Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

ANSCOMBE’S INTENTIONS

Constantine Sandis
(University of Hertfordshire, UK)

Prologue
Philosophers commonly talk of Anscombe’s book Intention as a single unified
account produced at a particular moment in time. Yet we share a tendency to refer to the
text of the second (1963) edition while citing the first (1957) one1. Between these two
there also lies the 1958 second impression of the first edition, which already contained
some significant alterations noted by Anscombe herself.
This paper offers a detailed consideration of the differences between these three
publications, as well as her earlier Proceedings of the Aristotelian Society article
« Intention ». In what follows I catalogue the various changes she made with the aim of
clarifying whether it makes sense, in light of them, to talk of an early and late Intention. I
conclude that while this would be going too far, Anscombe’s views on action description
and practical reasoning have changed in ways in which it would be profitable for
philosophers of action and moral psychology to track.

1Examples include many of the essays in Diamond & Teichman (Intention and Intentionality : Essays for
G. E. M. Anscombe Sussex, Harvester Press, 1979) and Ford et. al. (Essays on Anscombe’s Intention, Boston,
MA, Harvard University Press, 2011), as well as Moran (« Anscombe on Practical Knowledge », in J. Hyman
& H. Steward ed., Agency and Action, Cambridge, Cambridge University Press, 2004, p. 43-68), Mayr
(Understanding Human Agency, Oxford, Oxford University Press, 2011), Sandis (The Things We Do and Why
We Do Them, London, Palgrave Macmillan, 2012), & Dancy & Sandis (The Philosophy of Action - An Anthology,
Oxford, Wiley-Blackwell, 2015). Exceptions include Teichmann (The Philosophy of Elizabeth Anscombe,
Oxford, Oxford University Press, 2008) who explicitly refers only to the 1963 edition (his bibliography
distinguishes this from the first edition but makes no mention of the second impression), and Bayne
(Elizabeth Anscombe’s « Intention », Charleston, CA, BookSurge Publishing, 2010) who states that all the
numbering refers to that of the second edition, but then writes « 1957 » after each individual reference.
Davidson (« Actions, Reasons, and Causes », in Essays on Actions and Events, 2nd revised edition, Oxford,
Clarendon Press 2001, p. 3-19 & « Intending », in Ibid., p. 83–102) mystically cites a « 1959 » edition to ;
more recent American philosophers (e.g. Thompson, Life and Action : Elementary Structures of Practice and
Practical Thought, Boston MA, Harvard University Press, 2008) tend to refer only to the Harvard University
Press 2000 reprint which is identical to the 1963 edition. Finally, it is also worth noting that while all four
of the existing translations to date (German, Spanish, French, and Italian) are of the 1963 edition, all but the
German omit to translate the page in which Anscombe notes the significant changes between editions.
Indeed, the Spanish translation mistakenly professes to be a translation of the 1957 text and the Italian one
gives no date at all, save for the mention of « 1957 » in Mary Geach’s preface).

74
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

I. Before Intention
Before Intention there was « Intention », the Proceedings of the Aristotelian Society
talk and subsequent article. In her Introduction to the book Intention Anscombe writes :
« Excerpts, with small modifications, comprising the discussion of the difference
between “motive”, “intention”, and “mental cause” formed an Aristotelian Society
paper delivered on June 3rd, 1957. »

The paper which was published in the 1956-7 Proceedings includes a brief bio which
curiously dates Intention to 1958. The section the material overlaps with are §5 and §§10-
15, minus the now notorious pages on non-observational knowledge, the original article
simply maintaining the general view without explication :
« …the subject is able to give the cause of a thought in the same kind of way that he is
able to state the place of his pain or the position of his limbs. Such statements are not
based on observation2. »

The « small modifications » which Anscombe notes are mainly comprised of the odd
italic, a switch from i.e. to e.g. (p. 17), and from « Humeian » to « Humian » (p. 16 – neither
caught on).

II. Actions and their Results : The First Edition and the Second Impression
The Second Impression of Intention (1958) includes the following note by
Anscombe :
« I have made a few alterations ; the only ones of any significance are on pp. 29, 58,
59, and 61. »

There are indeed no other significant differences between the two editions. With the
exception of a small change on p. 29 (§19) all the changes that Anscombe notes are in the
discussion of Aristotle’s distinction between practical and theoretical reasoning (§§33). I
return to those passages in the next section. First, I want to focus on the small change
which occurs in §19 (p. 29). Anscombe writes, in all three versions :
« What makes it true that the man’s movement is one by which he performs such and
such an action will have absolutely no bearing on the I that occurs, unless we suppose
a mechanism by which an I appropriate to the situation is able to occur because of the
man’s knowledge of the situation – he guesses. »

In the first edition this is followed by :

« e.g. that the movement of his hand will result in its grasping the hammer and so

2 G. E. M. Anscombe, « Intention », Proceedings of the Aristotelian Society LVII : 1956-7, p. 176.

75
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

the right I occurs. »

The Second Impression, by contrast, changes the example from one of hand
movement to one of muscular contractions :
« e.g. that his muscular contractions will result in his grasping the hammer and so
the right I occurs. »

Why does Anscombe push the example inward (« inside the body » as Jennifer
Hornsby would later claim when locating all actions there) 3 ? One possible reason would
be that she realizes that she is already committed to the movement of the agent’s hand
being identical to his (or his hand’s) grasping the hammer. If what we have here is indeed
one action with several descriptions (see Intention §§23-6) then it is misleading – though
not altogether false – to say that the movement of his hand results in its grasping the
hammer, since the two bodily movements are identical. The claim would not be strictly
speaking, false, because the movement of my hand is best conceived not as an event but,
rather, as an ongoing process that results in the hammer being grasped by the hand (a
state of affairs which, in a mood of nuance, we may nonetheless wish to distinguish from
the hand’s actual grasping of the hammer) 4.
A more technical distinction between an action and its result would later be
introduced by G.H. von Wright5. A result, so understood, forms a logical part of an action
and is, accordingly, meant to be distinguished from any effects or consequences. In this
sense of result, the movement of a person’s hand may only result in the hand’s moving
and not in the hammer’s being grasped. It is of course tempting to think that if the two
movements are identical then their (von Wrightian) results will be identical too. But the
lesson to take home here, I think, is that whether or not R is a result of M is a matter of
description, too.
What of muscle contractions ? The change of example seems to indicate that
Anscombe does not see them as identical to (outer) bodily movements, under any
description. While she would have agreed with Davidson’s explicit claim that « all actions

3 J. Hornsby, Actions, London, Routledge and Kegan Paul, 1980.

4 For the neo-Aristotelian view that actions – as well as mere bodily motions – are non-causal processes see,
for example, Stout (Things That Happen Because They Should, Oxford, Oxford University Press, 1996) and
Steward (« Actions as Processes », Philosophical Perspectives, 26 (1), 2012, p. 373-388). I shan’t concern
myself here with whether Anscombe is using « bodily movement » in the transitive or intransitive sense
(see Hornsby, Actions, op. cit., Ch. 1) since the process view applies to both uses of the term « movement ».
5 G.H. Von Wright, Explanation and Understanding, London, Routledge & Kegan Paul, 1971, p. 76ff.

76
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

are intentional under some description6 », Anscombe wouldn’t have included muscular
contractions as examples of bodily movements which may be re-described as actions. For
even if my contracting my muscles was numerically identical to my moving my hand, it
wouldn’t follow that muscles contractions were themselves identical to my hand
movements, let alone to my hand’s grasping of the hammer. The 1958 example sheds
clarity, for it is less plausible to identify a grasping of the hammer with muscle
contractions than it is to do so with hand movements.

III. Expression of Intention and Action Explanation 1958-1963

In her note on the Second Edition, Anscombe writes :
« For this edition I have made some small alterations in §§2, 6, 17, 33 and 34. »

§§33-4 correspond to the remarks on practical reasoning on pages 57-63. I shall deal
with these in the next section, alongside the changes which she had already made in these
passages between 1957 and 1958. The noted alterations in the earlier sections are all
small ones. I take them in turn.
In §2, Anscombe queries the difference between two sorts of expression which share
the superficial grammatical form « I am going to ». The examples she gives are « I am going
to be sick » and « I am going to take a walk ». « It is not illuminating », she writes in all
editions, « to be told that one is a prediction and the other an expression of intention »
(p. 2). Rejecting the thought that – unlike a prediction – the expression of an intention is
not a statement about the future but, rather, a description or expression of a further
specifiable present state of mind, Anscombe retorts that this answer renders it « difficult
to see why intention should be essentially connected with the future, as it seems to be7 ».
In the 1963 edition she disambiguates by replacing the word « it » with « the intention ».
The next miniature alteration involves the removal of the bracket phrase and the insertion
of « most of » between « that » and « the earliest », from the following sentence :
« No one is likely to believe that it is an accident, a mere fact of psychology, that those
states of mind which are intentions always have to do with the future, in the way that
it is a fact of racial psychology, as one might say, that the earliest historical traditions
(except for the Hebrew ones) always concern heroic figures (p. 2). »

The safer revision suggests that she most likely came across evidence of some

6 D. Davidson, (1971), « Agency », in Essays on Action and Events, Oxford, Clarendon Press, 2001, p. 50.
7 G. E. M. Anscombe, Intention, first edition, Oxford, Basil Blackwell, 1957 and 1958 (2nd impression), p. 2.

77
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

dispute about Hebrew traditions or, alternatively, found one or more additional
exceptions to her generalization. But the main point of the sentence - viz. that the future-
orientated aspect of intention is no contingency – remains intact. A final small change in
§2 occurs in the following sentence :
« A command will be a description of some future action, addressed to the prospective
agent, and cast in a form whose point in the language is to make the person do what
is described8. »

The Second edition simply replaces « command » with « imperative », thereby

correctly widening the scope of the remark to include imperatives that are not commands,
such as prescriptions or requests9.
The early versions of §6 open as follows :
« To justify the proposed account, “Intentional actions are ones to which a certain
sense of the question ‘why ?’ has application”, I need to explain both this sense and
what I call the question’s not having an application10. »

In the Second Edition this becomes :

« To clarify the proposed account, “Intentional actions are ones to which a certain
sense of the question ‘why ?’ has application”, I will both explain this sense and what
I call the question’s not having an application11. »

This time Anscombe’s claim is weakened quite considerably ; what is on offer is no

longer a defence but a clarification.
The last and least in this series of smalls sets of changes occurs at the outset of §17 :
« I can now complete my account of when the question “Why ?” has no
application12. »

« I can now complete my account of when the question “Why ?” is shewn not to
apply13. »

The motivation behind this change seems purely stylistic, to avoid repetition of the
word « application » which occurs again in the next sentence.

8 Intention, 1957 & 1958, p. 3.

9 Anscombe retains « command » throughout the rest of this section, presumably because she is
concentrating on the specific speech-act of giving an order. Thanks to Valérie Aucouturier for reminding me
that in §33, p. 58 she appeals to the imperative form to criticize a misunderstanding of the efficiency of
practical reasoning.
10 p. 11, italics in the original.
11 Intention, second edition, Oxford, Basil Blackwell, 1963, p. 11.
12 Intention, 1957 & 1958, p. 25.
13 Intention, second edition, 1963, p. 25.

78
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

IV. Practical Reasoning 1957-1963

As previously mentioned, the bulk of the changes between the First Edition and the
Second Impression take place in Anscombe’s account of Aristotle’s distinction between
practical and theoretical reasoning (§§33). These passages are subjected to further
changes in the second edition of 1963, confirming that as her thought on the topic
continued to progress she became increasingly unsatisfied with some of her earlier
formulations14. I begin with the first set of changes.
Page 58 of Anscombe (1957) includes the following footnote, appended to the claim
that one of the reasons why Aristotle’s « discovery » of « Practical Reasoning » has « been
lost to modern philosophy through misinterpretation » is that « [his] denials that practical
reasoning is “apodeictic” are taken to indicate, if correct, a perhaps regrettable but
presumably unavoidable defect in this sort of syllogism » :
« Not all of what Aristotle meant by this concerns me here. I shall be maintaining that
many practical syllogisms do not contain entailments (and that this is no defect) ;
which shews that they are not “apodeictic”. But some do ; and Aristotle’s grounds for
the thesis that these too are not ‘apodeictic’ are of no interest to my present enquiry. »

This note disappears from all subsequent versions of the text. Why did Anscombe
retract it ? There is no evidence to suggest that she has changed her mind on whether
Aristotle denied that practical syllogisms are proofs or, indeed, whether he was wrong to
do so. Nor do any of the remaining alterations suggest a newfound interest in Aristotle’s
grounds for the anti-apodeictic thesis. What is clear, however, is that Anscombe revisits
all these questions in her paper on « Thought and Action in Aristotle : What is Practical
Truth ? », which makes no reference to Intention 15 . Most probably, her continuing
dissatisfaction with §§33 was the direct result of her revisiting Aristotle’s texts as she
began to think about and write the 1965 paper, though we must not discount the
possibility of mistaking cause for effect here.
In the later paper, Anscombe primarily concerns herself with the view that Aristotle
needs the notion of intention to say some of the things we wants to say, and is wrong to

14 Even if Anscombe never as such changes her mind on practical reasoning, she seems to distinguish
between misinterpretations that are entirely the reader’s fault (and so require no change in the text) and
one’s which may have arisen due to inadequate turns of phrase or uses of example. Up to a point, the
distinction can easily be captured in terms of her own work on directions of fit (text to interpretation and
vice versa), though the requirement of a change to the text does not preclude the equal necessity of change
in the original interpretation.
15 The paper came out in 1965 but she was presumably writing it for quite some time before. She returned

to other aspects of this topic again in 1974, 1992, and 1993.

79
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

think that the notion of choice will suffice. It is to this end that she returns to the question
of practical syllogisms in relation to the apodeictic, writing :
« Every man has got to walk

I am a man

I have got to walk

is a formally valid deductive argument – I will call such an argument a proof-syllogism.

Now Aristotle had special ideas about proof, so he would not have agreed to say what
I have just said. “Every man has got to walk” is not a changeless truth, so he would
have said this is not apodeictic (see e.g. 1140a33-5). Disregarding this let us merely
note the formal validity of the reasoning as a deduction. Further, let us grant that if I
agree to the premises and therefore to the conclusion, and say “I have got to walk”,
speaking quite seriously, it would be queer of me not to walk, if nothing prevented
me16. »

As she would write in a later paper in critical deference to G.H. von Wright’s
approach to the topic :
« If there is practical inference, there must be such a thing as its validity17. »

Further below, having directed the reader towards Aristotle’s famous example of
the cloak, she adds :
« Now it is hard to tell whether Aristotle reflected that “I need a cloak” is not a formally
valid deductive conclusion from “I need a covering and a cloak is a covering”. The fact
that it is not, is, I should contend, no criticism of the syllogism as a piece of practical
reasoning. But it is possible that if he had been challenged about this, he would have
said one could amend the syllogism by putting in that a cloak was the best covering
or the easiest to make or something of that sort (cf. Nicomachean Ethics 112b16). For
he is marked by an anxiety to make practical reasoning out to be as like as possible to
speculative reasoning. “They are just the same” he says in the Movement of
Animals…and seems to be referring to a necessitation of the conclusion18. »

Anscombe is here confirming her 1957 claim that Aristotle was, paradoxically, on
the one hand prone to go out of his way to emphasize the similarities between practical
and theoretical syllogisms and, on the other, adamant in maintaining that the former are
never proofs. She continues :
« A further sign is that when he is looking at practical syllogism in this light – as
necessarily yielding the conclusion – his examples of the first universal premise
always go “It’s needed”, “It’s expedient”, “such and such a kind of being ought to do

16 E. Anscombe, « Thought and Action in Aristotle », From Parmenide to Wittgenstein, Oxford, Blackwell,
1965, p. 73.
17 E. Anscombe, « Practical Inference », in M. Geach & L. Gormally, Actions and Ethics : Essays by GEM

Anscombe, Exeter, Imprint Academics, 1974, p. 110.

18 Ibid.

80
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

such and such a kind of thing”. He wants a “must” in the conclusion in the verbalized
form in which he gives it in19. »

This brings us to Anscombe’s next change, on p. 59. She has just given the reader
two examples of Aristotelian practical syllogisms, the first taking the categorical form and
the second the imperative. In the first edition she subsequently writes :
« Both this and the Aristotelian example given before would necessitate the
conclusion for someone who accepts the premises. »

The 1958 version expands this sentence to the following two :

« Both this and the Aristotelian example given before would necessitate the
conclusion. Someone professing to accept the opening order and the factual premise
in the imperative example must accept its conclusion, just as someone believing the
premises in the categorical example must accept its conclusion. »

The addition just seems to make explicit what was previously implicit, but note the
odd switch from accepting the premises to professing to do so (in the imperative example)
and believing them (in the categorical one). Either this is a mere decorative change or, less
reasonably, we are being told that – in the imperative case - whether or not we must
accept the conclusion depends not so much on whether we accept the opening order as
on whether we profess to do so in the categorical case and whether we believe the relevant
premise in the categorical one (note that there is no talk of professing to accept the
conclusion in either case). Alternatively, if no more persuasively, the change could
indicate that it makes no sense to talk of accepting an imperative (though one may of
course profess to do so), and that in categorical cases accepting a premise is a matter of
believing it (whereas perhaps one cannot believe an imperative but only agree with – or
adhere to – it).
The final change from the First Edition to the Second Impression occurs at the start
of the final paragraph of §33 (p. 61). Anscombe originally writes :
« Only negative general premises can hope to avoid insanity of this sort ; and of course
negative general premises, if accepted, don’t lead to any particular actions (at least,
not by themselves or by any formal process). »

In the 1958 text this has been changed to :

« Only negative general premises can hope to avoid insanity of this sort. Now these,
even if accepted as practical premises, don’t lead to any particular actions (at least,
not by themselves or by any formal process) but only to not doing certain things. »

19 Ibid.

81
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

The sort of insanity in question is that of positive universal premises of the form
« Every human being needs to eat all the dry food he ever sees » (p. 60). At first sight the
two main changes between the versions of the comment are fairly minor. The first adds
the qualification « as practical premises » to highlight that even this wouldn’t help to
bridge the noted gap between acceptance and action. The second adds that while negative
premises don’t lead to particular actions, they can lead to omissions and refrainings.
While not implied in (though compatible with) the original statement, this is a pertinent
addition, not least in the wake of moot debates about whether or not omissions are
actions.20
The Second (1963) Edition makes further departures in its discussion of the
reception of Aristotle’s account of Practical Reasoning. Anscombe’s views on practical
reasoning were clearly in transition during this period, culminating in her
aforementioned 1965 paper. The first of these new changes occurs at the outset of §33. In
the early versions Anscombe writes :
« “Practical reasoning”, or “practical syllogism”, which means the same thing, was one
of Aristotle’s best discoveries, but it has been lost to modern philosophy through
misinterpretation. I think it is true to say that the general impression of “practical
syllogism” is that it is a form of reasoning about what one ought to do, leading up to
some conclusion : “I ought to do this” : it is, in short, an ethical syllogism. Aristotle’s
denials that practical reasoning is “apodeictic” are taken to indicate, if correct, a
perhaps regrettable but presumably unavoidable defect in this sort of syllogism21. »

The 1963 version replaces this passage with the following :

« “Practical reasoning”, or “practical syllogism”, which means the same thing, was one
of Aristotle’s best discoveries. But its true character has been obscured. It is
commonly supposed to be ordinary reasoning leading to such a conclusion as : “I
ought to do such-and-such”. By “ordinary reasoning” I mean only reasoning ordinarily
considered in philosophy : reasoning towards the truth of a proposition, which is
supposedly shewn to be true by the premises. Thus : “Everyone with money ought to
give a beggar who asks him ; this man asking me for money is a beggar ; I have money ;
so I ought to give this man some”. Here the conclusion is entailed by the premises. So
it is proved by them, unless they are doubtful. Perhaps such premises never can be
certain22. »

The removed claim concerning Aristotle’s denial that practical reasoning is

« apodeictic » is that to which Anscombe had originally appended the footnote which had

20 For a critical overview see K. Bach, « Refraining, Omitting, and Negative Acts », in T. O’Connor & C. Sandis
ed., A Companion to the Philosophy of Action, Oxford, Wiley-Blackwell, 2010, p. 50-57.
21 Intention, 1957b &1958, pp.57-8)
22 Intention, 1963, pp.57-8.

82
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

already disappeared by the second impression. Perhaps Anscombe was dissatisfied with
her thoughts on this, or perhaps she increasingly felt that the negative apodeictic point
was distracting readers from the main argument here. Either way, what we see here is a
shift of emphasis from the claim that the cause of the misinterpretation is Aristotle’s
(mistaken) insistence that practical syllogisms are never apodeictic (coupled with the
thought that this is a defect of some kind), to the more explicit assertion of what is being
obscured and how viz. that it is a mistake to think that what renders practical syllogisms
akin to theoretical ones is that they too are arguing for the truth of a proposition (namely
a normative one), as opposed to more practical conclusions such as « I’ll catch the 9.0123 ».
The earlier text emphasizes that the mistake is to think that Aristotle’s own view
was that practical reasoning is defecting in that it is a failed attempt to produce
entailments for normative propositions. The later one, by contrast, simply states that this
is not the true nature of practical reasoning, without mentioning Aristotle’s views on the
apodeictic or the idea of defective syllogism. What Anscombe seems to have changed her
mind about, then, is the truth or (at the very least) importance of the thought the
misunderstanding – now toned down to obscurification – stems from Aristotle’s remarks
about practical reasoning not being apodeictic. This change was no doubt informed by the
closer study of Aristotle exemplified in the 1965 paper discussed above.
Anscombe’s new strategy of rendering explicit what was previously implicit
continues through the slight extension from :
« Everyone takes the practical syllogism to be a proof – granted the premises – of a
conclusion » (p. 58).

to :
« Everyone takes the practical syllogism to be a proof – granted the premises and
saving their inevitable uncertainty or doubtfulness in application– of a conclusion »
(p. 58).

She next returns to Aristotle’s role in facilitating the obscurification of his own view :
« The cause of the mischief, though it is not exactly his fault, is Aristotle himself. It
was, I am sure, perfectly clear to him that he had found a completely different form of
reasoning from theoretical reasoning, or proof syllogism ; but it pleased him to give
cases of it which made it as parallel as possible to the theoretical syllogism24. »

« The cause of the mischief, though it is not entirely his fault, is Aristotle himself. For

23 Thanks to Roger Teichmann for pointing out to me that Anscombe doesn’t seem to mind much whether
the conclusion of a practical syllogism was said to be an action or its verbalized form.
24 Intention, 1958, p. 59.

83
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

he himself distinguished reasoning by subject matter as scientific and practical.

“Demonstrative” reasoning was scientific and concerned with what is invariable. As if
one could not reason about some particular non-necessary thing that was going to
happen except with a view to action !25 »

In the second edition this is followed by a long new passage which marks the most
significant change across all three versions of the book, and in whose light the motivation
for some of the changes noted above becomes clearer :
« “John will drive from Chartres to Paris at an average of sixty m.p.h. he starts around
five, Paris is sixty miles from Chartres, therefore he will arrive at about six” – this will
not be what Aristotle calls a “demonstration” because, if we ask the question what
John will do, that is certainly capable of turning out one way or another. But for all
that the reasoning is an argument that something is true. It is not practical reasoning :
it has not the form of a calculation what to do, though like any other piece of
“theoretical” argument it could play a part in such a calculation. Thus we may accept
from Aristotle that practical reasoning is essentially concerned with “what is capable
of turning out variously”, without thinking that this subject matter is enough to make
reasoning about it practical. There is a difference of form between reasoning leading
to action and reasoning for the truth of a conclusion. Aristotle however liked to stress
the similarity between the kinds of reasoning, saying [n. De Motu Animalium VII] that
what “happens” is the same in both. There are indeed three types of case. There is the
theoretical syllogism and also the idle practical syllogism [n. Ethica Nicomachea
1147a, 27-8] which is just a classroom example. In both of these the conclusion is
“said” by the mind which infers it. And there is the practical syllogism proper. Here
the conclusion is an action whose point is shewn by the premises, which are now, so
to speak, on active service. When Aristotle says that what happens is the same, he
seems to mean that it is always the same psychical mechanism by which a conclusion
is elicited. He also displays practical syllogisms so as to make them look as parallel as
possible to proof syllogisms26. »

The shift of blame increases slightly – almost amusingly - from « not exactly » to
« not entirely », but the important change is the new placement of the idle practical
syllogism in between the theoretical syllogism and the practical syllogism proper. In the
1957 and 1958 versions we are then told « Let us imitate his example, giving it a plausible
modern content » :
« Vitamin X is good for men over 60

Pigs’ tripes are full of vitamin X

I’m a man over 60

Here’s some pigs’ tripes » (p. 60)

25 Intention, 1963, p. 59.

26 Intention, 1963, pp. 59-60.

84
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

The second edition prefaces the same syllogism with « Let us imitate one of his
classroom examples, giving it a plausible modern content » (p. 60). Anscombe here
disambiguates between two kinds of practical reasoning, the idle (« which is just a
classroom example ») and the proper (whose « conclusion is an action27 »). The former,
she implies, is closer to theoretical reasoning than the latter.
This new addition28 also serves to illustrate that there can be theoretical reasoning
that is about action. The difference between theoretical and practical reasoning, then, is
not in its object but its form. Indeed, we could push this further and make space for a
fourth kind of reasoning namely practical reasoning whose conclusion is (an act or state
of) believing as opposed to the proposition believed (if indeed propositions are the things
we typically believe). Whatever similarity remains across all cases is conveyed through
Anscombe’s use of the umbrella term « elicited » to modify both propositions deduced and
actions reached.
The distinction between actual and (merely) academic forms of practical reasoning
is of vital importance to understanding Anscombe’s infamous discussion of non-
observational knowledge in §45. Her example, present in all editions, of the man « merely
conceiving speculatively how a thing might be done 29 » should clearly be read as an
instance of idle practical reasoning. Upon combining this understanding with
Theophrastus’ point about mistakes which rest in the performance rather than the
judgment, we come to appreciate that Anscombe’s suggestion that in the case of
intentional action « my knowledge is independent of what actually happens » (ibid.), is
restricted to practical knowledge proper. Indeed, the point that any misalignment
between what happened and what I intended to do is due not to a false prediction but to
my act’s misfiring is directly related to Anscombe’s contention that practical truth is truth
which we create in acting, though we can obviosuly « have theoretical knowledge of what
our present practical knowledge is of30 » (Anscombe 1993 : 156).
This brings us to the final change to the second edition. Anscombe originally opened
§34 with the following words :

27 But see footnote 103.

28 Introduced more fully in Anscombe’s « Thought and Action in Aristotle », art. cit., 1965, p. 76.
29 Intention, 1963, p. 82
30 This is but a mall first step towards understanding Anscombe’s account agential knowledge. For the

various interpretational controversies surrounding it see J. Schwenkler, « Non-Observational Knowledge in

Action », Philosophy Compass, vol. 7, 2012, p. 731-740 & « Understanding “Practical Knowledge” »,
Philosopher’s Imprint, vol. 15, 2015, p. 1-32.

85
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

« But there is no reason, even if our one desire were to follow Aristotle, to confine the
term “practical reasoning” to pieces of practical reasoning which look very parallel to
proof-reasonings31. »

In the 1963 version they are transformed to :

« But, we may ask, even if we want to follow Aristotle, need we confine the term
“practical reasoning” to pieces of practical reasoning which look very parallel to
proof-reasonings32. »

The sentiment remains the same, but now takes the form of a rhetorical question
rather than a categorical statement. Anscombe’s softening of tone could reasonably be
taken to suggest that she came to allow that there might be some pro tanto reason to
confine the term in the specified way, it is hardly a compelling one.
While there are no further changes to the text, due to the largeness of the
penultimate insertion the pagination of the 1963 edition from p. 59 never recovers back
to that of the earlier editions ; indeed the 1963 edition is almost a full page longer. This
difference would be trivial were it not for the fact that most people who cite the 1957 text
quote page numbers from that of 1963 or the identical Harvard reprint (2000).

Epilogue
So how significant are the changes between the various versions of Intention ? While
it would clearly be going much too far to talk of an early and late Intention, some of the
changes – particularly those on practical reasoning – are substantial enough to be of
interest not only to Anscombe scholars but to anybody interested in the philosophy of
action, moral psychology, and practical reasoning (e.g. the ongoing debate over whether
the conclusion of a practical syllogism is an action or a belief33).
I have set out to say something about each and every change Anscombe has made, if
only for completist reasons. It is likely that some of the smaller new additions were made
for typesetting reasons relating to pagination. Others are evidence of her extreme
precision, attention to detail, and devotion to getting things exactly right. Even if I am
mistaken in my assessments of them, I trust that – given that no synoptic edition of
Intention exists – this text will at least remain valuable as a catalogue of all the various

31 Intention, 1957b & 1958, p. 61.

32 Intention, 1963, p. 62.
33 For a very helpful critical overview see R.J. Wallace, « Practical reason », Stanford Encyclopedia of

Philosophy, 2014.

86
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

changes made34.

References
 Anscombe G. E. M., « Intention », Proceedings of the Aristotelian Society LVII : 1956-
7, p. 175-92.

 Anscombe G. E. M., Intention, first edition, Oxford, Basil Blackwell, 1957.

 Anscombe G. E. M., Intention, second impression, Oxford, Basil Blackwell, 1958.

 Anscombe G. E. M., Intention, second edition, Oxford, Basil Blackwell, 1963 ;

reprinted 1966 & 1985.

 Anscombe G. E. M., « Thought and Action in Aristotle : What is Practical Truth ? »,

in J. R. Bambrough ed., New Essays on Plato and Aristotle (London : Routledge and
Kegan Paul), 1965, p. 143-158 ; reprinted in her From Parmenides to Wittgenstein :
Collected Philosophical Papers Volume I, Oxford, Blackwell, 1981, p. 66-80, to
which any page numbers refer.

 Anscombe G. E. M., « Von Wright on Practical Inference » (1974), in P. A. Schilpp &

L. E. Hahn (eds.), The Philosophy of Georg Henrik von Wright, La Salle, III, Open
Court, 1989) ; reprinted as « Practical Inference », in Geach & Gormally (eds.),
2005, p. 109-147, to which any page numbers refer.

 Anscombe G. E. M., Absicht, intr. and ed. by J.M. Connolly, T. Keutner, Freiburg –
München, Karl Alber, 1986.

 Anscombe G. E. M., Intención, intr. J. Mosterín, trans. A. I. Stellino, Barcelona,

Paidós, 1991.

 Anscombe G. E. M., « Practical Truth », in Ruch filozoficzny, vol. 49, n.1, 1992, p. 30-
33.

 Anscombe G. E. M., « Practical Truth », in J. M. Dolan (ed.), Working Papers in Law,

Medicine and Philosophy, University of Minnesota Press, 1993 ; reprinted in Geach
& Gormally ed., 2005, p. 149-158, to which any page numbers refer.

 Anscombe G. E. M., Intention, second edition, Cambridge, MA, Harvard University

34Many thanks to Valérie Aucouturier, Vincent Descombes, Bruno Gnassounou, Rémi Clot-Goudard,
Lorenzo Greco, Erasmus Mayr, Marta Moreno, Cyrille Michon, E.A. Sandis, and Roger Teichmann for very
helpful discussions and advice.

87
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

Press, 2000.

 Anscombe G. E. M., L’Intention, trans. by M. Maurice & C. Michon, Préface by

V. Descombes, Paris, Gallimard, 2002.

 Anscombe G. E. M., Intenzione, trans. and int. C. Sagliani, preface by M. Geach, Roma,
Edizione Università della Santa Croce, 2004.

 Bach K. (2010), « Refraining, Omitting, and Negative Acts », in T. O’Connor & C.

Sandis (eds.), A Companion to the Philosophy of Action, Oxford, Wiley-Blackwell,
2010, p. 50-57.

 Bayne S. R., Elizabeth Anscombe’s « Intention », Charleston, CA, BookSurge

Publishing, 2010.

 Dancy J. & Sandis C. (eds.), The Philosophy of Action – An Anthology, Oxford, Wiley-
Blackwell, 2015.

 Davidson D., « Actions, Reasons, and Causes », Journal of Philosophy 60, 1963,
p. 685-700. Reprinted in Essays on Actions and Events, 2nd revised edition, Oxford,
Clarendon Press 2001, p. 3-19, to which any page numbers given refer.

 Davidson D., « Agency », in R. Binkley, R. Bronaugh and A. Marras (eds.), Agent,

Action and Reason, Toronto, University of Toronto Press. Reprinted in Essays on
Actions and Events, p. 43-61, to which any page numbers given refer.

 Davidson D., « Intending », in Y. Yovel (ed.), Philosophy of History and Action, D.

Reidly and The Magnes Press, 1978. Reprinted in Essays on Actions and Events,
p. 83–102, to D. Davidson, Essays on Actions and Events, 2nd revised edition,
Oxford, Clarendon Press 2001.

 Diamond C. & Teichman J. (eds.), Intention and Intentionality : Essays for G. E. M.

Anscombe, Sussex, Harvester Press, 1979.

 Ford A., Hornsby J. & Stoutland F. (eds.), Essays on Anscombe’s Intention, Boston,
MA, Harvard University Press, 2011.

 Geach M. & Gormally L. (eds.), Human Life, Action and Ethics : Essays by G. E. M.
Anscombe, Exeter, Imprint Academic, 2005.

 Hornsby J., Actions, London, Routledge and Kegan Paul, 1980.

 Mayr, E. Understanding Human Agency, Oxford, Oxford University Press, 2011.

88
 Klesis – 2016 : 35 – Lectures contemporaines de Elisabeth Anscombe

 Moran R., « Anscombe on Practical Knowledge », in J. Hyman & H. Steward(eds.),

Agency and Action, Cambridge, Cambridge University Press, 2004, p. 43-68.

 Sandis C., The Things We Do and Why We Do Them, London, Palgrave Macmillan,
2012.

 Schwenkler J., « Non-Observational Knowledge of Action », Philosophy Compass,

vol. 7, 2012, p. 731–740.

 Schwenkler J., « Understanding “Practical Knwoledge” », Philosophers’ Imprint, vol. 15,

no. 15, June 2015, p. 1-32 (http ://quod.lib.umich.edu/p/phimp/3521354.0015.015/1).

 Steward H., « Actions as Processes », Philosophical Perspectives, 26 (1), 2012,

p. 373-388.

 Stout R., Things That Happen Because They Should, Oxford, Oxford University Press,
1996.

 Teichmann R. (ed.), Logic, Cause and Action, Cambridge, Cambridge University

Press, 2000.

 Teichmann R., The Philosophy of Elizabeth Anscombe, Oxford, Oxford University

Press, 2008.

 Thompson M., Life and Action : Elementary Structures of Practice and Practical
Thought, Boston MA, Harvard University Press, 2008.

 Von Wright G. H., Explanation and Understanding, London, Routledge & Kegan Paul,
1971.

 Wallace R. J., « Practical Reason », Stanford Encyclopedia of Philosophy, 2014,

retrieved 7 April 2015 (http ://plato.stanford.edu/entries/practical-reason)

89