Rosen - The Philosophy of Error and Liberty of Thought JS Mill on Logical Fallacies


J.S. Mill on Logical Fallacies  121 

The Philosophy of Error and Liberty of 
Thought: J.S. Mill on Logical Fallacies 

University College London FREDERICK ROSEN 

© Informal Logic Vol. 26, No. 2 (2006): pp. 121-147. 

Keywords: John Stuart Mill, philosophy of error, fallacies, liberty of thought, truth, 
history of logic, Bacon, Socrates, Aristotle 

 Abstract: Most recent discussions of John 
Stuart Mill’s System of Logic (1843) neglect the 
fifth book concerned with logical fallacies.  Mill 
not only follows the revival of interest in the 
traditional Aristotelian doctrine of fallacies in 
Richard Whately and Augustus De Morgan, but 
he also develops new categories and an original 
analysis which enhance the study of fallacies 
within the context of what he calls ‘the 
philosophy of error’.  After an exploration of 
this approach, the essay relates the philosophy 
of error to the discussion of truth and error in 
chapter two of On Liberty (1859) concerned 
with freedom of thought and discussion. 
Drawing on Socratic and Baconian perspectives, 
Mill defends both the traditional study of logic 
against Jevons, Boole, De Morgan, and others, 
as well as the study of fallacies as the key to 
maintaining truth and its dissemination in 
numerous fields, such as science, morality, 
politics, and religion.  In Mill’s view the study 
of fallacies also liberates ordinary people to 
explore the truth and falsity of ideas and, as 
such, to participate in society and politics and 
develop themselves as progressive beings. 

I 

Logic lays down the general principles and laws of the search after truth; the 
conditions which, whether recognised or not, must actually have been 
observed if the mind has done its work rightly. Logic is the intellectual 

Résumé: Les discussions les plus 
récentes sur le Systems of Logic (1843) de 
John Stuart Mill négligent le cinquième 
livre, qui traite des sophismes. Mill suit 
non seulement la reprise d’intérêt dans la 
doctrine aristotélicienne traditionnelle de 
Richard Whately et d’Augustus DeMor- 
gan, mais développe aussi des nouvelles 
catégories et une analyse originale qui 
améliore l’étude des sophismes dans le 
contexte de ce qu’il appelle «la 
philosophie de l’erreur». J’explore cette 
approche, et ensuite je relie la philosophie 
de l’erreur à la discussion de la vérité et de 
l’erreur dans le deuxième chapitre de On 
Liberty (1859) qui traite de la liberté de 
pensée et de discussion. Mill s’inspire 
des perspec-tives de Socrate et de Bacon 
pour défendre l’étude traditionnelle de la 
logique contre Jevons, Boole, DeMorgan, 
et d’autres, ainsi que l’étude des 
sophismes comme moyen clé pour 
maintenir la vérité et sa dissémination dans 
plusieurs champs, tels que la science, la 
moralité, la politique, et la religion. Selon 
Mill, l’étude des sophismes libère les gens 
ordinaires à explorer la vérité et la fausseté 
des idées, à participer dans la société et la 
politique, et à s’instruire. 



122     Frederick Rosen 

complement of mathematics and physics. Those sciences give the practice, 
of which Logic is the theory. It declares the principles, rules, and precepts, of 
which they exemplify the observance. 

——— 
Of Logic I venture to say, even if limited to that of mere ratiocination, the 
theory of names, propositions, and the syllogism, that there is no part of 
intellectual education which is of greater value, or whose place can so ill be 
supplied by anything else.  Its uses, it is true, are chiefly negative; its function 
is, not so much to teach us to go right, as to keep us from going wrong.  But 
in the operations of the intellect it is so much easier to go wrong than right; 
. . . Logic points out all the possible ways in which, starting from true premises, 
we may draw false conclusions. 

——— 
Logic is the great disperser of hazy and confused thinking:  it clears up the 
fogs which hide from us our own ignorance, and make us believe that we 
understand a subject when we do not. 

——— 
To those who think lightly of school logic, I say, take the trouble to learn it. 
You will easily do so in a few weeks, and you will see whether it is of no use 
to you in making your mind clear, and keeping you from stumbling in the dark 
over the most outrageous fallacies (CW XXI: 238, 239). 

These brief passages have been taken from a discussion of logic in J.S. Mill’s 
Inaugural Address Delivered to the University of St. Andrews (1867) (CW XXI: 
215-57). They contain not only advice to young students as to how to pursue their 
studies but also allusions to his own intellectual development.1 Here we find a 
restatement of his belief in the role of logic in the search for truth, the relationship 
between logic and other sciences like mathematics and physics, his Socratic belief 
in looking into and through ‘the fogs which hide us from our own ignorance’, and 
his acceptance of Aristotelian scholastic logic with its emphasis on avoiding 
‘stumbling in the dark over the most outrageous fallacies’. 

Mill’s major work on logic, A System of Logic, Ratiocinative and Inductive 
(1843) (CW VII, VIII) is not as carefully studied nowadays as are his more accessible 
works, On Liberty (1859) (CW XVIII: 213-310) and Utilitarianism (1861) (CW 
X: 203-39). Despite this recent neglect, there is a considerable body of philosophical 
literature which concentrates on the Logic, as, for example, in the work of Jackson, 
Anschutz, Ryan, Skorupski, and Scarre, to name a few.2 Unfortunately, none of 
these commentators have taken much, if any, notice of the fifth book of the Logic, 
which is a substantial essay concerned with fallacies (CW VIII: 733-830). Yet, as 
we have seen in the passages quoted above, Mill seemed to place the detection of 
fallacies at the heart of the study of logic. 

At the time Mill wrote the Logic, the inclusion of material on fallacies was 
somewhat contentious. That the study of fallacies (though often rejected in the 



J.S. Mill on Logical Fallacies  123 

modern period) persisted in the early part of the nineteenth century was due to the 
influence of Aristotelian logic, and, particularly, the influence of Aristotle’s On 
Sophistical Refutations (Aristotle 1955: 10-155). That this study of dialectical 
argument should accompany the main demonstrative or didactic arguments centred 
on the syllogism ensured forever, it seemed, that the study of fallacies should form 
part of logic generally. The scholastic tradition in logic, still followed in English 
universities in the eighteenth and early-nineteenth centuries and enshrined in the 
textbooks of Robert Sanderson and Henry Aldrich, faithfully repeated the Aristotelian 
list of thirteen fallacies.3 Instruction in fallacies was doubtless regarded as light 
and often amusing relief from the otherwise tedious study of the syllogism. 

Many modern philosophers, however, had followed their rejection of Aristotelian 
philosophy generally with a rejection or severe curtailment of the usual treatment 
of fallacies. Locke’s Essay Concerning Human Understanding, for example, not 
only failed to consider the traditional list of fallacies, but also contained a critique 
of Aristotle and of the relevance of the syllogism as the instrument of reason and 
source of knowledge (Locke 1975: 671 [IV.XVII.4]). Traditional logic, for Locke, 
was regarded as productive only of useless disputes, which were of little benefit to 
society. ‘For notwithstanding these learned Disputants, these all-knowing Doctors’, 
he wrote, ‘it was to the unscholastick Statesman that the Governments of the 
World owed their Peace, Defence, and Liberties; and from the illiterate and 
contemned Mechanic (a Name of Disgrace) that they received the improvements 
of useful Arts’ (Locke 1975: 495 [III.X.9]). For Locke, and for others in the 
eighteenth and nineteenth centuries, the study of problems in logic, and, particularly, 
those that concentrated on the syllogism, tended to be translated into the study of 
language and/or psychology. In this development (and Locke’s Essay is a good 
example) the traditional exposition of fallacies was omitted, as contributing more 
to disputes and their enjoyment than to the acquisition of knowledge. 

In contrast with the example of modern philosophers rejecting Aristotelian logic 
and the study of fallacies, as derived from the Sophistical Refutations, Mill, as we 
have seen, fully embraced the traditional study of fallacies. This somewhat unusual 
starting point for a nineteenth-century philosopher with thoroughly progressive 
views owed a good deal to Richard Whately’s Elements of Logic (1826) (see 
Whately 1975). Whately sought to revitalize the study of logic at Oxford by restating 
Aristotelian logic in a new context, and, in the process, abandoned what seemed to 
be the list of Aristotelian fallacies in favour of a new schema. The underlying 
momentum for the new context might be described as Coleridgean.  Whately’s 
Elements of Logic and companion work, Elements of Rhetoric (1827) (see Whately 
1828), were first published in the Encyclopaedia Metropolitana, a work inspired 
by Coleridge to advance progressive views which could be seen as emerging out 
of traditional thought and institutions (see Encyclopaedia Metropolitana 1845: i.1- 
43, 193-240, 241-303).Whately’s return to Aristotelian logic, set forth in a 
contemporary context with modern examples and without the baggage of 



124     Frederick Rosen 

scholasticism, seemed to be an ideal expression of this vision.4 I have discussed 
Mill’s encounter with Whately’s Elements of Logic and his essays on ‘Bentham’ 
(1838) (CW X: 75-115) and ‘Coleridge’ (1840) (CW X: 117-63) elsewhere (see 
Rosen 2006, 2007a, 2007b), and will simply note here that Mill’s two essays on 
the leading British thinkers of the age were written and published at the same time 
(between 1838 and 1840) as the book on fallacies in the Logic was drafted and 
completed (CW VII: xcviii-xcix). Mill’s retention and development of the study of 
logical fallacies was linked to his enthusiasm for the belief that the contrary ideas 
of Bentham and Coleridge might be successfully combined in thought and debate. 

There were other factors which might have led Mill to turn to and develop the 
traditional study of logical fallacies. Jeremy Bentham’s Book of Fallacies was 
published in 1824, and Bentham claimed that he was in certain respects the first 
person since Aristotle to reconsider the forms of logical fallacies (Bentham 1824: 
2). Although Bentham was strongly criticized by Whately for the one-sidedness of 
his radicalism (rather than for his treatment of fallacies), Mill thought that Whately 
was ‘unnecessarily severe’ in his criticisms (see Whately 1975: 194n-5n; CW XI: 
31). As we shall see, Mill also adopted one category from Bentham’s work. 
However, Mill never fully endorsed Bentham’s study of fallacies and tended to 
ignore it, as well as George Bentham’s elaborate discussion of fallacies based on it 
(see G. Bentham 1827: 220ff). A more general and pervasive influence of Bentham 
on Mill’s treatment of fallacies might be found in the conceptions of evidence 
developed by Bentham in the Rationale of Judicial Evidence (1827), edited by Mill 
(see Bentham 1827; Kubitz 1932: 51-2). 

Mill also might have claimed the authority of Francis Bacon for his approach to 
traditional logic. Although many philosophers believed that Bacon was the champion 
of the rejection of the syllogism and of Aristotelian logic generally in favour of a 
rigorous inductive approach, Mill saw in Bacon a forerunner of his own position 
that sought to combine induction with syllogistic logic (CW XI: 12-13). Bacon 
argued that, even though traditional logic ‘has done more to fix errors than to 
reveal truth’, there was no reason to attack traditional logic or even to challenge 
the ancient view of logic, as the new inductive system could easily exist side by 
side with the Aristotelian system (Bacon 1994: 38). ‘Let there be therefore, as a 
boon and blessing for each side, two sources of knowledge and two ways of 
organizing it; and likewise two tribes of thinkers and philosophers, two clans as it 
were, not in any way hostile or alien to each other, but linked in mutual support .. 
..’ (Bacon 1994: 40). Furthermore, Mill’s discussion of error, as we shall see, 
owed a good deal to Bacon’s analysis of ‘Idols’, and, particularly, to ‘Idols of the 
Theatre’ in the Novum Organum. 

Another factor that lay behind Mill’s study of fallacies was his commitment to 
a Socratic method in philosophy (see Urbinati 2002: 142-3). Mill had been attracted 
to Socrates, his hero and mentor, from his earliest days. ‘The Socratic method’, 
he wrote in the ‘Early Draft’ of the Autobiography, ‘. . . is unsurpassed as a 



J.S. Mill on Logical Fallacies  125 

discipline for abstract thought on the most difficult subjects. Nothing in modern 
life and education, in the smallest degree supplies its place’ (CW I: 24). After a 
brief account of the Socratic elenchus, he concluded that, even as a boy, the 
Socratic method ‘took such hold on me that it became part of my own mind; and 
I have ever felt myself, beyond any modern that I know of except my father and 
perhaps beyond even him, a pupil of Plato, and cast in the mould of his dialectics’ 
(CW I: 24). Mill wrote his translations and brief comments on nine Platonic dialogues 
for the Monthly Repository in 1834-5 (see CW XI: 37-238), just prior to his writing 
the essays on Bentham and Coleridge and the fifth book of the Logic. He also 
would have known that in the Sophistical Refutations Aristotle was reflecting and 
consolidating the earlier debates between Socrates and the Sophists and others, 
which appear in these dialogues. 

Mill was strongly impressed by Plato’s picture in the Gorgias and Republic ‘of 
the solitary and despised position of the philosopher in every existing society, and 
the universal impression against him, as at best an useless person, but more 
frequently an eminently wicked one’. . . (CW XI: 399). He adopted Grote’s arguments 
that while Plato might have despised the Sophists, because they were more concerned 
with appearances than with reality, and took money for their services, they were 
not the real enemy.The real corruptors of the young were not the Sophists, but 
society itself: 

. . . their families, their associates, all whom they see and converse with, the 
applauses and hootings of the public assembly, the sentences of the court of 
justice.  These are what pervert young men, by holding up to them a false 
standard of good and evil, and giving an entirely wrong direction to their 
desires.  As for the Sophists, they merely repeat the people’s own opinions. 
(CW XI:400) 

Mill not only adopted Grote’s view of the importance of the Sophists to public 
debate and the role of Socrates in this debate, but, in a review of Grote’s 
posthumously published study of Aristotle, he also acknowledged the importance 
of the Sophistical Refutations in the development of an important aspect of logic 
and in the pursuit of truth. Even though Aristotle’s work was a study of the art of 
arguing for victory rather than for truth, Mill saw nothing wrong with such public 
debates so long as they took place within established rules. Furthermore, he believed 
that Aristotle regarded such exercises in dialectical argument to be valuable in the 
pursuit of truth. Quoting from Grote, Mill gave three reasons for this belief (CW 
XI: 508). First, the debates constituted a kind of mental exercise that was valuable 
and stimulating.  Second, the debates put one in touch with the ordinary opinions 
held by society, and hence led the philosopher to attempt to understand and modify 
them. Third, such dialectical debate influenced science and philosophy, encouraging 
one to look at both sides of questions and to determine which answers were true 
or false. Mill believed that Aristotle’s method in dialectical argument ‘was greatly 
in advance not only of his own time, but of ours’: 



126     Frederick Rosen 

His general advice for exercise and practice in Dialectic is admirably adapted 
to the training of one’s own mind for the pursuit of truth.  “You ought to test 
every thesis by first assuming it to be true, then assuming it to be false, and 
following out the consequences on both sides.”  This was already the practice 
of the Eleatic dialecticians, as we see in the Parmenides. (CW XI: 508) 

Mill’s study of fallacies was based on the lonely philosophical mission of Socrates 
and the framework of Aristotle’s various categories of fallacies provided by 
generations of logicians.  He proposed a major revision of the Aristotelian categories 
without rejecting Aristotle’s achievement and without abandoning the perspective 
of the Socratic dialectician on which it was based. 

II 

Mill’s analysis of fallacies was intended to form part of what he called the 
‘Philosophy of Error’ (CW VIII: 737 [V i §3]5).  This phrase may seem strange to 
us, though it would not have been odd for Mill to use it. At the beginning of the 
book on fallacies, Mill quoted briefly from Malebranche and Hobbes, both of whom 
wrote on the theme of error (see CW VIII: 734; intro. quotation). Malebranche, 
for example, began his The Search after Truth as follows: 

Error is the cause of men’s misery; it is the sinister principle that has produced 
the evil in the world; it generates and maintains in our soul all the evils that 
afflict us, and we may hope for sound and genuine happiness only by seriously 
laboring to avoid it. (Malebranche 1997:1) 

For Malebranche, the main source of error was not one that lay outside the 
individual, but was closely connected with the misuse of freedom. This misuse of 
freedom consisted of a failure to regulate the will in its eagerness to embrace the 
truth. Such eagerness led one to embrace the appearance of truth and to go beyond 
the understanding to judge of matters and give consent to the truth of propositions 
about which one had no clear perception (Malebranche 1997: 16). 

According to Hobbes, we might distinguish between falsity and error. Falsity 
arose from ‘pronouncing rashly’, as when in seeing the sun in the sky and its 
reflection in the water we say that both are the sun and then conclude that there 
are two suns. Such a false conclusion was represented by Hobbes as a departure 
from the agreed conventions of names and not the commission of error. Error 
arose from our senses and thinking, as, for example, when we saw the sun’s 
reflection in the water, we imagined that it was actually there (Hobbes 1997: i.55- 
6). The deception here, as in Malebranche, proceeded from within the individual, 
even though Hobbes did not subscribe to the doctrine of freedom of the will. For 
Hobbes, error was a deception, to which we consented, that proceeded from 
reasoning itself. Such an error was further distinguished from absurdity (‘senseless 
speech’), as when one referred to a round triangle, etc., which was nonsense 
rather than error (though often commonly referred to as error) (see Hobbes 1997: 
iii.32). 



J.S. Mill on Logical Fallacies  127 

For both Malebranche and Hobbes, error could be combated by the individual 
through the exercise of freedom of thought and the use of reason.  However, not 
all philosophers accepted this position. Locke, for example, included a chapter ‘Of 
Wrong Assent, or Error’ (IV.XX) in the Essay, which seemed to diminish the 
scope of freedom and the occasions for error, leading him to the paradoxical 
conclusion that ‘There are not so many Men in Errours and wrong Opinions, as is 
commonly supposed’ (Locke 1975: 719). The conclusion was paradoxical not 
because Locke believed that we generally knew the truth, but despite our ignorance, 
we did not actually make erroneous judgments.  For most ordinary people the need 
to labour left them unable to consider the validity of the proofs of arguments to 
establish whether or not they were erroneous or true (Locke 1975: 706ff). For 
those who had leisure, Locke found them ‘confined to narrowness of Thought, 
and enslaved in that which should be the freest part of Man, their Understandings.’ 
Required religious observance, it seemed, supplied the opinions of the leisured 
few, and filled up the minds of those who had only a little leisure from the drudgeries 
of labour. Locke referred to this phenomenon as the propagation of ‘Truth without 
Knowledge’, ‘where Men are forced, at a venture, to be of the Religion of the 
Country; and must therefore swallow down Opinions, as silly People do Empiric’s 
Pills, without knowing what they are made of, or how they will work, and have 
nothing to do, but believe that they will do the Cure. . .’ (Locke 1975: 708-9). 
Furthermore, most people could not follow a syllogism or other process of reasoning 
and in this respect ‘there is a greater distance between some Men, and others. . . 
than between some Men and some Beasts’ (Locke 1975: 709). Others did not wish 
to cultivate the understanding due to weakness of will, the ‘hot pursuit of pleasure’, 
drudgery in their business and domestic lives, laziness, oscitancy, an aversion to 
books and study, and simply took on trust what they found convenient and ‘in 
fashion’. They also tended to care for their bodies but not for their souls (Locke 
1975: 710). 

Logic and hence fallacies, for Locke, thus played a small role in the lives of 
most people.  As Locke wrote: 

‘Tis enough for him to obey his Leaders, to have his Hand and his Tongue 
ready for the support of the common Cause, and thereby approve himself to 
those, who can give him Credit, Preferment, or Protection in that Society. 
Thus Men become Professors of, and Combatants for, those Opinions, they 
were never convinced of, nor Proselytes to; no, nor even had so much as 
floating in their Heads: and though one cannot say, there are fewer improbable 
or erroneous Opinions in the World than there are; yet this is certain, there 
are fewer, that actually assent to them, and mistake them for truths, than is 
imagined. (Locke 1975: 719) 

If Locke seemed to provide little scope for the work of the individual logician, 
and no estimation of that interior freedom apparently necessary to detect errors in 
reasoning, Mill certainly did. For Mill, however, not all errors were relevant to the 
study of fallacies. He would not include casual mistakes, such as the incorrect 



128     Frederick Rosen 

addition of a sum, the remedies for which would be simply paying greater attention 
and further practice (CW VIII: 736-7 [V i §2]). Nor did he include moral (as 
opposed to intellectual) sources of erroneous opinions, which were due to 
indifference to attaining truth on the one hand and bias on the other. Both of these 
were highly powerful and influential, but, in Mill’s opinion, they could not operate 
on their own. They predisposed one to erroneous opinions, but did not actually 
determine them. Indifference to the truth led to error, because it prevented the 
mind from collecting evidence and making careful inductions. Bias would dispose 
the mind towards accepting erroneous theories, but would not in itself lead to 
errors (CW VIII: 737-8 [V i §3]). 

This analysis of the human condition led Mill to an apparently gloomy conclusion 
regarding philosophical theories: 

The natural or acquired partialities of mankind are continually throwing up 
philosophical theories, the sole recommendation of which consists in the 
premises they afford for proving cherished doctrines, or justifying favourite 
feelings: and when any one of these theories has been so thoroughly 
discredited as no longer to serve the purpose, another is always ready to 
take its place. (CW VIII: 738 [V i §3]) 

This apparently endless succession of false and misleading philosophical theories 
is bad enough, but to make matters worse, those who attempted to use their 
intelligence to gather and assess evidence and make inferences from such evidence 
were often ‘stigmatised’ with names like ‘scepticism, immorality, coldness, hard- 
heartedness, and similar expressions’ (CW VIII: 738 [V i §3]). 

In spite of this view of the human condition Mill believed that whenever bias 
operated successfully, it was first necessary to undermine the understanding. Anyone 
sufficiently on guard regarding evidence and inference would be able to resist 
such subverting pressures of bias.  As he wrote: 

There are minds so strongly fortified on the intellectual side, that they could 
not blind themselves to the light of truth, however really desirous of doing 
so; they could not, with all the inclination in the world, pass off upon 
themselves bad arguments for good ones. (CW VIII: 739 [V i §3]) 

Mill thus proposed an opposing force of the incorruptible intellect that could 
withstand the presence and influence of bias. There were a few people who could 
resist the appeal of bias and both the illusions and enticements of fashionable 
philosophical theories. ‘If the sophistry of the intellect could be rendered impossible’, 
Mill wrote, ‘that of the feelings, having no instrument to work with, would be 
powerless’ (CW VIII: 739 [V i §3]). He did not propose a campaign to eliminate 
bias; that would have been impossible. Bias was derived from human nature itself. 
One might have concluded that the inevitability of the succession of one false but 
fashionable theory after another meant that we were condemned to live wholly in 
a world of false consciousness. But, as we have seen, Mill allowed full scope for 
the search for truth in all fields where the intellect might bring enlightenment. 



J.S. Mill on Logical Fallacies  129 

Logic itself would enable truth to be discovered and maintained across all 
subjects, but the study of fallacies had a special role. It not only supported the 
pursuit of truth, but it equally could be used to undermine this apparently endless 
succession of bogus philosophy based on bias and an indifference to truth. Even 
though new theories might be waiting in the wings to succeed those discredited by 
philosophical analysis, the philosophy of error itself might develop and become 
more widespread, thus limiting the effect of these fashionable theories. For Mill, 
then, the study of fallacies provided an important ally for philosophical analysis in 
preparing the mind and assisting in the development of mental habits that protected 
the individual and enabled one more easily to detect error. 

Although Mill’s contribution to the ‘Philosophy of Error’ is striking, it is not 
necessarily original.  His depiction of the succession of erroneous philosophical 
theories might well have been taken from Bacon’s ‘Idols of the Theatre’ in the 
Novum Organum (Bacon 1994: 55-6; see Urbach 1987: 105-6). Bacon’s depiction 
of the ‘various dogmas of philosophies’ as ‘so many stage plays creating fictitious 
and imaginary worlds’ must surely have been Mill’s source for his own striking 
vision of the endless succession of false philosophical theories. 

In addition to Bacon, one should also mention the Port-Royal Logic of Arnauld 
and Nicole, which Mill generally regarded favourably (see Arnauld and Nicole 
1996: 135, 189-203ff). Here one can find an account of fallacies that included, but 
also looked beyond, the Aristotelian list. The discussion of error sought to account 
for how one acquired false principles as much as, following Aristotle, how one 
reasoned incorrectly from these principles. Like Arnauld and Nicole, Mill was also 
concerned with how self-love, interest, or passion might lead one to adopt various 
principles, and he was also willing to take up the challenge of understanding such 
influences or thought from the perspective of fallacious reasoning. 

Mill’s discussion of the philosophy of error fed into his other writings on 
numerous levels. To take one example, it enables us to appreciate more fully his 
analysis of ‘Of the Liberty of Thought and Discussion’—the title and topic of 
Chapter 2 of On Liberty (CW XVIII: 228). This chapter, usually referred to in 
terms of freedom of expression or freedom of opinion, contains two elements 
with which freedom is concerned:  thought and discussion. It is clear from the 
opening paragraphs that Mill’s philosophy of error was lurking in the background. 
The pursuit of truth, which liberty would facilitate, was inconceivable for Mill 
without the presence and beneficial use of error. If liberty was denied, an erroneous 
opinion could not be exchanged for a true one; but if an original opinion was true, 
the denial of liberty meant that one could not benefit from ‘the clearer perception 
and livelier impression of truth, produced by its collision with error’ (CW XVIII: 
229 [L ii para.1]). The study of fallacies was one important way of encouraging 
the beneficial dialogue between truth and error, which enabled truth ultimately to 
replace error in inference, deduction, and in argument generally. 



130     Frederick Rosen 

Mill also repeated in On Liberty his belief that error could be corrected (CW 
XVIII: 231 [L ii para.7]), but there were new elements in his argument that were 
not present in the Logic. On the one hand, consistently with the Logic, he could 
write: 

It is a piece of idle sentimentality that truth, merely as truth, has any inherent 
power denied to error, of prevailing against the dungeon and the stake. Men 
are not more zealous for truth than they often are for error, and a sufficient 
application of legal or even social penalties will generally succeed in stopping 
the propagation of either (CW XVIII: 238-9 [L ii para.17]). 

Although Mill continued by qualifying his remarks, he provided here, as in the 
Logic, no assurance that truth would triumph over error. At one point he also 
alluded to the tiny minority concerned with truth in any society and the near 
impossibility of struggling against error among this elite: 

. . . for, on any matter not self-evident, there are ninety-nine persons totally 
incapable of judging of it, for one who is capable; and the capacity of the 
hundredth person is only comparative; for the majority of the eminent men of 
every past generation held many opinions now known to be erroneous, and 
did or approved numerous things which no one will now justify. (CW XVIII: 
231 [L ii para.7]) 

Not only do we see in this material an allusion to Bacon’s ‘Idols of the Theatre’ 
and Mill’s own use of it, but also to such great figures as Socrates and Jesus who, 
as individuals, died for truth (CW XVIII: 235-6 [L ii paras.12,13]). One might 
have expected Mill to endorse the position in the Logic regarding the struggle 
against the endless procession of false philosophical theories and to confirm his 
ringing statement in the introduction to the Logic that logic was concerned with 
the cultivation and development of one’s own thoughts: ‘If there were but one 
rational being in the universe, that being might be a perfect logician; and the science 
and art of logic would be the same for that one person as for the whole human 
race’ (CW VII: 6 [intro. §3]). 

Nevertheless, just following the passage quoted above concerning ninety-nine 
out of a hundred people, Mill posed a different question and gave to it a very 
different and apparently more optimistic answer: 

Why is it, then, that there is on the whole a preponderance among mankind 
of rational opinions and rational conduct? If there really is this preponderance 
—which there must be unless human affairs are, and have always been, in an 
almost desperate state—it is owing to a quality of the human mind, the 
source of everything respectable in man either as an intellectual or as a moral 
being, namely that his errors are corrigible. (CW XVIII: 231 [L ii para.7]) 

One might admit that Mill’s assertion of the preponderance of rational opinion 
and conduct was set forth somewhat tentatively, but still wonder how his belief 
that error was correctable (a belief at the heart of the philosophy of error and 
linked to the very idea of liberty of thought) could sustain the thesis that mankind 



J.S. Mill on Logical Fallacies  131 

was mainly rational in opinions and conduct. What happened to the moral causes 
of error, i.e., indifference to truth and bias that led so many to accept error? 

Mill’s sudden ascription of rationality to mankind reveals an important duality 
in On Liberty that fits together only tentatively or, we might even say, hypothetically. 
On the one hand, to overcome error, the great thinker must be free to follow his 
intellect wherever it led. The person who thought for himself or herself, and despite 
errors, did not give up, was superior to those who held only true opinions due to 
an indifference to the truth. On the other hand, Mill was not satisfied with the 
social consequences of what the great thinker might achieve, because this 
achievement might take place in a society where most people suffered mental 
enslavement. As Mill wrote: 

Not that it is solely, or chiefly, to form great thinkers, that freedom of thinking 
is required.  On the contrary, it is as much and even more indispensable, to 
enable average human beings to attain the mental stature which they are 
capable of.  There have been, and may again be, great individual thinkers, in 
a general atmosphere of mental slavery.  But there never has been, nor ever 
will be, in that atmosphere, an intellectually active people. (CW XVIII: 243 [L 
ii para.20]) 

Clearly, for Mill, in On Liberty an ‘intellectually active people’ was as important 
as that ‘one rational being in the universe’ of the Logic.  But in On Liberty Mill also 
attempted to bring them together. It is important to be clear about what Mill sought 
to bring together. O’Rourke has argued correctly against Himmelfarb, Skorupski, 
and others, who see different and conflicting principles of liberty in On Liberty, 
with freedom of expression being primarily other-regarding, while liberty generally 
is mainly concerned with self-regarding acts (O’Rourke 2001: 107ff). He has also 
emphasized not only that freedom of thought is primarily self-regarding, but also 
that it is the driving force behind freedom of expression generally (O’Rourke 
2001: 77ff, 96 and n). Thus, for O’Rourke, there is no conflict or inconsistency 
between the liberty of expression emphasized in Chapter 2 of On Liberty and the 
discussion of liberty generally in that work. 

My problem with On Liberty is of a different kind. The development of the 
philosophy of error in the Logic clearly adds to and supports O’Rourke’s thesis in 
showing what Mill meant by the very idea of liberty of thought. My problem arises 
from Mill’s clear acceptance of the role of the great individual thinker, like Socrates, 
in a world that despises or rejects such a person. How then did Mill shift the 
argument from the great thinker to the ‘average human being’?  He did so by again 
invoking the philosophy of error. All human beings have the capacity to correct 
error. Indeed, if freedom of thought, and this aspect of it, provides the dynamic 
for the whole of On Liberty, and if such liberty is the only bulwark against the 
tyranny of the majority, it is no wonder that Mill remained, as we have seen from 
the quotations from his Inaugural Address, the great advocate of the study of 
logic. This study, and particularly that of fallacies, is one important way for the 



132     Frederick Rosen 

logic of a Socrates to become that of an average human being. Even if only a few 
can aspire to greatness in philosophy, everyone is capable of searching out and 
correcting error. For Mill, therefore, as evident in the Inaugural Address, the study 
of logic is the bedrock of the search for truth generally in any society that values 
liberty. 

III 

If some of Mill’s contemporaries, like Whately and De Morgan, were content to 
expound, more or less, the original Aristotelian list of fallacies (see Whately 1975: 
131-203; De Morgan 1926: 276ff), Mill set out an entirely new classification (CW 
VIII: 743 [V ii §2]). This classification ultimately created five categories with 
many forms illustrated within each category.  Most of the Aristotelian list reappear 
in Mill’s discussion in various ways, but Mill’s aims and treatment of the fallacies 
were original and distinctive. A full survey of all of the fallacies might be somewhat 
tedious, so I shall confine myself to a general survey of his main categories and 
some estimation of his achievement. 

Mill reached his five categories by making a series of distinctions between 
various kinds of belief and argument. The first category was based on a distinction 
between beliefs held to be true simply and without evidence and those based on 
inferences from evidence. Fallacies, related to the former, were called fallacies of 
simple inspection or a priori fallacies which were held without direct evidence or 
experience, but created a presumption in favour of a proposition which allowed 
the holder to discard rules of induction (CW VIII: 741 [V ii §2]). The six examples 
given by Mill covered an enormous amount of ground, and Alexander Bain thought 
that the category might be included in a separate treatise on its own, and, additionally, 
need not be the special province of the logician (Bain 1870: ii.375). 

Let us look briefly at some of the fallacies Mill discussed.  The first consisted 
of mistaking subjective laws for objective ones. This category is somewhat 
reminiscent of Bentham’s dramatic account in An Introduction to the Principles of 
Morals and Legislation, much admired by Mill, of the way most previous 
philosophers had used subjective ideas as objective principles and foundations for 
their systems. Bentham mentioned, for example, such common ideas in philosophy 
and theology, as moral sense, common sense, understanding, rule of right, fitness 
of things, law of nature, law of reason, natural justice, natural equity, good order, 
truth, and the doctrine of election (Bentham 1996: 21ff, 26n-7n; see Rosen 2003: 
221ff). Mill had a similar object in mind at the level of logic in seeking to dismiss 
much of what he called ‘intuitionism’ in favour of his own empiricism. But he 
believed, and expressed this belief at a number of points in the Logic (e.g. CW VII: 
8-9, 14 [intro. §4]; VIII: 644 [IV i §2], 746-7 [V iii §1]), that logic could not and 
need not resolve metaphysical issues or determine fundamental philosophical 
positions. Nevertheless, the exposure of logical fallacies might undermine such 
positions and require those who took them to restate or abandon their arguments. 



J.S. Mill on Logical Fallacies  133 

Although Mill fully appreciated this way of thinking among intuitionist 
philosophers, he turned instead to illustrate the fallacy by considering several popular 
beliefs where subjective thoughts were wrongly given objective status. For example, 
he took the common remark, ‘talk of the devil and he will appear’, to illustrate 
how an idea was supposed to establish a reality (CW VIII: 748 [V iii §2]). The 
work of alchemists to make gold potable and exploit its miraculous properties, Mill 
believed, was based on the idea that because gold was considered something precious 
and to be marvelled at, it must have marvellous properties as a physical substance. 
The fallacy that wonderful events must have wonderful properties was illustrated 
by the example of showers of objects falling mysteriously in northern Italy being 
regarded as manna from heaven and eaten avidly by the local population, even 
though analysis revealed such manna to consist of the excrement of insects. So 
eagerly was the so-called manna consumed that it was difficult to obtain samples 
for analysis (CW VIII: 750 [V iii §3]). 

Mill then went on to consider how this fallacy infected scientific and philosophical 
thinking. The fallacy consisted in the belief that ‘our ideas of things must be true 
of the things themselves’ (CW VIII: 751 [V iii §3]). At the level of philosophy one 
example was the tendency to believe that things that we think of together must 
exist together, and a second example: that which is inconceivable must be false 
(CW VIII: 750 [V iii §3]). 

The first part of this fallacy (that A must accompany B because it is involved in 
the idea of B) was first exhibited, he believed, by Descartes, but was later taken up 
by Spinoza, Leibniz, and was widespread in modern German philosophy. Mill then 
wrote: 

I am indeed disposed to think that the fallacy now under consideration has 
been the cause of two-thirds of the bad philosophy, and especially of the bad 
metaphysics, which the human mind has never ceased to produce. Our general 
ideas contain nothing but what has been put into them, either by our passive 
experience, or by our active habits of thought; and the metaphysicians in all 
ages, who have attempted to construct the laws of the universe by reasoning 
from our supposed necessities of thought, have always proceeded, and only 
could proceed, by laboriously finding in their own minds what they themselves 
had formerly put there, and evolving from their ideas of things what they had 
first involved in those ideas. In this way all deeply-rooted opinions and 
feelings are enabled to create apparent demonstrations of their truth and 
reasonableness, as it were out of their own substance. (CW VIII: 752 [V iii 
§3]) 

This passage has been quoted at length to show that Mill regarded any philosophy 
that did not engage with a rigorous empiricism as being fundamentally flawed. 
Although a work on metaphysics might be necessary to dispatch that particular 
doctrine, the analysis of fallacies was intended to undermine particular aspects of 
it. 



134     Frederick Rosen 

Mill illustrated the second part of this fallacy, that things that cannot be conceived 
together cannot exist together, by using an example taken from Isaac Newton. 
Newton responded to Descartes’ attack on the theory of gravitation on the grounds 
that the sun could not act upon the earth as it was not on the earth, by supposing 
that there must exist an ether that filled up the void between sun and earth and 
enabled gravity to operate. Mill quoted Newton, himself, as writing: 

That gravity should be innate, inherent, and essential to matter, so that one 
body may act on another, at a distance, through a vacuum, without the 
mediation of anything else, by and through which their action and force may 
be conveyed from one to another, is to me so great an absurdity, that I believe 
no man, who in philosophical matters has a competent faculty of thinking, 
can ever fall into it. (CW VIII: 754 [V iii §3]) 

‘This passage’, Mill continued after the quotation from Newton, ‘should be 
hung up in the cabinet of every cultivator of science who is ever tempted to 
pronounce a fact impossible because it appears to him inconceivable’ (CW VIII: 
754 [V iii §3]). Mill then dealt with a string of a priori propositions that did not 
depend on evidence, e.g.: matter cannot think; space or extension is infinite; nothing 
can be made out of nothing. His point was that we do not know if these maxims 
are true or false, but that they are often presumed to be true merely because we 
think that they must be so. ‘A large proportion of all the errors ever committed in 
the investigation of the laws of nature’, he wrote, ‘have arisen from the assumption 
that the most familiar explanation or hypothesis must be the truest’ (CW VIII: 756 
[V iii §3]). 

Mill’s discussion of the remaining five fallacies in the category of simple 
inspection ranged widely from common beliefs to scientific and philosophical 
theories. He criticized such beliefs as that the lungs of a fox would relieve or cure 
asthma and that turmeric (which is yellow) was good for jaundice; he rejected the 
view that some substances bore the ‘signatures’ of the humours, e.g., red roses of 
the blood and flowers of saffron for the bile. There were also extensive criticisms 
of philosophical doctrines associated with ‘mysticism’ (the Hindu Scriptures, 
Platonism, and Hegelianism), and also with the Epicureans and Bacon (CW VIII: 
757ff. [V iii §4]). 

Bain was clearly impressed with Mill’s exposition.  He regarded Mill as the 
direct successor to Socrates and Bacon. ‘Socrates’, he wrote, ‘was the first person 
to urge strongly the natural corruption of the human intellect, and the need of a 
very severe remedial discipline, which, in the shape of personal cross-examination, 
he was wont to apply to his fellow Athenians’ (Bain 1870: ii.375). From Socrates, 
Bain moved directly to Bacon, whose discussion of the ‘Idols’ in the Novum Organum 
he thought bore a striking resemblance to Mill’s discussion of a priori fallacies. 



J.S. Mill on Logical Fallacies  135 

IV 

Mill’s second major distinction was between those fallacies that arose where evidence 
was distinctly conceived (categories 2-4) and those (category 5) where evidence 
was indistinctly conceived. He took the name of this last category, ‘fallacies of 
confusion’, from Bentham’s Book of Fallacies (CW VIII: 742 [V ii §2]; Bentham 
1824: 213-58).6 Mill stated that this class of fallacy occurred where ‘. . .the source 
of error is not so much a false estimate of the probative force of known evidence, 
as an indistinct, indefinite, and fluctuating conception of what the evidence is’ 
(CW VIII: 809 [V vii §1]). Bain considered this category as ‘boundless and 
indefinable’. ‘No man’, he wrote, ‘can foreshadow the intricacies, the incoherences, 
the perplexities, the entanglements, possible to the human understanding’ (Bain 
1870: ii.376).  What was considered under this category by Mill was ‘extra-logical’ 
and the fallacies could not be absorbed into the main principles of induction and 
deduction. For Bain, ‘a considerable stretch of the logician’s province is implied in 
the taking up of this class of errors’ (Bain 1870: ii.375). As with the first category 
Bain also proposed to consider this one separately, as not wholly part of inductive 
and deductive logic (Bain 1870: ii.376). 

The three classes of examples—ambiguous terms, petitio principii, and 
ignoratio elenchi—were familiar parts of Aristotle’s schema of fallacies in thought 
(extra dictionem). But Mill’s grouping them in this larger category of fallacies of 
confusion gave them an additional importance, not all of which was captured by 
Bain’s suggestion that they could be separated from the ‘boundless and indefinable’ 
mass of fallacies in this category, because they occurred most frequently. 
Nevertheless, by examining these categories, we shall be able to assess Mill’s own 
objects with this material. 

The first of the fallacies was that of ambiguous terms (CW VIII: 809ff. [V vii 
§1]). This fallacy occurred in reasoning either when there was an ambiguity in the 
middle term or when terms were used in different ways in premises and conclusions. 
To illustrate the range of fallacies under this heading, Mill quoted at length from 
Whately who provided numerous examples of the fallacy. One was the assumption 
that words coming from the same root necessarily had the same meaning. Hence, 
the argument—projectors (entrepreneurs) are unfit to be trusted; this man has 
formed a project; therefore, he is unfit to be trusted—was fallacious because not 
everyone who developed a project was a projector. Another fallacy displaying 
ambiguity in the premise and conclusion might be: To be acquainted with the guilty 
is a presumption of guilt; this man is so acquainted; therefore, we may presume 
that he is guilty. Here, there was a difference in the meaning of ‘presumption’ and 
‘presume’. 

Mill devoted a good deal of attention to numerous confusions in philosophy. 
One was connected with the term ‘theory’, whose meanings ranged from ‘a 
completed philosophical induction from experience’, which was put into a form 



136     Frederick Rosen 

related to practice so that theory was the explanation of practice. In another and 
‘more vulgar’ sense, theory ‘means any mere fiction of the imagination’, which 
attempted to explain how a thing might have been produced rather than explaining 
how it was produced. When an attempt was made to discredit philosophy, theory 
and theorists in this latter sense were usually the objects of discussion (CW VIII: 
812; [V vii §1]). 

The second fallacy was that of petitio principii, begging the question, including 
reasoning in a circle (CW VIII: 819ff. [V vii §2]). Mill quoted Whately’s definition 
of the fallacy—‘in which the premise either appears manifestly to be the same as 
the conclusion, or is actually proved from the conclusion, or is such as would 
naturally and properly be so proved’ (CW VIII: 820 [V vii §2]). One example, 
which Mill took from scholasticism, was that the mind always thinks, because it is 
the essence of the mind to think (CW VIII: 822-3 [V vii §2]). The conclusion here 
was the same as the premise and was fallacious also because of the ambiguity in 
the term ‘essence’. 

Of the examples where one assumed what was to be proved, Mill drew directly 
on Bentham’s conception of ‘question-begging appellatives’ in the Book of Fallacies. 
Bentham argued that although petitio principii was well known to Aristotle, Aristotle 
had not appreciated how such a fallacy could be committed with the use of a 
single term (Bentham 1824: 213; CW VIII: 823 [V vii §2]). Mill (and Bentham) 
used the example of ‘innovation’ which, when applied to a practice, meant not 
only novel, but also bad (as opposed to ‘improvement’ which also meant novel but 
good) (Bentham 1824: 218). A defender of a new practice could not object to its 
being called an innovation, but in the process was forced by the use of the term to 
accept that it was a bad practice and not worthy of defence. Hence, the use of a 
single word was made to overrule all of the facts and force a major concession 
from those who defended a practice.7 

A final example connected with petitio principii was the justification of the 
doctrine of the social compact or contract in Hobbes and Rousseau, which Mill 
believed involved reasoning in a circle (CW VIII: 827 [V vii §2]). The social compact 
was supposedly based on a promise made by one’s ancestors. Hence, to the question, 
why obey the government formed via the social compact, the answer was given 
that we should do so because of our promise to obey. But Mill argued that the 
promise to obey was only based on and was enforceable by the social compact 
itself. Thus, the argument was circular: why obey the social compact? Because 
we must keep our promise. Why keep our promise?  Because we have agreed to 
the compact (see also Hume 1985: 480ff). Mill pointed out that the argument 
might have escaped such circularity by the additional argument that the compact 
was enforceable because of its utility, i.e., it led to happiness and a better life. 

The third fallacy was ignoratio elenchi, which consisted of reaching the wrong 
conclusions from the premises (CW VIII: 827-30 [V vii §3]). This differed from 
the fallacy of ambiguity, where the meaning of the premises was misconceived, or 



J.S. Mill on Logical Fallacies  137 

from petitio principii, where the conclusion was mistaken for what was to be 
proved. Mill again turned to lengthy quotations from Whately to illustrate the fallacy, 
which used numerous but irrelevant means to defeat an opponent and thereby 
establish one’s own thesis. In his own examples, Mill was more concerned with 
arguments in philosophy, which he took from Malthus, Berkeley, and others. For 
example, the argument, that Berkeley’s theory of the non-existence of matter could 
be defeated by knocking one’s head against a post, ignored or mistook the fact that 
Berkeley did not deny the experience of the senses, but that he was sceptical of the 
underlying cause of the appearances suggested by the senses. 

The fallacies of confusion thus ranged widely from public debate to philosophical 
argument, and from a number of propositions to the deft employment of a single 
term. In spite of Bain’s remark that the number of fallacies of confusion was 
endless, he believed that the study of logic could provide some discipline to 
understand and resist them, and the exposure of some of the most obvious 
categories might have ‘a slightly fortifying influence’ (Bain 1870: ii.386). 

V 
Mill’s third distinction among fallacies occurred where evidence was distinctly 
conceived, as opposed to the category of fallacies of confusion where the evidence 
was indistinctly conceived. Within this heading he distinguished between inductive 
and deductive fallacies. For the former (inductive fallacies) he included two 
categories:  fallacies of observation (category 2) and fallacies of generalization 
(category 3). 

Under the heading of fallacies of observation Mill included those arising from 
non-observation and mis-observation. For the former he distinguished between 
non-observation arising from overlooking instances and non-observation arising 
from overlooking circumstances (CW VIII: 773 [V iv §1]). An example of a fallacy 
arising from overlooking instances was the use of almanacs for predicting the 
weather. Even though there were many erroneous predictions each year, there 
were a few correct ones that sustained the faith in almanacs. The false ones were 
ignored and people continued to stick to the false prophet through the elimination 
of chance (CW VIII: 774 [V iv §3]). Mill emphasized that the greatest cause of this 
fallacy was ‘preconceived opinion’, which led people not to see pertinent and 
relevant facts.  He surveyed a variety of common opinions and scientific theories 
to demonstrate the widespread use of this fallacy. He pointed to its employment in 
religious disputes, where texts, which could not be reconciled with one’s views, 
were simply overlooked, and among historians where various events were seen 
through preconceived opinions to favour Protestants or Roman Catholics, Royalists 
or Republicans, etc. (see CW VIII: 776-8 [V iv §3]). 

The non-observation of circumstances was also shown by Mill to be the source 
of much fallacious thinking. He showed how the false belief in the phlogistic 
theory of combustion persisted, because no one looked carefully at the gaseous 



138     Frederick Rosen 

products of combustion. Eventually, chemists were led to the correct explanation 
that, instead of a substance (phlogiston) being separated and released through 
combustion, a substance was actually being absorbed. Another example was the 
widespread use of powders and ointments to effect often miraculous cures when 
other factors were actually responsible for the cure. Mill mentioned the example of 
John Wesley who believed he had been cured of a malady by the application of a 
brown paper plaster of egg and brimstone, and omitted to reckon the importance 
of four months’ rest, with country air, repose, ass’s milk, and exercise, also 
prescribed by his doctor. Another example was the widespread use of mercury in 
America to cure yellow fever, when those to whom it was applied were already 
cured of the disease (CW VIII: 778-80 [V iv §4]). 

Fallacies based on mal-observation (as opposed to non-observation) arose not 
from something not seen or ignored, but from something seen wrongly (CW VIII: 
782-4 [V iv §5]). It occurred primarily where one mistook perception for inference. 
One important example was the opposition to the Copernican system which was 
based on the common perception of the sun rising and setting and the stars revolving 
in the heavens. Mill linked to this fallacy the tendency to mistake inferences for 
intuitions. A visionary might think that God had spoken directly to him or her, and 
that this was known by intuition. What was at stake, however, was most likely a 
number of inferences based on mal-observation.  In making this assertion Mill was 
careful to add: ‘A caution, therefore, against this class of errors, is not only needful 
but indispensable; though to determine whether, on any of the great questions of 
metaphysics, such errors are actually committed, belongs not to this place, but as 
I have so often said, to a different science’ (CW VIII: 784 [V iv §5]). If Mill was 
reluctant to consider here the great questions of metaphysics and theology, 
concerned with the existence of God or other supreme principle, he clearly cast 
doubt on the logical standing of those mystics who claimed direct contact with 
God on the basis of legitimate inferences from experience. 

Fallacies of generalization constituted the second kind of inductive fallacy and 
the third category in Mill’s schema. He stated that these are ‘the most extensive of 
all’ fallacies, but if the earlier books of the Logic were followed, the fallacies 
should not occur and in this category there would only be mistakes and blunders 
(CW VIII: 785 [V v §1]). He then depicted a number of common examples of 
types of fallacies that have arisen from a failure to understand and apply the rules 
of induction. 

He began by asserting that certain kinds of generalization must be groundless, 
because ‘experience cannot afford the necessary conditions for establishing them 
by a correct induction’ (CW VIII: 785 [V v §2]). Inferences based on conditions 
on earth or in the solar system could not be assumed to extend everywhere in the 
universe. Similarly, despite some exceptions, propositions that asserted a universal 
impossibility could not be established by induction. 



J.S. Mill on Logical Fallacies  139 

Among particular fallacies he listed that of mistaking empirical for causal laws 
(CW VIII: 788-91 [V v §4]). For example, two widely-held beliefs in Mill’s day 
were: as blacks have never been as civilized as whites sometimes were, they shall 
never be so; and as women were supposed to be inferior to men in intellect, they 
will necessarily be inferior. These beliefs contained several errors, such as the 
absence of a comparison of instances or the ascertainment of material facts and 
the recognition that such instances could not be the basis of causal laws. Another 
example of this fallacy was that made by ‘common-sense’ people of a ‘practical’ 
disposition who believed that human beings would continue to act as they did at 
present. These ‘common-sense’ people wrongly denied the doctrine of the 
progressiveness of the human species which, despite numerous errors associated 
with this doctrine, did not assume wrongly that certain behaviour could not occur 
because it had not previously been exhibited. Mill believed that the doctrine of the 
progressiveness of the species was an important constituent of social science and 
though not a cause of anything, it stood as ‘a summary expression for the general 
result of all the causes’ (CW VIII: 791 [V v §4]). 

Another fallacy consisted of the confusion of sequence with consequence (post 
hoc, ergo propter hoc—‘after this and therefore owing to this’) (CW VIII: 792-4 
[V v §5]).  As examples Mill listed the following:  (1) England owed her industrial 
pre-eminence to restrictions on commerce; (2) the national debt was a cause of 
national prosperity; (3) England owed her prosperity to the excellence of the Church, 
Houses of Parliament, legal procedure, etc. In these examples historical sequences 
were fallaciously transformed into causal sequences. The fallacy was based on 
bad generalization a posteriori but could exist a priori with the latter attempting to 
explain a complicated phenomenon by a simpler theory than its nature admitted. 
For example, the belief that all disease depended on a hostile acid or alkali in the 
chemical composition of the body committed this fallacy. In politics there were 
doctrines that ascribed absolute goodness to particular forms of government, social 
arrangements, and even to particular modes of education without any reference to 
stages of civilization and the character of the societies for which they were intended. 
This became an important theme later in the Logic concerned with the social 
sciences, but here he discussed aspects of the issues in terms of fallacies. 

The remainder of this section was concerned with fallacies associated with 
analogies and metaphors (CW VIII: 794-801 [V v §§6-7]). These fallacies did not 
deal with attempts to establish inductions, but operated where a full proof was not 
a possibility. Mill depicted an argument from analogy as ‘an inference that what is 
true in a certain case, is true in a case known to be somewhat similar, but not 
known to be exactly parallel, that is, to be similar in all material circumstances’ 
(CW VIII: 794 [V v §6]). All that could be expected from analogies was a small 
increase in probability, and there were ample opportunities for fallacies. 



140     Frederick Rosen 

Metaphors tended to ‘assume the proposition which they are brought to prove. 
. . . For an apt metaphor, though it cannot prove, often suggests the proof’ (CW 
VIII: 800 [V v §7]). After considering a number of examples, Mill concluded: 

A metaphor, then is not to be considered as an argument, but an assertion 
that an argument exists; that a parity subsists between the case from which 
the metaphor is drawn and that to which it is applied. This parity may exist 
though the two cases be apparently very remote from one another . . . .  (CW 
VIII: 801 [V v §7]) 

The final category (category 4) contained the fallacies of ratiocination and was 
derived from the distinction between inductive and deductive fallacies (CW VIII: 
803-8 [V vi §§1-4]). If inductive fallacies led to fallacies of observation and 
generalization, deductive fallacies led to fallacies of ratiocination. Mill granted that 
in the ordinary textbooks of logic fallacies were mainly confined to this category. 
He, however, devoted only five pages to the topic, referred the reader mainly to 
discussions in Whately and DeMorgan, and claimed that the rules of the syllogism 
provided almost complete protection against many of the traditional fallacies (CW 
VIII: 803 [V vi §1]).8 Thus, in a curious way Mill managed to agree with those 
who rejected the separate study of traditional fallacies on the grounds that one 
needed to study the rules of logic rather than ways of breaking the rules, while at 
the same time he developed an extensive and novel treatise on fallacies. 

VI 
In the Autobiography Mill characteristically referred to the Logic as supplying the 
‘opposite doctrine’ to the German or a priori school of human knowledge. His 
doctrine ‘derives all knowledge from experience, and all moral and intellectual 
qualities principally from the direction given to the associations’. The a priori 
doctrine, ‘that truths external to the mind may be known by intuition or 
consciousness, independently of observation and experience’, was considered by 
Mill ‘as the great intellectual support of false doctrines and bad institutions’ (CW I: 
233).  In the Logic Mill particularly sought to clarify the foundations of mathematics 
and physical science in order to dislodge the a priori school from its supposed 
foundations. Given his belief that intuitionist philosophy had deleterious effects in 
‘morals, politics, and religion’ (CW I: 233), there may be some justification for 
Anschutz’s remark to the effect that Mill was fighting a political battle through his 
logical and metaphysical writings (see Anschutz 1953: 61-2; cf. Scarre 1989: 13- 
14). But, as we have seen, there was a second theme in the Logic, amply illustrated 
in the fifth book, which suggested that logic served truth rather than one doctrine 
as opposed to another.  Indeed, the material on fallacies established how, between 
opposing doctrines, the truth might be ascertained, presented, and defended.9 

Mill’s subtle and comprehensive study of logical fallacies took the subject away 
from the realm of deceitful tricks and a source of amusement for students in an 
otherwise dull subject to oppose false thinking in philosophy, science, morality, 



J.S. Mill on Logical Fallacies  141 

politics, and many aspects of ordinary life. But such false thinking might also be 
restated in more valid forms, and, as Mill argued against Whately, the detection of 
fallacies should be but one aspect of the use of fallacies in logic. The other aspect 
was an attempt to restate the argument without fallacies, to clarify the argument, 
and further assess its validity. If, in the essays on Bentham and Coleridge, Mill 
attempted to bring together the opposing doctrines of the two supposedly leading 
thinkers of the age, in the Logic he provided the philosophical reader with some of 
the equipment necessary to restate, clarify, and assess such opposed ideas. 

One need not accept Passmore’s view that ‘there could be no lasting compromise 
between Mill and the Coleridge school; Coleridge and his followers’, he wrote, 
‘were beyond the pale as “intuitionists”—Mill’s favourite term of all-encompassing 
abuse. . .’ (Passmore 1966: 13). Although Mill consistently rejected intuitionism in 
philosophy, he found much to admire in Coleridge’s ideas: the emphasis on the 
history of countries and institutions as keys to understanding their character, the 
admiration of Bacon, placing the pursuit of truth above a blind adherence to religious 
or philosophical creeds, and a willingness to explore opposing poles in thought and 
politics to determine the truth. This use of opposing points of view enabled Mill to 
advance what he called in a letter to the Coleridgean, John Sterling, ‘a catholic 
spirit in philosophy’ which he felt under ‘a special obligation to preach’ (CW XIII: 
411). Although one might be somewhat puzzled as to how an emphasis on polarities 
could generate catholicity or universality in philosophy, Mill probably meant simply 
that by appreciating opposing sides of issues, one could approach more closely to 
a universal perspective than if one saw the world simply from one perspective. To 
take a practical example, Mill dealt with the opposing poles in politics of 
cosmopolitanism on the one hand and nationalism on the other by arguing for a 
principle of cohesion within society, which did not embrace a hatred of foreigners, 
an indifference to humanity, or the preference of the national interest over the 
human race in general. Mill thus attempted to draw from the polarities a new 
perspective—what has been called ‘universalist cosmopolitan patriotism’, which 
was uniquely his and arguably closer to the truth than either cosmopolitanism or 
nationalism (see Varouxakis 2002: 126). 

Mill’s account of fallacies and, indeed, the study of logical fallacies generally 
are not now in fashion.  As a faithful follower, Bain, with some exceptions, treated 
Mill’s approach to fallacies in his own study of logic with great respect (see Bain 
1870: i. pp. iv-v, ii. 374ff; 1904:  299-300). Nevertheless, many others who followed 
Mill were not as dutiful. In spite of their widespread critique of Mill’s logic, Idealist 
philosophers wholly ignored the material on fallacies. For F.H. Bradley, Bosanquet, 
and Green, the truth seemed to dwell in the realm of metaphysics, for which the 
study of fallacies in Mill’s version (and apparently in any other version) was not 
appropriate (see Bradley 2002, Bosanquet 1895, Green 1886: ii). When H.W.B. 
Joseph wrote his Introduction to Logic early in the twentieth century, he was not 
convinced that the study of fallacies should be part of logic. He argued that logic 



142     Frederick Rosen 

was like mathematics and that there were not two ways of thinking about 
mathematics. If logic was an empirical science, like psychology, one might argue 
that the study of fallacies was like studying mental illness as a distinct phenomenon 
within a larger science of psychology (Joseph 1916: 566-7). Nevertheless, Joseph 
decided to include an ‘Appendix’ on fallacies and to repeat the traditional thirteen 
fallacies originally set forth by Aristotle. He justified this move on the grounds that 
it had tradition in its favour, and would enable students to understand the language 
of fallacies which remained in common use. ‘And if we consider not the 
enumeration of types of fallacy, but their classification’, he wrote, ‘ it will appear, 
I think, that there is no such merit in any alternative scheme as justifies us in 
sacrificing the advantage of keeping to the standard and traditional scheme of 
Aristotle.’ Mill’s scheme was consigned to a brief footnote, where a question was 
raised as to whether or not the a priori fallacies were actually fallacies at all (Joseph 
1916: 574 and n). 

The main opposition to Mill’s approach to fallacies, however, came from a 
different source, which was more indifferent than hostile. This was the development 
of symbolic logic, and, particularly, the application of algebra to logic. In a recent 
edition of George Boole’s The Mathematical Analysis of Logic, originally published 
in 1847, just four years after Mill’s Logic, the editor has written: 

Boole’s work freed logicians, once and for all, from the endless repetition of 
a few principles, mostly derived from Aristotle, and, by example, challenged 
them to develop their subject in new directions. Many gifted thinkers, among 
them John Venn, William Stanley Jevons, Charles Sanders Peirce, Ernst 
Schröder, Alfred North Whitehead, and Bertrand Russell, rose to the 
challenge, and the vast and autonomous subject of modern logic gradually 
emerged from their combined efforts. It is a glorious success story. (Boole 
1998: xi) 

Mill is excluded from this illustrious list of logicians. And few recent textbooks 
on logic either mention Mill or discuss fallacies generally (see, for example, Tomassi 
1999). It is fair to say that modern symbolic logic has moved in new directions. 
Boole, himself, knew Mill’s logic and quoted from him at the beginning of his own 
work to the effect that where possible the mechanical side of logic might employ 
a technical notation and where logic did not involve mechanical reasoning, such a 
technical language should not be employed. Boole accepted this view and wrote 
that ‘to supersede the employment of common reason, or to subject it to the rigour 
of technical forms, would be the last desire of one who knows the value of that 
intellectual toil and warfare which imparts to the mind an athletic vigour, and 
teaches it to contend with difficulties and to rely upon itself in emergencies’ (Boole 
1998: 2). 

Mill, however, did not appear to accept Boole’s comment. When he was asked 
to respond to the work of Jevons, he replied in a letter that, like Boole and 
DeMorgan, ‘he [Jevons] seems to me to have a mania for encumbering questions 



J.S. Mill on Logical Fallacies  143 

with useless complications, and with a notation implying the existence of greater 
precision in the data than the questions admit of’.  Mill called it ‘a vice’—a failure, 
one supposes, of intellectual virtue. He depicted this vice as ‘one preeminently at 
variance with the wants of the time, which demand that scientific deductions 
should be made as simple and as easily intelligible as they can be made without 
ceasing to be scientific’ (CW XVII: 1862-3). In a note added in 1856 to the Logic, 
commenting on the work of De Morgan (CW VII: 171n-3n [II ii §1 note]]), Mill 
also criticized the use of technical forms in logic as ruling out the study of fallacies, 
which were concerned with ‘the incautious use of the common forms of language’. 
The consequence was to leave ‘the enemy . . . in possession of the only ground on 
which he [the logician] can be formidable’ (CW VII: 172n). 

Mill’s brief comments raise as many questions as they answer (see Scarre 
1989: 1ff). It is arguable that the development of symbolic logic, particularly using 
algebra and grounded in mathematics generally, was more closely linked to modern 
science than Mill’s own approach. From another perspective, however, modern 
symbolic logic may be regarded as more conventional and traditional than the logic 
developed by Mill. Arguably, it replaced Mill’s logic of truth with the older view of 
logic as an art aimed at consistency in reasoning. Furthermore, Boole’s somewhat 
stoical comments about logic training the mind, though not foreign to Mill, were 
more appropriate to the teaching of traditional Aristotelian logic. In place of ‘athletic 
vigour’, Mill (in his Inaugural Address) emphasised ‘making your mind clear and 
keeping you from stumbling in the dark over the most outrageous fallacies’. His 
metaphors evoke Socratic and Platonic themes concerned with illumination and 
truth. His object was never to develop a gymnasium of the mind but to suggest 
that logic was the key to truth in science, morality, politics, and religion, and 
without logic, we were condemned to dwell in error forever. So strongly was he 
committed to this Socratic doctrine that he extended it, as we have seen, to his 
defence of liberty generally in society. He frequently invoked the importance of 
public debate and the study of fallacies as the best way to keep truth alive and 
opposed to the parade of false philosophical theories. For Mill, the abandonment of 
the study of fallacies would not only impoverish the mind but also would diminish 
the concern with truth.  In politics one is left with competing ideologies, but with 
little ambition to transcend them. Mill’s account of fallacies restated such Socratic 
ambitions in a modern idiom.10 

Notes 

1 See the Autobiography (CW I: 287), where he wrote concerning the Inaugural Address:  ‘. . . I 
gave expression to many thoughts and opinions which had been accumulating in me through life 
. . .’. 
2 See Jackson 1941; Anschutz 1953; Ryan 1987; Skorupski 1989; Scarre 1989.  For a discussion 
of Mill on fallacies, see Hansen 1997: 125-43. 



144     Frederick Rosen 

3 See Sanderson 1618; Aldrich 1691.  The Aristotelian and Scholastic fallacies are: 
I. Fallacies in language (in dictione): 

a) Homonymy (Equivocation):  double meaning in a single term. 

b) Amphiboly:  double meaning in a combination of words or sentences. 

c) Conjunction:  e.g. because 3 and 2 are 5, it does not follow that 5 is two 
           numbers 

d) Disjunction:  e.g. 3 and 2 are 5, but one cannot then say that 5 is both an odd 
             and even number. 

e)  Accentuation:  changing the meaning of a sentence by changing the accent on a 
              word, e.g. incense and incense. 

f) Figura Dictionis:  similar inflections in unlike things, e.g. an African has white 
             teeth and black skin; therefore, he is black and white. 

II. Fallacies in thought (extra dictionem): 

g) Fallacia accidentis  or a dicto simpliciter ad dictum secundum quid:  assuming 
                that subject and predicate have their attributes in common; taking a predicate 
             as coextensive with a subject when it is not. 

h) Fallacia a dicto secundem quid ad dictum simpliciter:  confusion of an absolute 
     statement with a statement limited in manner, place, time, or relation, e.g. 

          what you bought yesterday, you eat today; you bought raw meat yesterday; 
           you eat raw meat today. 

i) Ignoratio elenchi:  proceeding to argue a different position from that under 
    consideration. 
j) Fallacia consequentis, non sequitur:  numerous cases of non-sequitur, e.g. rain wets 
    the ground; therefore, wet ground implies that it has rained. 
k) Petitio principii:  arguing in a circle and begging the question. 
l) Non causa pro causa (also post hoc, ergo propter hoc):   an inductive fallacy wrongly 
    attributing to causation what merely occurs in sequence. 
m) Fallacia plurium interrogationum:  asking several questions at once in order to 
     confuse. 
See Bain 1870: i.275-8; Grote 1880: 385ff; Aristotle 1955; Hamblin 1970: 52ff. 

4 Coleridge concluded his ‘General Introduction, or Preliminary Treatise on Method’ which 
introduced the Encyclopaedia Metropolitana 1845: i.43, as follows:  ‘Without advocating the 
exploded doctrine of perfectibility, we cannot but regard all that is human in human nature, and all 
that in nature is above herself, as together working forward that far deeper and more permanent 
revolution in the moral world, of which the recent changes in the political world may be regarded 
as the pioneering whirlwind and storm.  But woe to that revolution which is not guided by the 
historic sense; by the pure and unsophisticated knowledge of the past: and to convey this 
methodically, so as to aid the progress of the future, has been already announced as the distinguishing 
claim of the Encyclopaedia Metropolitana.’ 
5  The Editors have inserted, following references to the Collected Works of Mill, edition-neutral 
references, as follows: for the Logic, e.g., “V i §3” means “Book V, Chaper 1, Section 3” and for 
On Liberty, e.g., “L ii para. 7” means “On Liberty, Chapter 2, Paragraph 7.” 
6 The name of this class of fallacy had actually been devised by Étienne Dumont for the earlier 
French edition of this work.  See Bentham 1816: ii.132ff; Bentham 1824: 10. 
7 Bentham’s theme here reflected his earlier distinction between eulogistic, dyslogistic, and neutral 
terminology in A Table of the Springs of Action (1817), where he argued that the estimation of 
objects, practices, and theories depended in part on the choice and use of terminology.  See 
Bentham 1983: 1-115, esp. 95ff. 



J.S. Mill on Logical Fallacies  145 

8 Nevertheless, Mill did consider individually a small number of fallacies of ratiocination.  The 
first was called ‘fallacies in the conversion and aequipollency of propositions’.  A second was the 
erroneous conversion of a hypothetical proposition.  A third was concerned with changing the 
premises from one argument to another, and was called in Scholastic philosophy ‘a dicto secundum 
quid ad dictum simpliciter’. 
9 For an example of how fallacies might be applied successfully to metaphysics without challenging 
any ultimate foundations of an opposing school, see CW IX: 424-7. 
10 On Socrates, see further: Urbinati 2002; Irwin 1998; Riley 1996; Demetriou 1996; Demetriou 
1999; Rosen 2004. 

References 

Aldrich, H. (1691). Artis Logicae Compendium. Oxford: Oxford University Press. 

Anschutz, R.P. (1953). The Philosophy of J. S. Mill. Oxford: Oxford University Press. 

Aristotle (1955).  On Sophistical Refutations, On Coming-To-Be and Passing-Away, On 
the Cosmos.  E.S. Forster and D.J. Furley (Trans.).  Cambridge, MA and London: The 
Loeb Classical Library. 

Arnauld, A. and Nicole, P. (1996). Logic or the Art of Thinking. J. Vance Buroker, (Ed.). 
Cambridge: Cambridge Texts in the History of Philosophy. 

Bacon, F. (1994).  Novum Organum. P. Urbach and J. Gibson (Trans.). Chicago and La 
Salle: Open Court 

Bain, A. (1870). Logic. 2 vols. London: Longmans & Co. 

Bain, A. (1904). Autobiography. London: Longmans, Green & Co. 

Bentham, G. (1827). Outline of a New System of Logic with A Critical Examination of Dr. 
Whately’s “Elements of Logic”. London: Hunt and Clarke. 

Bentham, J. (1816). Tactique des assemblées législatives, suivie d’un Traité des sophismes 
politiques. 2 vols. É. Dumont (Ed.). Geneva and Paris: J.J. Paschoud. 

Bentham, J. (1824). The Book of Fallacies: from Unfinished Papers of Jeremy Bentham. 
By a Friend [P. Bingham]. London: John and H.L. Hunt. 

Bentham, J. (1827). Rationale of Judicial Evidence, specially applied to English 
practice. 5 vols.  J.S. Mill (Ed.). London: Hunt and Clarke. 

Bentham, J. (1983). Deontology together with A Table of the Springs of Action and 
Article on Utilitarianism. A. Goldworth (Ed.). Oxford: Clarendon Press. 

Bentham, J. (1996).  An Introduction to the Principles of Morals and Legislation. J.H. 
Burns and H.L.A. Hart (Eds.), with a New Introduction by F. Rosen. Oxford: 
Clarendon Press. 

Boole, G. (1998).  The Mathematical Analysis of Logic. J. Slater (Ed.). Bristol: Thoemmes 
Press. 

Bosanquet, B. (1895). The Essentials of Logic. London: Macmillan & Co. 

Bradley, F.H. (2002). The Principles of Logic. Uxbridge: Cambridge Scholars. 

Demetriou, K. (1996). ‘Grote on Socrates: An Unpublished Essay of the 1820s in its 
Context’, Dialogos 3: 36-50 

Demetriou, K. (1999). George Grote on Plato and Athenian Democracy: A Study in 
Classical Reception. Frankfurt: Peter Lang. 



146     Frederick Rosen 

De Morgan, A. (1926). Formal Logic (1847). A.E. Taylor (Ed.). London: The Open Court 
Company. 

Encyclopaedia Metropolitana; or, Universal Dictionary of Knowledge, On an Original 
Plan (1845). 29 vols. E. Smedley, Hugh J. Rose, and Henry J. Rose (Eds.) London: B. 
Fellowes et al. 

Green, T.H. (1886). The Works of Thomas Hill Green. 3 vols. R L. Nettleship (Ed.). 
London: Longmans, Green. 

Grote, G. (1880). Aristotle, 2nd ed. A. Bain and G. Croom Robertson (Eds.). London: John 
Murray. 

Hamblin, C.L. (1970). Fallacies. London: Methuen. 

Hansen, H.V. (1997).  ‘Mill on inference and fallacies’ in D. Walton and A. Brinton (Eds.), 
Historical Foundations of Informal Logic. Aldershot: Ashgate. 

Hobbes, T. (1997). The Collected Works of Thomas Hobbes. 12 vols.W. Molesworth 
(Ed.) with a New Introduction by G.A.J. Rogers. London: Routledge/Thoemmes. 

Hume, D. (1985). Essays, Moral, Political and Literary. E.F. Miller (Ed.). Indianapolis: 
Liberty Press. 

Irwin, T. (1998). ‘Mill and the Classical World’ in The Cambridge Companion to Mill, J. 
Skorupski (Ed.). Cambridge: Cambridge University Press. 

Jackson, R. (1941). An Examination of the Deductive Logic of John Stuart Mill. London: 
Oxford University Press. 

Joseph, H.W.B. (1916). An Introduction to Logic, 2nd ed. Oxford: Clarendon Press. 

Kubitz, O. (1932). ‘Development of John Stuart Mill’s System of Logic’ in Illinois Studies 
in the Social Sciences, volume 18, nos. 1-2. Urbana, IL: University of Illinois Graduate 
School. 

Locke, J. (1975). An Essay concerning Human Understanding.  P. Nidditch (Ed.). Oxford: 
Clarendon Press. 

Malebranche, N. (1997). The Search after Truth. T. Lennon and P. Olscamp (Trans.). 
Cambridge: Cambridge Texts in the History of Philosophy. 

O’Rourke, K. (2001). John Stuart Mill and Freedom of Expression, The genesis of a 
theory. London: Routledge. 

Passmore, J. (1966). A Hundred Years of Philosophy, 2nd ed. London: Duckworth. 

Riley, J. (1996). ‘On Liberty and the Periclean Ideal’, Qwerty 6: 241-8. 

Rosen, F. (2003). Classical Utilitarianism from Hume to Mill. London and New York: 
Routledge. 

Rosen, F. (2004). ‘J.S. Mill on Socrates, Pericles and the Fragility of Truth’, Journal of 
Legal History 25: 181-94. 

Rosen, F. (2006, forthcoming). ‘Mill on Coleridge’, Telos. 

Rosen, F. (2007a, forthcoming). ‘From Jeremy Bentham’s Radical Philosophy to J.S. 
Mill’s Philosophic Radicalism’ in G. Stedman Jones and G. Claeys (Eds.), Cambridge 
History of Nineteenth Century Political Thought. Cambridge: Cambridge University 
Press. 

Rosen, F. (2007b, forthcoming). ‘Parallel Lives in Logic: The Benthams and the Mills’. 
Volume of essays from the London Mill Conference, April 2006. 



J.S. Mill on Logical Fallacies  147 

Ryan, A. (1987). The Philosophy of John Stuart Mill, 2nd ed. Basingstoke: Macmillan. 

Sanderson, R. (1618). Logicae Artis Compendium, 2nd ed. Oxford: J. Lichfield. 

Scarre, G. (1989). Logic and Reality in the Philosophy of John Stuart Mill, Dordrecht: 
Kluwer Academic Publishers. 

Skorupski, J. (1989). John Stuart Mill. London and New York: Routledge. 

Tomassi, P. (1999).  Logic. London and New York: Routledge. 

Urbach, P. (1987).  Francis Bacon’s Philosophy of Science:  An Account and Reappraisal. 
Chicago and La Salle: Open Court. 

Urbinati, N. (2002). Mill on Democracy, From the Athenian Polis to Representative 
Government. Chicago and London: University of Chicago Press. 

Varouxakis, G. (2002). Mill on Nationality. London and New York: Routledge. 

Whately, R. (1828). Elements of Rhetoric, 2nd ed. Oxford and London: J. Parker and John 
Murray. 

Whately, R. (1975). Elements of Logic (2nd ed., 1927). Ray E. McKerrow (ed.). Delmar, NY: 
Scholars’ Facsimiles and Reprints. 

Frederick Rosen 
40 Priory Gardens 

Highgate 
London N6 5QS 
United Kingdom 

f.rosen@acl.uc.uk