This Critical Notice was published in Dialogue, (1995) vol. 34, no. 4, pp. 797-806.

Topology in Informal Logic: Slippery Slopes and Black Holes

Douglas Walton, Slippery Slope Arguments
(Oxford: Oxford University Press, 1992), xiii+296, and
James Davies, Ways of Thinking
(New York: Peter Lang, 1991), xiv+258.

NORMAN SWARTZ Simon Fraser University

The commonalities of Douglas Walton's Slippery Slope Arguments and James Davies's Ways of Thinking are obvious: both are written by Canadian philosophers; both lie within the broad field of informal logic; and both make appeals in support of dialogical reasoning. But there the similarities end. The former is the work of a prolific author writing a treatise focusing narrowly on one topic within informal logic; the latter is the product of a newcomer to book-writing, and his is a textbook intended for beginning students.

     Judging from my own experience and from conversations with colleagues, it seems that philosophers, generally, tend to be slow, methodical, readers. But if ever a book warranted speed-reading, it is Walton's Slippery Slope Arguments. Walton is, to be sure, a high-minded philosopher; he writes serious philosophy; there is good philosophy within the covers of Slippery Slope Arguments. Even so, the book is heavily padded, and – like some of his previous work – maddeningly repetitious. Here is one of numerous examples (this one, you will notice, occurring within the span of just two pages):

Practical reasoning is a goal-directed type of argumentation which takes knowledge of an agent's situation into account in guiding the agent on how it is reasonable to act in a situation of incomplete knowledge (pp. 87-88).

The conclusion of a practical inference is an imperative to a course of action for a particular agent (p. 88, top of page).

Practical reasoning is a goal-driven, knowledge-based, action-guiding kind of argumentation that concludes in a practical imperative of action in relation to the (usually imperfect) knowledge that an agent has of ways and means to proceed in a particular situation (p. 88, middle of page).

The conclusion of an instance of practical reasoning is an imperative that directs a planner (or agent) to a specific course of action (or inaction) in a particular situation, as it is known to the planner (p. 89).

     There are also occasional lapses into banality:

The planner tries to forecast or imagine the possible good consequences of the planned action, and also the possible bad consequences. The good or favorable consequences tend to support the action being considered, while the bad or negative consequences have the opposite effect (p. 90).

     This book cries out for the scissors of a ruthless editor who would have left much of the text on the cutting room floor.[Note 1] In an earlier review[Note 2], I characterized another of Walton's books[Note 3] as containing enough good material for a book half its size. In the present instance, Walton calls this volume a monograph To this I would reply that it contains enough good material for a stout pamphlet.

     Another failing common in Walton's writings is his predilection to use standard philosophical terminology in sloppy ways, especially when writing of causation. Example: "causality is not a matter of necessity, especially when a long sequence of causation is concerned" (p. 102). One must be puzzled over the appending of the qualifying clause. How could necessity depend on the number of steps in a causal sequence?

     Yet, in spite of the dross and the frequent clumsiness, there is – as I have said – a core of good philosophy in this book. The chief merit is that – probably for the first time – a writer pulls together material from many writers (esp. Govier; Little, Groarke, and Tindale; and Perelman and Olbrechts-Tyteca) and synthesizes from it a sustained examination of slippery slope arguments.

     Walton begins his study by emphasizing that slippery slope arguments are not invariably, or even for that matter usually, fallacious arguments.[Note 4] For example, I recall my own father warning me of the dangers of beginning smoking, how the habit may become addictive and be extremely difficult to break. His was a slippery slope argument: "If you begin to smoke, you may become a heavy smoker and find it very difficult to quit." There was nothing fallacious about his argument; it was a perfectly sound piece of practical reasoning.

     The problem then becomes that of attempting to classify the kinds of slippery slope arguments; of trying to distinguish sound, or proper, uses of slippery slope arguments from fallacious ones; and, finally, to review strategies for refuting or rebutting fallacious uses of such arguments.

     In the classificatory scheme Walton devises, there are four kinds of slippery slope arguments: (1) the Sorites; (2) the Causal; (3) the Precedent; and (4) the so-called Full. Oddly, Walton never raises the question whether this scheme is exhaustive.

     The Sorites version of the Slippery Slope argument involves the gradual extending of a vague term from cases where it clearly applies to cases where it does not apply or does so only problematically. A typical example is to be found in disputes raging over the morality of abortion. If it is conceded that it is immoral to kill a fetus just prior to birth, how can one draw a line after which it is immoral to kill a fetus but prior to which it is morally permissible? As we all know, some persons have used slippery slope arguments, in particular the Sorites version, to insist that no such line exists: that abortion, from the very moment of conception, is morally impermissible.

     The Causal version is illustrated in the 'Domino-theory' put forward to justify American involvement in the War in Vietnam: "If Vietnam is allowed to fall to the Communists, other countries in Southeast Asia will similarly fall."

     The Precedent version (typically) involves arguing against a certain proposal because there are similar cases which would seemingly merit similar treatment and thus would expand, undesirably, the scope of the initial proposal. For example: "Allowing French parents to choose to send their children to French language schools which would have public support would be a precedent for allowing parents of Indian, Ukrainian, German and many other language groups to do the same thing. …" (p. 125).

     The descriptive title "full" that Walton adopts (borrowing from Govier) for his fourth category is somewhat misleading (p. 162). For it carries with it the implicit suggestion that the other three kinds of slippery slope arguments are not 'full', i.e. are somehow sketchy or incomplete. Perhaps a more apt term might have been something of the sort "multidimensioned" or "composite". In this fourth, last, kind of slippery slope argumentation, no fewer than eight features are present: "(1) argument from gradualism, (2) argument from consequences, (3) practical reasoning, (4) argument from analogy, (5) argument from popular opinion, (6) argument from precedent, (7) causal argumentation, and (8) the sorites type of argumentation" (p. 161). (Needless to say, such complex arguments will tend to be rather lengthy, not lending themselves to quotation in a review.) Walton offers examples of 'full' slippery slope arguments in case studies involving euthanasia, censorship, and medical research using fetal tissue.

     Throughout his discussions, Walton is concerned with laying out the logic of slippery slope arguments and with the kinds of moves one might make in practical argument to support or rebut a slippery slope argument. He is much less concerned to examine the particular strengths and weaknesses of specific examples. Thus readers hoping or expecting to find in this book detailed and lengthy examinations of the substantive issues involved in major contemporary disputes in which slippery slope arguments feature prominently will be disappointed.[Note 5] For example, after having invoked the abortion issue several times in the book for illustrative purposes, Walton's final word leaves the matter hanging in an extraordinarily indecisive fashion:

Both [disputants] had equally strong arguments, and as a result, neither argument gained an ascendancy that could resolve the dispute. A good tactic to break the deadlock would be for one side or the other to take the initiative and make a case for the beginning of personhood at some point on the continuum other than the two poles of conception and birth, rejected by the other side.

     For example, if one of them were to argue that a fetus becomes a person at the last point in the first trimester, it would break the deadlock. But of course this argument would have to be backed up by good reasons for picking this point, or else the other party might insist that this point is arbitrary, using the slippery slope argument again. (pp. 263-4)

In short, this book examines the dialogical structures of four kinds of slippery slope arguments; it does not, but neither does it pretend to, examine the contents of specific disputes that turn on slippery slope arguments.

     Where Walton's book, to date anyway, has the field pretty much to itself (a computer-search of several major university libraries turned up no other book devoted entirely to the logic of slippery slope arguments), Davies's Ways of Thinking must compete in a market supersaturated with books, many in late editions and by well-known authors (and in some cases, teams of authors) backed by the resources of several of the largest textbook publishers. To carve out any piece of this cutthroat market, Davies's book would have to offer something both unique and worthy.

     There is no question that Davies's book is unique. Indeed it attempts to create a wholly new sub-genre within informal logic. Davies explains:

This is a "thinking book" with a difference – besides developing a method of critical thinking in general conformity with college and university practice, it sets such a method within the larger context of different ways of thinking. Such "ways of thinking" are, surprisingly, alien to contemporary academic practice with its dependency on one way of thinking, that which has become enthroned in science and technology, namely linear thinking. This dependency has been largely unreflective … (p. xi).

The frame of reference of this book is phenomenological and existential. 'Existential' means that philosophy is related to history and existence and it is not adequate thinking and perception without reference to these realms. … 'Phenomenology' means openness to the phenomena as they are, approaching them without preconceptions; no theories or 'descriptions' are accepted which prescribe antecedently how phenomena are to be perceived. (p. 246).

In addition to 'linear' thinking, Davies distinguishes five others: lateral, literary, meditative, historical-existential, and metaphysical.

     If Davies were able to make good on his stated program, Ways of Thinking would be one of the most important philosophical books of the decade. For the thesis he is advancing is audacious, and, if sustained by evidence and argument, would be nothing short of revolutionary. Unfortunately, the support for his thesis is so weak, unconvincing, and clumsy that it actually undermines his program.

     Having expressed his displeasure at the prominence traditionally accorded what he calls 'linear' thinking, Davies begins by devoting six (out of ten) chapters precisely to advancing this mode of thinking (some 58% of the total pages).

     The book is a shambles: it is naive (in the worst sense of that term) and is error-ridden. No chapter escapes presenting some major falsehood. "A conclusion, in philosophical argument [which Davies illustrates with a trivial syllogism], is that proposition which follows from the reasons and breaks new ground" (p. 8). Davies appears to have forgotten, or overlooked, Mill's and others' worries whether every deductively valid inference is a petitio principii. "An argument contains two or more propositions, one of which follows logically from the other(s) as a conclusion" (p. 9). Indeed?! On such an account, there would be no deductively invalid arguments, no inductively valid ones, and no inductively invalid ones.

     In his explication of Venn diagrams (p. 23), Davies writes as if the invalidity of the inference from "All S is P" to "All P is S" validates the inference "All S is P ├ (All P is S)" (p. 23).

     In his explanation of necessary and sufficient causal conditions, Davies includes some mighty queer examples: "Election, in a democracy, [is] a necessary condition of sovereignty" (p. 43). Did I overlook the Queen's being elected? "Education is a necessary condition of professional certification" (p. 43). Legally necessary, perhaps, but surely not causally necessary. "Ice is a sufficient rather than necessary condition of hockey because hockey can be played on grass and floor as well as ice" (p. 43). Here Davies seems to be presupposing that if a condition is not a necessary condition, then it is sufficient! But not only is that general presupposition false, his specific conclusion is likewise, since clearly there are instances of ice, e.g. the surface of my driveway two winters ago, where no hockey whatever has been played. "A quarter is a sufficient, but not necessary condition, of there being a penny in my pocket" (p. 45). That's strange: here in Canada, quarters are made of nickel, while pennies are made of copper.

     The symbolism of the sentential calculus — viz. "", "•", "∨", "⊃", and "≡" — is cursorily introduced (p. 39), seemingly more as a gesture than with conviction. The symbols are not given truth-table definitions; no inference rules are offered for their use; and they are actually used subsequently on only four pages (40-41 and 43-44).

     On page 75, Davies explicates the denial of the universal affirmative, "All politicians are dishonest", with the ambiguous barbarism, "All politicians are not dishonest".

     Davies argues that deductively valid arguments are both truth-preserving and falsity-preserving: "If premise-propositions are false or problematic, the conclusion of the argument will also be false or problematic unless the reasoning is faulty (that is, fallacious, in which case it might fallaciously be true)" (p. 170). Nonsense. Any reader who has had even a modicum of exposure to a good logic course will find it a trivial matter to generate counterexamples, e.g., "Former Prime Minister Trudeau has six sons. Therefore Trudeau has at least two sons." The premise(-set) is false; the conclusion is true; and the reasoning is not fallacious.

     In chapter seven, Davies offers explications of four different theories of truth, to which he gives the names, "correspondence", "coherence", "pragmatic", and "ontological". Although in the first three cases he claims to be explicating traditional theories, the analyses given can be considered, if not simply bizarre, then, at the very least, glaringly inept. For example, in what Davies calls the "correspondence" theory, we find something remotely akin to a Tarskian theory, but only remotely. Davies's version seems to be a cocktail of semantic features along with epistemological ones (in particular, verificationist ones). "If one says 'E=mc2', for the formula to be true, there should be a correspondence between it and a physical experiment, or series of experiments, in which it is demonstrated" (p. 184). To this he adds, "there are a number of different kinds of correspondence in surprisingly diverse areas. We cite three: scientific, superstitious, and religious" (p. 184).

     In his discussion of coherence, Davies attributes to Gödel (whose given name Davies consistently misspells as "Curt"), a perfectly preposterous proof, viz. "that all mathematical-logical systems will eventually generate paradox" (p. 185). He then continues: "The consequence of the truth of this theorem is that coherence must be broadened sufficiently to include it [i.e. paradox]. It is to be noted, also, that paradox differs from inconsistency because it is a special situation in which a proposition is both true and false. The fact that a system may generate paradox does not, however, mean that inconsistency is accepted into a system as a general rule. Paradox is an exception to the rule. It appears to be the case that, in a complete system, even paradox coheres" (pp. 185-6). I am sure that this quotation does not itself cohere [sic].

     Jumping ahead to the 'ontological' theory of truth, we read: "The ontological theory provides a theory by which literary expression can be related to truth. Heidegger believes that truth is 'the unveiling of being'. Using this belief, it is possible to understand literary thinking as making possible such an unveiling through analogy, metaphor, myth and symbol" (p. 187). I wonder, will beginning students in critical thinking – the intended audience – be able to make sense of this? I doubt it.

     Chapter eight is devoted to dialogical thinking, a topic which has occupied other writers, e.g. Walton, much in recent years and whose writings are far more penetrating and illuminating.

     Chapters nine and ten are, finally, given over to arguing that attention should be paid to 'other modes' of reasoning: the aforementioned lateral, literary, meditative, and metaphysical. If there were any merit to be found in this book, it would have to be in these last two chapters.

     Not surprisingly these latter chapters are of no better quality than their predecessors. Chapter nine suffers from a complete ignoring of the difference between the context of discovery and the context of justification. Few, if any, other authors nowadays present the fare of standard informal logic textbooks as canons of discovery, and to extol the (supposed) creative virtues of 'lateral' thinking, etc. in contrast to what Davies calls 'linear' thinking, is to egregiously misrepresent the claims made on behalf of the latter.

     No one who has read Popper on falsifiability could ever seriously advance the inane argument Davies does on page 223. There, Davies is trying to commend the technique of 'free association'. In a reportedly actual case (no reference given), someone 'free associates' on the word "air". The first few words, we are told, have 'obvious' connections to "air", but then succeeding words – viz. "artery", "fair", and "art" – seem not to. But Davies has no trouble finding the (supposed) connection: "There is, initially, no apparent connection but, on reflection, it is noted that arteries carry oxygen, a vital component of air" (p. 223). Davies needs to read Popper's critique of Adler: "What, I asked myself, did it confirm? No more than that a case could be interpreted in the light of the theory. But this meant very little, I reflected, since every conceivable case could be interpreted in the light of Adler's theory…"[Note 6] Like Adler, Davies ingenuously believes that the impossibility of disconfirmation is powerful support for a theory. He then continues: "Everything is related rationally to everything else, so that, however disparate any two things may appear to be, they turn out to be related. … The idea here is that a (rational) relationship can be discovered between any two entities whatsoever. … If everything is somehow related to everything else, lateral thinking is not necessarily unrelated, in some way, to what you are reasoning about no matter how remote or unconnected it may appear to be. If this is so, lateral thinking is an essential and powerful catalyst to thinking" (pp. 224-5). I will not even attempt to catalog the cascade of blunders in this passage.

     The ensuing few pages contain advice to "relax in the consciousness of infinity"; repeat a mantra; and "utter quietly the word 'OM', an 'unstruck sound' or 'primordial vibration'" (p. 230). In the final section, "Metaphysical Thinking", Davies thoroughly obscures the act/content distinction, arguing that thinking about metaphysical issues is a different manner of thinking than exhibited in, e.g., thinking about scientific matters.

     In short, what one finds in this penultimate chapter is the sort of blather contained in scores of books which line the shelves of third-class bookstores in shopping malls. We have here, certainly not all, but many of the buzzwords of new-age 'metaphysics' and its 1960's precursors.

     The final chapter, "Thinking and Existence", while containing some viable claims about collective reasoning, remains, for the most part, vapid and distressingly obscurantist. "For thinking to be vital at all, it must issue from consciousness and being which are sufficiently centered and powerful enough to be able to generate authentic individuality" (p. 251). Davies asserts that "a centered and strong consciousness and being can be created only through a community in which they are celebrated and nurtured" (p. 252). He thinks that certain communities are better suited to this task than others. His favored nominees – but with strict reservations – include communities of philosophers, trade unions, and theatre companies (p. 252).

     I am in no position to gauge the worth of this book relative to all other textbooks in informal logic. But this much I can say with assurance: of the fifty or more such books currently on my shelves, every one of them is significantly better than Ways of Thinking. Its being reviewed along with Slippery Slope Arguments, does nothing, I hope, to tarnish Walton's book. Walton's book, as explained, has its flaws (mostly stylistic), but it also has very considerable merit and warrants a place in research bibliographies in modern informal logic. Davies's book, in contrast, deserves to be ignored.


  1. It would also have profited from a copy-editor who was sensitive to misplaced modifiers. For example:

      This technique developed into the technique used by Socrates in the early dialogues called the elenchus (p. 23).
      … in an everyday discussion of a controversial topic like drug abuse by nonexperts… (p. 78).[ Resume ]
  2. Norman Swartz, "A Guide for the Disputatious", Dialogue, 30 (1991), pp. 123-8.[ Resume ]
  3. Douglas Walton, Informal Logic: A Handbook for Critical Argumentation (Cambridge: Cambridge Univ. Press, 1989).[ Resume ]
  4. He points out that some informal logic texts treat the slippery slope argument exclusively as a fallacy. But then, true to his style, he repeats the point time and again, e.g., on pages 2, 13, 15, 29, 103, 207, 242, 280, etc.[ Resume ]
  5. For the latter, 'issues-focused', approach to slippery slope arguments, see David Lamb's Down the Slippery Slope: Arguing in Applied Ethics (London: Croom Helm, 1988). Walton's and Lamb's books are remarkable complements of one another.[ Resume ]
  6. Karl Popper, Conjectures and Refutations (New York: Basic Books, 1962), p. 35, italics added.[ Resume ]

Return/transfer to Norman Swartz's Philosophical Notes

Return/transfer to Norman Swartz's Home Page