nection of informal logic with the informal fallacies.

B. Walton
In his 1990 paper, ‘What is Reasoning? What is an Argument?’ Walton
deals explicitly with the issue of the relationship between formal and
informal logic:
Formal logic has to do with the forms of argument (syntax) and truth values (seman-
tics). . . . Informal logic (or more broadly, argumentation, as a field) has to do with the
uses of argumentation in a context of dialogue, an essentially pragmatic undertaking.
(pp. 418–419)

Thus for Walton, informal logic deals with the pragmatic aspects of
argument, whereas formal logic deals with the semantic and syntactic
aspects of it. The important point, it seems to me, is that Walton takes argu-
mentation to be the domain of informal logic.
Govier and Walton thus agree that informal logic is concerned with argu-
mentation. However, this insight won’t go very far in differentiating it from
formal logic which is also, it is still widely believed, concerned with argu-
mentation. So let us ask how informal logic’s approach to argumentation
differs from that of formal logic. In the next section, I introduce the afore-
mentioned Barth-Krabbe distinction to see what illumination it offers.
 THE RELATION BETWEEN FORMAL AND INFORMAL LOGIC 269

IV. THE BARTH-KRABBE DISTINCTION

An obvious point is that ‘informal’ derives its meaning in contrast to its

counterpart – ‘formal.’ I will follow Barth-Krabbe’s account (1982, p. 14f )
where they distinguish three senses of the term ‘form.’
By ‘form1,’ Barth and Krabbe mean the sense of the term which derives
from the Platonic idea of form, where form denotes the ultimate meta-
physical unit. Barth and Krabbe claim that most traditional logic is formal
in this sense (i.e., syllogistic logic). That is, syllogistic logic is a logic of
terms, where the terms might naturally be understood as names for Platonic
(or Aristotelian) forms. In this first sense of ‘form,’ almost all logic would
be informal (not-formal). Certainly neither propositional nor the predicate
logic can be construed as a term logic. So the understanding of informal
logic this leads to is too broad to be useful. For it would mean that the
propositional calculus was informal logic – clearly not a helpful result. This
sense of ‘form,’ then, will not help us pinpoint the nature of informal
logic.
By ‘form2,’ Barth and Krabbe mean the logical form of statements as
these are understood in modern logic. In this sense, one could say that the
syntax of the language to which a statement belongs is very precisely for-
mulated or ‘formalized’; or that the validity concept is defined in terms of
the logical form of the statements that make up the argument. In this sense
of ‘formal,’ most modern and contemporary logic is ‘formal.’ That is, such
logics rely heavily on the notion of logical form when it comes to under-
standing the structure of argument, and they embrace validity as the central
construct in normative theory: validity is a necessary condition for a good
argument.
In this second sense of form, informal logic is not formal, because it
abandons the notion of logical form as the key to understanding the struc-
ture of argument, and it likewise abandons validity as constitutive for the
purposes of the evaluation of argument. When Govier speaks of informal
logic, I believe that it is this second sense of ‘formal’ that stands in the
background as the point of contrast. I believe this is the appropriate sense
of ‘form’ to use in our attempt to pinpoint the identity of informal logic.
It is informal2 logic.
By ‘form3,’ Barth and Krabbe mean to refer to ‘procedures which are
somehow regulated or regimented, which take place according to some set
of rules.’ Barth and Krabbe say that ‘we do not defend formality3 of all
kinds and under all circumstances.’ Rather ‘we defend the thesis that verbal
dialectics must have a certain form (i.e., must proceed according to certain
rules) in order that one can speak of the discussion as being won or lost’
(p. 19). In this third sense of ‘form,’ informal logic can itself also be said
to be formal. That is, there is nothing in the informal logic enterprise that
stands opposed to the idea that argumentative discourse should be subject
to norms, i.e., subject to rules, criteria, standards or procedures. What is
270 RALPH H. JOHNSON

rejected is the view that the appropriate criteria for evaluating arguments
are to be obtained by reflection on the logical form.
I propose the following as a way to construe informal logic that fits with
what both Govier and Walton have said and incorporates the Barth-Krabbe
intervention:
Informal logic designates that branch of logic whose task is to develop non-formal2
standards, criteria, procedures for the analysis, interpretation, evaluation, criticism and
construction of argumentation in everyday discourse. (Johnson and Blair, 1987, p. 148)

If this is how informal logic is best understood, then certain consequences

ensue for the conceptions tabled earlier. For example, it follows that Ryle’s
conception is off kilter. Indeed, as I stated earlier, there appears to be a
ready-made slot for what Ryle calls informal logic. It is what many would
call philosophical or conceptual analysis. The idea that informal logic is
connected with the study of informal fallacies fits well with this concep-
tion, whereas Goldman’s view about informal logic does not. Those who
urge an epistemological approach could be right if, in order to achieve the
tasks mentioned in the definition above, it turns out that informal logic
has must have recourse to epistemology.
Having stated what I take informal logic to be, I now move discuss its
relationship to formal logic.

V. INFORMAL LOGIC AND ITS RELATIONSHIP TO FORMAL LOGIC

If the above position is on the mark, then the relationship between informal
and formal logic is complementary rather than oppositional. Here is how
Walton puts it:
Hence the strongly opposed current distinction between informal and formal logic is really
an illusion, to a great extent. It is better to distinguish between the syntactic/semantic
study of reasoning, on the one hand, and the pragmatic study of reasoning in arguments
on the other hand. The two studies, if they are to be useful to serve the primary goal of
logic, should be regarded as inherently interdependent, and not opposed, as the current
conventional wisdom seems to have it. (p. 418)

Walton suggests one plausible way of viewing the relationship: each studies
argument but they take different approaches to it. While it is tempting to
take this road I believe it should be resisted because (unless one holds the
view that argument is essentially identical to implication) such a resolution
masks a deeper difference. Formal logic (as traditionally understood) and
informal logic (as defined above) have in fact different domains.
Formal logic (FDL) is not really concerned with argumentation. There
are a number of reasons to think this. If we locate the contemporary origins
of FDL in Frege’s attempt to solve problems in the foundations of mathe-
matics, in Russell and Whitehead’s program of showing the reducibility
of mathematics to logic, then it becomes evident that FDL as historically
 THE RELATION BETWEEN FORMAL AND INFORMAL LOGIC 271

defined) is not really concerned with argumentation considered as an instru-

ment of rational persuasion.
When FDL is construed as a theory of argument, it is clear that it fails.
(This is not really a criticism of FDL since, in my view, it was never
intended to achieve this.) To see this, I start with the assumption that there
can be good arguments both for and against a given proposition: P. Hamblin
(1970) is clear on this point: ‘There are often good arguments for a given
conclusion and also good arguments against it’ (p. 232). The history of
philosophy bears witness to it. There are good arguments for idealism
[Plato] and against it [Aristotle]; for phenomenalism [Berkeley, Ayer] and
against it [Hume, Wittgenstein]; for skepticism [Descartes, Montaigne] and
against it [Moore, Wittgenstein]. Looking now outside the realm of philo-
sophical argumentation, I will note that the Supreme Court of the United
States not infrequently renders 5-4 verdicts. These are individuals who are
practiced in both constructing and appraising arguments, the arguments
are made by highly competent lawyers, the justices are impartial, they
have as much time as they need, they have unlimited research resources
available. Still they very often come with split verdicts which again
suggests that there can be very good arguments both for and against a given
proposition.
FDL fails as a theory of argument because it cannot accommodate this
important condition: i.e., that there can be good arguments for and against
a given proposition. It is true that there can be a valid argument for P and
a valid argument for –P. But there can not be a sound argument for P and
a sound argument for –P. This is obvious by reductio reasoning. If there
were a sound argument for P, then P would be true. If there were a sound
argument for –P, then –P would be true. But it cannot be that P and –P are
true. Hence there cannot be a sound argument for P and a sound argument
for –P.4
If, then, we take goodness in argumentation to be soundness, it follows
that there cannot be good arguments for P and –P. But we know that this
is not true: there can be good arguments for both P and –P. Hence the
goodness exhibited in the practice of argumentation is not the goodness
prescribed by FDL, at least when we understand that as the doctrine of
soundness. If we want a theory that illuminates our best argumentative prac-
tices, then we must look elsewhere. That ‘elsewhere’ is informal logic.
If FDL is not conerned with argumentation, then what is its province?
The province of FDL might be specified as the study of the entailment
relationship between propositions.5 FDL is concerned to clarify such rela-
tions as logical truth, logical consistency and logical equivalence.6 Its
approach to these matter is form2al in that it relies on the notion of logical
form.
I need to add one further claim here: arguments are not entailments/
implications, though they may sometimes contain them or have the struc-
ture of one. I cannot argue for this point here.7
272 RALPH H. JOHNSON

The province of informal logic is argumentation, or as I prefer, the

practice of argumentation. Specifically, it looks at arguments as instruments
of rational persuasion. Its approach to these matters is not form2al because
it does not rely on either the standard notion of logical form nor does it
take validity as normative.
In sum, informal logic and formal logic differ in their in their subject
matter. FDL studies the entailment relationship between propositions;
informal logic is about argumentation. They likewise differ in their
approaches: formal as against informal, as explained above. Hence informal
logic and FDL are complementary rather than oppositional.8

VI. CONCLUSION

There was a time when many of us who call ourselves informal logicians
thought that there was some opposition or tension between formal and
informal logic; we were, it now seems, caught in a false dilemma: either
formal logic or informal logic. The time has come to put these turf wars
behind us. Obviously we need both logics, for there is plenty out there
that requires our collective attention.

NOTES

* This paper was presented at the AILACT session of the Western Division meetings
of the American Philosophical Association, Chicago, May 7, 1998. I am grateful to my co-
discussants, John Woods and Gerald Massey, for their comment and criticisms, and to John
Hoaglund for his comments and for organizing the session. I am also grateful to two referees
of this journal for their very helpful comments.
1
But see the Walton passage quoted below on page 16.
2
The very name of the association sponsoring this session – the Association for Informal
Logic and Critical Thinking – suggests such a relationship.
3
See my (1992) for my account of the nature of critical thinking.
4
I am grateful to David Hitchcock for this shorter way of demonstrating the inadequacy
of FDL as a theory of argument.
5
Since these formal systems were developed in close association with research into the
foundations of mathematics, this inquiry has come to be known as mathematical logic or
symbolic logic.
6
Thanks again are due to David Hitchcock for his comments on an the paper referred to
in note 4 in which this issue arose.
7
As a result of what I call the mathematicization of logic in the 20th century, the idea
that argument and implication/ entailment are the same became a fairly standard view. I argue
this point in Manifest Rationality.
8
This position is developed in greater detail in my Manifest Rationality, forthcoming from
Lawrence Erlbaum.
 THE RELATION BETWEEN FORMAL AND INFORMAL LOGIC 273

REFERENCES

Barth, E. M. and E. C. W. Krabbe: 1982, From Axiom to Dialogue: A Philosophical Study

of Logics and Argumentation, Walter de Gruyter, New York/Berlin.
Bentham, J. van et al. (eds.): 1996, Logic and Argumentation, Royal Netherlands Academy
of Arts and Sciences/North Holland Publishing Co., Amsterdam.
Blair, J. A. and R. H. Johnson (eds.): 1980, Informal Logic: The First International
Symposium, Edge Press, Inverness.
Blair, J. A. and R. H. Johnson (eds.): 1994, New Essays in Informal Logic, Informal Logic,
Windsor.
Carney, J. D. and R. K. Scheer: 1964, Fundamentals of Logic, Macmillan, New York (3rd
ed., 1980).
Copi, I.: 1982, Introduction to Logic, Macmillan, New York (6th ed.).
Copi, I.: 1986, Informal Logic, Macmillan, New York.
Eemeren, F. H. van, R. Grootendorst and F. Snoek Henkemans: 1996, Fundamentals of
Argumentation Theory, Lawrence Erlbaum Associates, Mahwah NJ.
Finocchiaro, M.: 1984, ‘Informal Logic and the Theory of Reasoning', Informal Logic 6,
3–8.
Finocchiaro, M.: 1996, ‘Informal Factors in the Formal Evaluation of Arguments’. In: J.
van Bentham (1996).
Freeman, James: 1994, ‘The Place of Informal Logic in Logic’. In: Blair and Johnson (1994),
36–49.
Goldman, A.: 1986, Epistemology and Cognition, Harvard, Cambridge, MA.
Govier, T.: 1987, Problems in Argument Analysis and Evaluation, Foris, Dordrecht.
Hamblin, C. L.: 1970, Fallacies, Methuen, London.
Hansen, H. V. and R. C. Pinto (eds.): 1995, Fallacies. Classical and Contemporary Reading,
Pennsylvania State University Press, University Park, PA.
Hintikka, J.: 1989, ‘The Role of Logic in Argumentation’, The Monist 72, 1–24.
Johnson, R. H.: 1992, ‘The Problem of Defining Critical Thinking’, in S. P. Norris
(ed.), The Generalizability of Critical Thinking, Teachers College Press, New York,
38–53.
Johnson, R. H.: 1996, The Rise of Informal Logic, Vale Press, Newport News, VA.
Johnson, R. H. and J. A. Blair: 1987, ‘The Current State of Informal Logic’, Informal Logic
9, 147–151.
Johnson, R. H. and J. A. Blair: 1996, ‘Informal Logic and Critical Thinking’. In: van Eemeren
and Grootendorst (1996).
Johnson, R. H. and J. A. Blair: 1997, ‘Informal Logic in the 20th Century’, in D. Walton
and A. Brinton (eds.), Historical Foundations of Informal Logic, Ashgate, Aldershot,
158–173.
Kahane, H.: 1971, Logic and Contemporary Rhetoric: The Use of Reasoning in Everyday
Life, Wadsworth, Belmont, CA (6th ed., 1992).
Massey, G.: 1981, ‘The Fallacy behind Fallacies’, Midwest Journal of Philosophy 6, 489–
500.
McPeck, J.: 1981, Critical Thinking and Education, Martin Robertson, Oxford.
Perelman, Ch. & L. Olbrechts-Tyteca: 1958, La Nouvelle Rhetorique, Presses Universitaires
de France, Paris.
Ryle, G.: 1954, Dilemmas, Cambridge University Press, Cambridge.
Scriven, M.: 1980, ‘The Philosophical Significance of Informal Logic’. In: Blair and Johnson
(1980), 148–160.
Scriven, M.: 1993, Defining Critical Thinking, (Unpublished manuscript).
Siegel, H.: 1988, Educating Reason: Rationality, Critical Thinking and Education, Routledge,
New York and London.
Toulmin, S.: 1958, The Uses of Argument, Cambridge University Press, Cambridge
274 RALPH H. JOHNSON

Walton, D.: 1990, ‘What is Reasoning? What is an Argument?’, The Journal of Philosophy
87, 399–419.
Walton, D. and A. B. Brinton (eds.): 1997, Historical Foundations of Informal Logic,
Ashgate, Aldershot.
Walton, D. and A. Brinton: 1997, ‘Introduction’. In: Walton and Brinton, 1–11.
Weinstein, M.: 1994, ‘Informal Logic and Applied Epistemology’. In: Blair and Johnson
(1999), 140–161.
Wittgenstein, L.: 1951, Philosophical Investigations. Trans. G. E. M. Anscombe. Macmillan,
New York.
Woods, J.: 1980, ‘What is Informal Logic?’ In: Blair and Johnson (1980), 57–68.
Woods, J.: 1995, ‘Fearful Symmetry’. In: Hansen and Pinto (1995), 181–193.