Content uploaded by Douglas Walton
Author content
All content in this area was uploaded by Douglas Walton
Content may be subject to copyright.
File
of
Fallacies
The Appeal to Ignorance, or
Argumentum Ad
Ignorantiam
DOUGLAS WALTON
Department of Philosophy
University of Winnipeg
515 Portage Avenue
Winnipeg
,
Manitoba
R3B
2E9
Canada
The
argumentum ad ignorantiam
is
one of the twenty-something famous
informal fallacies treated (usually in one page or less) in so many of the
current and traditional textbooks in logic and critical thinking. However,
this particular fallacy was not in Aristotle's original list of fallacies in his
On Sophistical Refutations.
As indicated below (section four), it appears
to
have come into the curriculum at some later point
.
It is not an easy
fallacy to teach to beginners, and some of the reasons behind this diffi-
culty are explained below.
1.
FALLACIOUS CASES
One of the most famous cases of
argumentum ad ignorantiam
concerns
the classic case of the McCarthy hearings in the early 1950's. In a series
of televised hearings, Senator Joseph R. McCarthy slanderously accused
many innocent people of being Communists in a witch hunt atmosphere in
which unfounded but highly damaging accusations were often made.
McCarthy would show up with a bulging briefcase full of files on accused
individuals. But, in many cases, little or no real evidence was presented,
and a person was accused on the grounds that there was nothing in
McCarthy's files to disprove his Communist sympathies. In one of their
'find and identify the fallacy' exercises, Copi and Cohen (1994, p. 134)
quote from Richard H. Rovere's book,
Senator Joe McCarthy
:
On the Senate floor in 1950, Joe McCarthy announced that he has penetrated 'Truman's
iron curtain of secrecy.' He had 81 case histories of persons whom he considered to
be Communists in the State Department. Of Case 40, he said, 'I do not have much
information on this except the general statement of the agency that there is nothing in
the files to disprove his Communist connections.'
McCarthy was committing the
argumentum ad ignorantiam
fallacy in this
case. The Latin expression literally means 'argument to ignorance', but
Argumentation
13
: 367-377, 1999.
1999
Kluwer
Academic
Publishers. Printed in the Netherlands.
368
DOUGLAS WALTON
the fallacy is usually called 'argument from ignorance' in the textbooks,
or even more often, 'appeal to ignorance'. What seems to be involved in
this kind of case is a shift in the burden of proof. Instead of proving his
claim by giving evidence to support it, McCarthy based his claim on the
lack of any evidence to disprove the claim. The fallacy is thought to have
been committed in this case because McCarthy argued from a premise of
lack of knowledge (ignorance) to a positive conclusion claiming he knew,
or could prove, that the person in question was guilty of being a Communist
sympathizer.
The argumentum ad ignorantiam
or argument from ignorance is usually
defined as a type of argument of the following form: proposition A is not
known (proved) to be true (false), therefore A is false (true). Arguments
of this form, as indicated by the McCarthy case above, are generally held
by the traditional logic textbooks to be fallacious. The McCarthy case illus-
trates very well how the argument from ignorance can be used as a fallacy.
The tactic, as noted above, appears to involve a reversal of burden of proof.
People were accused of being Communists, which is a serious accusation,
one that should have a burden of proof attached to it. But the accusation
tended to stick if the accused party could not disprove the claim.
In other kinds of examples of this fallacy frequently cited by the text-
books, part of the problem is that the claim is very difficult, or even
i
mpossible to verify, by any acceptable evidence. The ghost example from
Copi (1953, p. 56) will give the reader an indication of why this form of
argument has generally held to be fallacious. According to this example,
we conclude that ghosts exist because no one has ever been able to prove
that they don't exist. Copi goes on to add that the argument from ignorance
is fallacious whenever it is argued that 'a proposition is true simply on the
basis that it has not been proved false, or that it is false because it has not
been proved true.' This statement has remained the same, through all nine
editions, and appears in bold print in the ninth edition (Copi and Cohen,
1
994, p. 116). According to the Copi (or Copi and Cohen) account, our
ignorance of how to prove or disprove a proposition does not establish
whether this proposition really is true or false. But why not? The answer
may be that such an argument is an invitation to accept a conclusion based
on mere speculation, as opposed to evidence. According to Adler, 1998a),
arguments from ignorance have the following form: No one has disproved
A, so it is possible that A is true, and therefore we should keep our minds
open with respect to A, and so we may conclude that it is reasonable to
believe that A is true. This type of argument is fallacious. Adler (1998)
cites examples of arguments of this fallacious kind of reasoning to support
claims of alien abduction.
What is characteristic of this kind of case is that it is very
difficult
to
know what would count as good evidence for or against the claim made -
for example, the existence of ghosts, or the reality of alien abductions. In
such cases, there is characteristically a verifiability problem, because any
2.
NONFALLACIOUS CASES
THE APPEAL TO IGNORANCE
369
kind of observational evidence, or evidence which would be reproducible
enough to meet scientific standards of evidence, does not seem to be
available. Or, if it is available, it tends to be very controversial and a lot
of questions are raised about whether it really is evidence. At any rate, it
is characteristic of many of the fallacious arguments from ignorance cited
i
n the logic textbooks that they tend to be about UFO's, the existence of
God, ghosts, the paranormal, and so forth - all subjects in which there is
a verifiability problem in the sense that it would be hard to know what
counts exactly as evidence either for or against the claim. The use of such
examples imports many related, but complicated issues about verifiability
and reproducibility of scientific evidence. In these cases, there are multiple
logical faults involved, but the analysis given by Adler cites the main fault
with them as arguments from ignorance. These are arguments that present
no real evidence, but then use the claim of absence of counter-evidence to
invite a hasty leap that has not been supported by the kind of evidence
that should be required to secure acceptance. However they are analyzed,
such arguments about ghosts and alien abductions are easy to classify as
fallacious appeals to ignorance. But there are many other cases that are
not so easy dismiss as fallacious.
There seem to be many cases where, as a presumptive guide to action, the
ad ignorantiam argument appears to be quite reasonable. One cited in
(
Walton, 1996, p. 86) involves the use of the common rule of safety in the
handling of firearms: if you do not know for sure that a weapon is unloaded,
you should act in accord with the presumption that it is (or may be) loaded.
In a given case, if I walk up to the firing range and pick up a pistol, the
right thing to do before waving it around is to open the chamber and check
to be sure that the weapon is unloaded. The reason justifying my conclu-
sion to act in this way is the practical consideration of safety. Many other
nonfallacious cited were cited by Robinson (1971, pp. 107-108), including
the following argument: 'There is no evidence that quintozene is dangerous
to human beings.' (p. 107). Robinson comments, 'the inference that
quintozene is not dangerous is fair because of the tacit premise that bad
consequences have been searched for, and would have been observed, if
they occurred.' This kind of case must surely be common in scientific
research, where it is called negative evidence (see section three, below).
In other cases, negative evidence, that is, the failure to find some
anticipated evidence, does seem to count as a good reason for concluding
that a proposition is likely to be true or false. Another example cited in
earlier editions of Copi (1953, p. 56) is very interesting to consider in this
connection. In this case, Copi considers a serious FBI investigation that
fails to unearth any evidence at all that a certain person called 'Mr. X' is
370
DOUGLAS WALTON
a communist. It seems reasonable that this failure to find evidence or
negative finding is a good reason to conclude that, in fact, Mr. X is at
least probably not a communist. This case is comparable to many common
kinds of cases where a security agency investigates allegations that someone
who has a security clearance is a spy, or has leaked information. Take note
that the context might be quite different from that of the McCarthy case.
It might be that
of
a normal investigation by a security agency. If the
security agency is competent, and finds no evidence of a transgression,
after a serious and thorough investigation, then most of us might agree
that it could be reasonable to conclude that the suspected party has
(probably, or as far as can be known) not committed the transgression.
Presumably, this type of argumentation is extremely common, and is often
nonfallacious. Copi (p. 56) agrees that, in the FBI case, the argument cited
could be a reasonable argument. But, and here is the problem, he excludes
it
as being an argument from ignorance. Copi would describe such a case
(p. 56) as one where the proof is 'not based on ignorance but on our knowl-
edge that,
if it had occurred, it would be known.'
His
claim is that since
it is not really an argument purely from lack of knowledge, we don't have
to worry about it as constituting a nonfallacious instance of the argument
from ignorance.
Initially however, the FBI case seems to be an
argumentum ad igno-
rantiam
that is not fallacious. It is an argument from lack
of
evidence or
from a negative finding, but nevertheless, provided the search for positive
evidence has been diligent and serious, it seems to be a reasonable argument
to lead to the conclusion that Mr. X is not a communist. However, Copi is
unwilling to categorize this case as an instance of the
argumentum ad
ignorantiam
because, in this case, there is a conditional to the effect that,
if evidence
of
Mr. X's
being a communist were present, the serious search
conducted by the FBI would be likely to find it. It seems then there is a
dispute in the textbook treatments concerning what counts as an
argu-
mentum ad ignorantiam.
According to Copi, the FBI case does not count as a genuine argument
from ignorance. However, it seems that, according to the normal textbook
treatments, many
of
the other texts would count this type
of
case as an
argument from ignorance. And, indeed, it does fit the form
of
argument
we set down above as defining
the
argumentum ad ignorantiam.
Copi
classifies it, however, as not being an argument based on ignorance, but
an argument based on knowledge, as well as perhaps partly on ignorance,
or a
failure to find some outcom
e. Looking back to the analysis
of
the
argument from ignorance given by Adler, (1998, 1998a) above, it would
seem that he would agree with Copi that the FBI case is not an instance
of the argument from ignorance. It would seem that for Copi and Adler, a
true argument from ignorance must be based on pure ignorance, and cannot
be partly based on knowledge as well.
But is this standard for the argument from ignorance so high that no real-
THE APPEAL TO IGNORANCE
371
istic case will ever meet it? For surely most arguments from ignorance in
realistic cases will be partly based on a conditional that, in context, makes
the argument partly also an appeal to positive evidence (knowledge) of
some sort. Robinson (1971, p. 107) even cites this 'redeeming feature' of
arguments from ignorance: 'in good science our ignorance that our gener-
alizations are false is combined with the knowledge that they do explain a
great deal we know'. In general, it would seem to be too narrow, as a
general approach to classification, to require that for an argument to be a
genuine case of an argument from ignorance, it must be based on total igno-
rance, and not be combined with what we do know. Surely the FBI case is
an argument from ignorance, and a typical one at that, even though the
sequence of argumentation in it combines ignorance with knowledge.
However it is to be classified, the FBI case does seem to be a reason-
able argument. If the FBI investigation into the question of Mr. X is a
communist was based on a very thorough, serious and professional search
of Mr. X's background, then it would be reasonable to conclude that,
probably or at least plausibly, Mr. X is not a communist. Of course, such
an argument could be erroneous. It could be that Mr. X is a mole, like
Kim Philby who has left very little or no evidence that an FBI investiga-
tion could find that would indicate that he was a communist. Because of
the existence of so-called moles, this possibility in any security search
always exists. However, if the FBI search was very thorough, then it would
be a reasonable presumptive conclusion that Mr. X is plausibly not a com-
munist and this might be a reasonable basis, for example, for giving Mr.
X a certain level of security clearance or clearing his name from some
scandal or allegation that he is a communist.
It is argued in (Walton, 1996) that spy cases like the case of Mr. X do
contain argumentation of a form that should properly be classified under
the heading of the
argumentum
ad ignorantiam,
that such arguments are
typically based on a combination of ignorance and positive
findings,
of a
kind that can be described as 'knowledge', and that such arguments are
frequently reasonable (as opposed to being fallacious). These three claims
are supported not just by cases like that of Mr. X, but also by arguments
commonly used in scientific research and in a variety of academic disci-
plines.
3
.
USES OF THE ARGUMENT IN
ACADEMIC
DISCIPLINES
The type of negative argument considered above, whether we call it the
argumentum ad ignorantiam,
or use one or another of various names for
it, is a familiar form of argumentation in various academic disciplines. In
history, this type of negative argumentation is called an
ex silentio
argument. For example, it has been argued that the Romans did not award
medals posthumously by citing negative evidence of such posthumous
372
DOUGLAS WALTON
decorations (Maxfield, 1981). Decorations on tombstones and written
writings record no evidence of decorations ever having been given to
soldiers who died in battle. On the other hand, there are many cases of
soldiers who survived who were given decorations. So, we could argue on
negative grounds that, if such a practice had existed, it would probably
have been reflected in some way in the existing evidence of the giving of
awards. But, since there is no known single instance of such an award
having been given, then, on the basis of an
ex silentio
argument, we can
conclude that it appears plausible, generally, that the Romans did not award
medals posthumously.
What is called
negative evidence
in scientific research is the kind of
evidence where an outcome is tested for and does not occur. Negative
evidence in science is regarded as not entirely worthless but, generally,
research articles reporting positive results tend to be highly favored over
those that report negative results. Even so, there appears to be an increasing
number of scientific articles reporting negative evidence. Some feel that
more publishing of negative evidence is good because it cuts down costs of
wasteful research. Others feel that too much publishing of negative results
is
wasteful because the finding of a negative result is not as significant,
generally, as the finding of a positive one. Scientists generally prefer
positive results to negative ones when it comes to publishing their articles
and, in fact, there may be a marked bias in scientific research towards
concentrating on getting positive results (Walton, 1996, p. 68). There seems
to be agreement that negative results are not worthless but they are gen-
erally less worthwhile or less impressive than getting positive results.
In computer science and in the social sciences, the argument from
ignorance is known as the lack of knowledge inference which occurs where
a queried item of information in a database is not found and then the
negative inference is drawn that the proposition is presumably false
.
Whether such a negative or lack of knowledge inference is justified depends
on how complete the database is. An example from Collins, Warnock,
Aiello and Miller (1975, p. 398) from an AI program called Scholar will
illustrate a lack of knowledge inference. Scholar was asked the question,
'Does Guyana produce rubber?' Scholar does know that Peru and Columbia
produce rubber and Scholar is very well informed about the rubber pro-
duction in South America, generally, so Scholar has good reason to think
that, if a country were a major rubber producer, then Scholar would know
that.
However, Scholar does not have any knowledge of whether Guyana
produces rubber or not, that is, neither this proposition nor its negation
are in Scholar's knowledge base, so how should Scholar reply to the
question? Scholar replies as follows: 'I know enough that I am inclined to
believe that rubber is not an agricultural product of Guyana.' Drawing a
lack of knowledge inference, Scholar concludes that, since Guyana is not
in its database as a rubber producer, it can conclude with a moderate degree
of confidence that Guyana does not produce rubber. Of course, this con-
THE APPEAL TO IGNORANCE
373
clusion is only presumptive in nature, but, given that Scholar knows a lot
about rubber producers in South America, it is a conclusion that can be
recommended with a high degree of confidence by Scholar.
Is this lack of knowledge inference an
argumentum ad ignorantiam
in
the logician sense? Once again, there seems to be disagreement on this
issue. According to the forms of argument, we initially proposed Scholar's
inference does seem to be an argument from ignorance, but, according to
Copi's criterion, it would probably not be because Scholar does have some
positive knowledge about rubber producers in South America that enables
it to exclude Guyana. Therefore, Copi might classify this lack of knowl-
edge inference as not being based on ignorance, but rather, as being based
on knowledge.
4.
ORIGINS OF THE ARGUMENTUM AD IGNORANTIAM
Which was the first logic textbook to include the
argumentum ad igno-
rantiam
as a fallacy is not known. The first occurrence of it as a textbook
fallacy that I know of is in Watts'
Logick,
as cited by Hamblin (1970, p.
164).
Watts based his treatment on that of Locke. The first known appear-
ance of the
argumentum ad ignorantiam
as a distinctive type of argument
dates from Locke's
An Essay Concerning Human Understanding
(1690)
cited in Hamblin (1970, pp. 159-160). According to Hamblin (p. 161),
Locke claims to have given the name
argumentum
ad ignorantiam
to this
type of argument, but he does not condemn it as a fallacy. According to
Hamblin, Locke 'stands poised between acceptance and disapproval' (p.
161). Locke described the
argumentum ad ignorantiam
as a way that 'men
ordinarily use to drive others and force them to submit their judgements
and receive the opinion in debate.' Locke defined this type of argument as
the kind of move where one party in such a debate requires the other party
to admit what the first party alleges as a proof or assign a better. In other
words, what the arguer is saying is, 'I offered you what I think constitutes
a proof, so we have to tentatively accept it unless you can offer a proof to
the contrary.' In other words, the arguer is saying he has a right to put this
proposition forward as a judgment that both parties should receive or
accept, at least tentatively, until the other party can disprove it, or put some
proposition in its place that is proved.
What Locke is clearly talking about here is a kind of situation in a debate
where one party asks the other party to tentatively adopt a particular propo-
sition as an
assumption
or a presumption even though full evidence to
support it is not available, provided that better evidence against it is also
not available. This seems like it could frequently be quite a reasonable
and a familiar type of argumentation. In fact it is such a common type of
argument in everyday reasoning that it is surprising that it was not recog-
nized as a 'topic' or commonplace type of argument tactic before Locke.
374
DOUGLAS WALTON
The question then is whether Aristotle might have recognized it as such,
even though it was not listed as one of his set of fallacies in
On Sophistical
Refutations.
I have stated myself (Walton, 1996, p. 36) that although
Aristotle frequently wrote of syllogisms that are fallacious because both
premises are negative, and Sextus Empiricus wrote much on the topic of
suspending belief in cases of insufficient evidence, I found no passage
expressing the idea of argument from ignorance after searching through the
works of Aristotle and Sextus. Now, however, I can report that I have found
such a passage. In
Topics
(158a3-5), when Aristotle explains many methods
to persuade a respondent in dialectic, he includes the following advice:
'One should put forward propositions that hold true of several cases, and
to which either no objection whatever appears or at least not any on the
surface; for when people cannot see any case in which it is not so, they
admit it for true.' (Barnes, 1984, p. 266). In this passage, I would say that
Aristotle does recognize what amounts to the form of argumentation Locke
called the
argumentum ad ignorantiam.
It is worth emphasizing, however,
that
Aristotle is not saying that this
form
of
argument is fallacious. He
sees it as an argumentation tactic that can be used reasonably to persuade
an audience, or an opponent in a dialectical exchange. Of course, there is
the possibility that it could be used to deceive, as well.
Locke clearly indicates that, in his view, it could be used fallaciously
where the one party tries to drive the other or force him to submit to a
judgment by using this kind of negative argumentation. The problem is that
Locke doesn't give us any criteria for determining, in a particular case,
when such an argument for ignorance is reasonable versus when it is used
fallaciously. The only hint given by Locke is that such an argument is used
fallaciously when it is used too forcefully in the debate, that is, where the
one party somehow drives the other or forces the other to submit to the
argument suggesting that the one party is being somehow unfairly aggres-
sive in the debate. Locke also indicates, however, that the argument from
ignorance is a weak type of argument which should give way to arguments
arising from probability or from the nature of things themselves. He sees
the argument from ignorance as a kind of argument that does not help me
directly to truth, in the same way that an argument based on hard evidence
might, but, at best, disposes me for the reception of truth, as he puts it
(see Hamblin, 1970, p. 160).
5
.
SHIFTING OF BURDEN OF PROOF
Two problems have been posed by the consideration of even the small range
of standard cases considered above. One is the problem of how to define
the
argumentum ad ignorantiam
precisely, and, in particular, to determine
whether cases like the case of Mr. X cited above really should be classi-
fied as arguments from ignorance or not. The second problem is to provide
THE APPEAL TO IGNORANCE
37
5
a
method to evaluate arguments from ignorance, in order to help us
determine in a particular case whether such an argument, once identified
as such, is fallacious or not. The cases studied above suggest that the route
to solving both problems lies in viewing the
ad ignorantiam
argument as
a dialectical exchange between two parties engaged in argumentation with
each other, conventionally called the proponent and the respondent. Krabbe
(1995, pp. 254-257) has shown how the
argumentum ad ignorantiam
can
be modeled as a dialectical shift of the kind that is illustrated by the fol-
lowing short sequence of dialogue.
Proponent: Assertion A.
Respondent: Why A?
Proponent: Why not -A?
As Krabbe points ou
t (p. 256),
if the rules of the dialogue permitted
answering a challenge with another challenge, the dialogue could continue
indefinitely, starting with a next move where the respondent asks 'Why not
not -A?' The problem posed by such possible dialogues is one of shifting
of the burden of proof by one party in a dialogue onto the side of the other
party. A short sequence of dialogue of this kind, used to model a pattern
of dialectical argumentation has been called a
profile
of dialogue
(
Krabbe,
1992, p. 277), and the technique of applying profiles of dialogue to many
common cases of the
argumentum ad ignorantiam
has recently been used
in (Walton, 1998). But what, in particular, marks out the fallacious uses of
the argument from ignorance in individual cases? Does the profile in the
fallacious cases have some characteristic pattern or marking?
One such dialectical pattern has been studied by Krabbe (1995, p. 258),
of a kind that corresponds to the fallacy van Eemeren and Grootendorst
(1987, p. 291) call
absolutizing the success of a defense,
as characterized
by the argumentative move, 'You didn't prove your point, so your point is
false.' The argument from ignorance, analyzed in this way as a dialectical
fallacy, could be described as an exaggerated statement of the results of a
discussion.
It is the tactic of implying that the discussion has already
successfully reached the closing stage, whereas in reality, it should be seen
as still being in the argumentation stage.
Another clue to the fallaciousness of
ad ignorantiam
arguments may be
sought in the fact that they are often weak and presumptive arguments, at
best,
which need to be evaluated on a balance of considerations basis in
light of a larger picture of evidence. When this type of argument is used
fallaciously, as indicated by the McCarthy case in section one above, for
example, it appears that somehow, because of the context, even the small
bit of evidence presented the proponent (or it could even be no real evidence
at all), has such a powerful effect the respondent is put into the position
of having to try to disprove the accusation, in order to defend against it.
It seems then that fallacious arguments from ignorance are often connected
376
DOUGLAS WALTON
with first, a reversal of burden of proof, and second, a difficulty in
fulfilling that burden, once it has been reversed, especially in cases where
genuine evidence isdifficult to find. In such cases, a failure to find evidence
that might help to defend one against the charge may result in the charges
going ahead purely on a basis of innuendo. Instead of fitting into the larger
body of evidence to play its correct role in shifting a balance of consider-
ation by presumption, the
argumentum ad ignorantiam,
i
n such a case, has
an impact far out of proportion to its real weight, and functions as a basis
for leading to a conclusion solely on the basis of slander and innuendo.
Part of the explanation of how the fallaciousness of this type of argument
works then, is that it can, as Locke notes, be pursued too aggressively so
that it is used to drive opinion too forcefully to a conclusion that is not
really warranted by the evidence in a given case. In this way, Locke's
explanation of the fallacy turns out to be strikingly similar (or perhaps even
coextensive) with Krabbe's dialectical explanation of it as an exaggerated
statement of the results of a discussion.
ACKNOWLEDGEMENTS
I would like to thank Rob Grootendorst, Hans Hansen, Erik Krabbe, Frans
van Eemeren and John Woods for helpful comments on a previous draft of
this article, and the Social Sciences and Humanities Research Council of
Canada for support in the form of a research grant.
REFERENCES
Adler, Jonathan: 1998, 'Open Minds and the Argument from Ignorance',
Skeptical Inquirer
(Jan.-Feb.), 41-44.
Adler, Jonathan: 1998a
,
'
Arguing from Ignorance
', in
Argumentation and Rhetoric:
Proceedings of the
Second
OSSA Conference,
St.
Catherines, Ontario, OSSA.
Aristotle: 1928,
On Sophistical Refutations,
Loeb Classical Library, Harvard University Press,
Cambridge, Mass.
Aristotle: 1984, 'Topics', in Jonathan Barnes (ed.),
The Complete Works
of
Aristotle, vol.
1,
Princeton University Press, Princeton.
Collins, Allan, Eleanor H. Warnock, Nelleke Aiello and Mark L. Miller: 1975, 'Reasoning
from Incomplete Knowledge,' in Daniel G. Bobrow and Allan Collins (eds.)
,
Representation and Understanding: Studies in Cognitive Science,
Academic Press, New
York, pp. 383-415.
Copi, Irving M.: 1953,
Introduction to Logic, first ed., Macmillan, New York.
Copi, Irving M. and Carl Cohen: 1994,Introduction to Logic,
9th ed., Macmillan, New York.
Hamblin, Charles L.: 1970,
Fallacies,
Methuen, London.
Krabbe, Erik C. W.: 1992
, '
So What? Profiles for Relevance Criticism on Persuasion
Dialogues',
Argumentation
6
,
271-283.
Krabbe, Erik C. W.: 1995, 'Appeal to Ignorance', in Hans V. Hansen and Robert C. Pinto
(eds.),
Fallacies:
Classical and Contemporary Readings, The Pennsylvania State
University Press, University Park, Pa., pp. 251-264.
THE APPEAL TO IGNORANCE
377
Locke
,
John: 1961
,
An Essay Concerning Human Understanding
(1690), ed. John W. Yolton,
Dent, London.
Maxfield, Valerie: 1981,
The
Military
Decorations of the Roman Army,
University of
California Press, Berkeley.
Robinson, Richard: 1971, 'Arguing from Ignorance',
The Philosophical Quarterly
21
,
97-108.
Eemeren, Frans H. van and Rob Grootendorst: 1987, 'Fallacies in Pragma-Dialectical
Perspective',
Argumentation
1
,
283-301.
Walton, Douglas: 1996,
Arguments from Ignorance,
Penn State Press, University Park,
Pennsylvania.
Walton, Douglas: 1998, `Profiles of Dialogue for Evaluating Arguments from Ignorance',
Argumentation
13
,
53-71.
Watts, Isaac: 1725,
Logick: Or, The Right Use of Reason in the Inquiry After Truth,
15th
ed., London, Printed for J. Buckland et al., 1772.
