flew-the philosophical quarterly-1959-.pdf

8/13/2019 Flew-The Philosophical Quarterly-1959-.pdf

http://slidepdf.com/reader/full/flew-the-philosophical-quarterly-1959-pdf 1/18

TH

PHILOSOPHICAL

QUARTERLYVOL. 9. N o. 34 JA NU AR Y 1959

HUME'S CHECK

Int ere st in Hu m e's essay Of Miracles is stil l very m uch alive.1

Inspite, or perhaps because, of this his position is constantly misunderstood.

Both the hostile A. E. Taylor, who argued that the only possible motive

for publishing it was a simple craving for no torie ty at an y cost ;2 and the

sympathetic T. H. Huxley, seem completely to have missed the point of

the more distinctive features of the argum ent. While every other commen

tator whom I have been able to consult seems to have made some mistakes,

mo re or less im po rtan t. We proceed to examine the text, to find out w hat

precisely Hume did actually say.I

The first thing to notice is that this, like five other essays in the Enquiry

concerning Hum an Understanding, is divided into P ar ts . Though gen erally

overlooked, this division should suggest natural sections in the argument :

which in this case, as in the others, we find. Part I moves, with momentary

lapses, a priori : and proceeds to a very general conclusion, about the evi

dence required to establish tha t a miraculous event had occurred. P ar t I I

summons mainly a posteriori considerations to determine the practical im-'Thus in 1955 the journa l o f the Amer ican Assoc ia t ion for the Advancement o f

Sc ience publ i shed D r . G . R. P r ice urg ing : th a t the ' ps i -phen om ena ' be long to theca teg ory of th e miracu lou s ; an d th a t , app ly ing a rgu m en ts from th is e ssay , we shouldente r ta in the idea of sys temat ic f r aud by a l l compe ten t exper imente r s r epor t ing pos i t iveresul ts (See Science for Au g. 1955 an d Ja n . 1956) . Th en in 1957 we find Mr. Chr is toph erHol l i s a rgu ing wi th in an impl ic i t ly Humean f ramework to an oppos i te conc lus ion .We must accep t tha t Chr is t rose f rom the dead; s ince i t would have been an evengrea te r p rod igy had a l l our au thor i t ie s been d ishones t o r mis taken (See Spectator for20/4/57 and succeeding weeks) .

a< D a v id H um e a nd the M i r a c u lous ' i n h i s Philosoph ical Studies (Macmillan,

1934), p. 332.

* *» < - _ _ -a

b y g u e s t on J a n u a r y 3 0 ,2 0 1 4

h t t p : / / p q . ox

f or d j o ur n a l s . or g /

D o wnl o a d e d f r om

http://pq.oxfordjournals.org/

































ANTONY FLEW

plications of this it is concerned primarily to establish one more particular

proposition, th a t a miracle can never be prove d, so as to be th e fo und ation

of a sys tem of religion .

1. Characteristically Hume starts by referring respectfully to an argu

m en t developed against th e peculiarly R om an Catholic dogm a of th e realpresence (tran sub stantia tion) by D r. Tillotson. Tillotson ha d been Arch

bishop of Ca nterbury and one of Locke's La titud ina rian friends. H um e

proceeds :

Nothing is so convenient as a decisive argument of this kind,

which must at least silence the m ost arrog ant bigo try and super

stition , an d free us from their im pe rtin en t solicitatio ns. I flatter

myself that I have discovered an argument of a like nature, which

if just, will, with the wise and learned, be an everlasting check toall kinds of superstitious delusion, and consequently, will be useful

as long as the world endures. 3

Three points here. Fir st : Hu m e offered a defence aga inst th e im pe rtin en t

solicitations of bigotry and supe rstition , an d no t an offensive weap on

capable of disproving any claims m ade . Second : even as a defence it is

supposed to serve only as a check , an d no t as an insuperab le bulw ark.

Th ird : even these limited functions it can fulfil on ly w ith th e wise and

learned .2. Hume now argues :

Though experience be our only guide in reasoning concerning

matters of fact, it must be acknowledged that this guide is not al

together infallible, but in some cases is apt to lead us into errors. . .

A wise m an therefore proportions his belief to the evidence . . .

(p. 110).

His presentation is in terms of his own peculiar psychological-logical arith

metical-mechanical theory of belief.

B ut this can be ignored—at leasttem pora rily. H e goes on :

To apply these principles to a particu lar instan ce ; we m ay observe,

that there is no species of reasoning more common, more useful,

and even necessary to human life, than that which is derived from

the testimony of men, and the reports of eye-witnesses and spec

ta to rs (p. 111).

Now this involves an appeal to experience.

Were not th e m emory tenacious to a certain degree ; ha d n ot m en

commonly an inclination to truth and a principle of probity; were

they not sensible to shame when detected in falsehood : Were not

these, I say, discovered by experience to be qua lities, inhe ren t in

human nature, we should never repose the least confidence in human

testi m on y (p. 112).

Of course assessment of testimony is a complicated and tricky business.

*E.H.U., p . 110 , i t a l ic s H um e 's : a l l r e fe rences wi l l be g iven to th e s t an da rd O.U .P .edi t ion of 1902, by L. A. Selby-Bigge.


h t t p : / / p q . ox




































HUME'S CHECK 3

Where

experience is not entirely uniform on any side, it is attended with

an unavoidable contrariety in our judgements, and with the same

opposition and mutual destruct ion of argument as in every other

kind of evidence (p. 112).Thus :

Suppose, for instance, that the fact which the testimon y endeavours

to establish, partakes of the extraordinary and the marvellous ; in

that case, the evidence, resulting from the testimony, admits of a

diminution, greater or less, in proport ion as the fact is more or less

unusual . The reason why we place any credit in witnesses and his

torians, is not derived from any connexion, which we perceive a

priori, between testimony and reality, but because we are accustomed

to find a conformity between them.B u t in this special case :

The very same principle of experience, which gives us a certain

degree of assurance in the test imony of witnesses, gives us also . . .

another degree of assurance against the fact, which they endeavour

to establish ; from which c ontrad iction the re necessarily arises a

counterpoise, and mu tual destruct ion of belief and a uth ority (p. 113).

Most appositely he quotes the Roman proverb : ' I should not believe such

a story were it told me by Cato '.

But in order to increase the probability against the test imony of

witnesses, let us suppose, that the fact, which they affirm, instead

of being only marvellous, is really miraculous ; and suppose also,

t h a t the testimony considered apart and in itself, amounts to an

entire proof; in that case there is proof against proof, of which the

strongest must prevail, but still with a diminution of its force, in

proportion to t h a t of its an tagonis t . For a miracle is a violation

of the laws of na ture ; and as a firm and unalterable experience has

established these laws, the proof against a miracle, from the verynature of the fact, is as entire as any argument from experience can

possibly be imagined (p. 114). There must . . . be a uniform

experience against every miraculous event, otherwise the event would

not merit that appellation (p. 115 : italics mine).

Hume concludes :

The plain consequence is (and it is a general maxim worthy of our

at tent ion) , ' Tha t no test imony is sufficient to establish a miracle,

unless the test imony be of such a kind, that its falsehood would be

more miraculous, than the fact, which it endeavours to establish ;

and even in that case there is a mutual destruct ion of arguments,

and the superior only gives an assurance suitable to that degree of

force, which remains, after deducting the inferior ' (pp. 115 f.).

3. We can now better appreciate the force of the three points underlined

earlier, (a) First : Hum e's argument is essentially defensive. He is not

t rying to prove a priori t h a t any event is impossible, but to show that there

mus t be peculiar and important difficulties inherent in any a t t empt to


h t t p : / / p q . ox




































4 ANTONY FLEW

establish that an event properly describable as miraculous has in fact occur

red. H e has not forgotten his funda m ental conte ntion : wh ateve r is in

telligible, and can be distinctly conceived, implies no contradiction, and can

never be proved false by any demonstrative argument or abstract reasoning a

priori (p. 35 : italics mine). His ab stra ct reasonings are concerned withthe concept of the miraculous. W hat he is trying to dem onstrate a priori

is : no t tha t, as a m att er of fact, miracles do no t happ en ; bu t th at , from th e

very nature of the concept, there must be a conflict in the evidence required

to show that they do.

(6) Second : H um e's arg um ent in P a rt I is offered precisely and only

as a check ; an d the O.E.D . affords no reason for believing th a t this

word carried a stronger meaning then than now.4 Taylor is m istake n to

charge inconsistency : Hu m e originally proposed to show superfluousany examination of evidence in certain cases, but later abandoned this

conten tion th a t there is a whole class of cases in which testim on y m ay

properly be dismissed without exam ination .5 The procedure which Hum e

actua lly is recomm ending he describes in the last pa rag rap h of P a rt I. I t

is his favourite example.

When anyone tells me, that he saw a dead man restored to life,

I immediately consider with myself, whether it be more probable,

that this person should either deceive or be deceived, or the fact,which he relates, should really hav e hap pen ed. . . If th e falsehood

of his testimony would be more miraculous, than the event which he

relates ; then, and n ot ti l l then, can he pretend to comm and m y

belief or opinion .6

H e makes very clear which estim ate he favou rs. B ut this does no t show

th at he is propounding the paradox th at the principle th at ' belief should

always be proportioned t o evidenc e ' justifies, in a certa in class of cases,

rejection of testimony without exam ination. . . . If anything , the very

reverse. I t is only later, in P a rt I I , th a t he advo cates a general resolution

to dismiss testimony in one category of cases as undoubtedly unreliable

(p. 129. See I I (4) below). As thi s is only after an d on th e basis of an ap pe al

to a posteriori considerations, it is not exactly, though it is like, dismissing

testimony without exam ination .

(c) Third : H um e's arg um ent is directed specifically to the wise an d

learned . There seem to be several reasons for inserting this app eal into

this essay in particular. There is the sub tlety of the argu m ent itself, which

has since betrayed some even of the wise and learned into misconstructions.'Co m pare : B u t a l l th e checks he received were insuff icient to m od era te his career

(Smollett , 1751).

'Ta ylo r , p . 339 : i ta l ics his . Co m pare three oth er passages in wh ich he insis ts s im ilar lyon misread ings of H um e 's care fu l ly s ta ted prog ram m e : Th e proposed d ismissa l o fte s t im ony wi th ou t exam ina t io n has th us led to no th in g . . . (p . 338 ) ; W e mi ghtthe re fore suppose tha t the benefi t p romised . . . to the ' wise ' would prove to be theabol i t ion of Chr is t ian i ty (pp . 341-2) ; an d th a t ve ry sc ru t iny of the te s t im ony inth e par t icu lar case f rom which Hu m e prom ises to del iver ' the wise ' (p . 345) .

'E.H.V., p . 116 : no te H um e' s s ignificant choice of th e wo rd co m m an d ; whic htends to conf irm our f i rs t , and, to a lesser extent , our third points .


h t t p : / / p q . ox




































HUMB'S CHECK 5

Hume later remarks how after a miracle story has been accepted by the

credulous m ultit ud e, no m eans of detection rem ain, but those which m ust

be drawn from the very testimony itself of the reporters : and those, though

always sufficient with the judicious and know ing, are commonly too fine to

fall under the comprehension of the vulgar (p. 127, italics mine). The re isthe fact that only those holding aloof from such credulous passions will be

in the sceptical defensive postur e to welcome some check on the im

pe rtine nt solicitations of arro gan t bigotry an d superstition . Then

presumably it would be on his view a mark, even a defining mark of this

elite to pro portio n th eir beliefs to the ir evidence ; and not to accept any

views concerning m att er s of fact on an yth ing else bu t evidence. A wise

m an , therefore, pro po rtion s his belief to th e evidence (p. 110).

I I

In Part II Hume calls upon some a posteriori assertions for an attempt

to establish certain conclusions of fact.

In the foregoing reasoning we have supposed, that the testimony,

upon which a miracle is founded, may possibly amount to an entire

proof, and that the falsehood of that testimony would be a real

prodigy. B ut it is easy to show, th a t we hav e been a great deal too

liberal in our concession, and that there never was a miraculouseven t estab lished on so full an evidence (p. 116).

1. (a) H e lists four sorts of considera tion. Of these th e first three are

straightfo rwa rdly em pirical. Althoug h controversial the y raise no special

herm eneutic problems. N um ber one is a categorical denial th at there has

ever in fact been a case in which various specified requirements have been

met and the evidence has been sufficient to justify belief. Num ber two

consists in a development of the theme of human credulity, starting with

a reference to a principle in hu m an na tur e which, if strictly examined, will

be found to diminish extremely the assurance, which we might, from human

testimon y, have, in an y kind of prodigy . This might be nicknamed th e

credo quia absurdum principle. Fo r whereas our usual maxim is, th a t the

objects, of which we have no experience, resemble those, of which we have ;

th a t wh at we ha ve found to be m ost usual is always mo st probable ;

nevertheless the hum an mind when any thing is affirmed utte rly absurd

and miraculous . . . the more readily admits of such a fact, upon account

of th a t ver y circum stance, which oug ht to destroy all its au tho rity (p. 117).

The thir d is the strong presum ption against all supe rnatura l and miraculous relations, that they are observed chiefly to abound among ignorant

and barbarous nations. . . It is strange, a judicious reader is apt to say

. . . that such prodigious events never happen in our days (pp. 119 f).

(b ) Th e fou rth reason is m ore novel, and of a different kind from th e

first three.

In m att er s of religion, wh ateve r is different is con trary . . . Ev ery

miracle, therefore, pretended to have been wrought in any of thesf'


h t t p : / / p q . ox




































6 ANTONY FLEW

religions . . . as its direct scope is to establish the particular system

to w hich it is att rib ut ed ; so ha s it the same force, thou gh m ore

indirectly to overthrow every other system . In destroying a rival

system, it likewise destroys the credit of those miracles, on which

th a t sy stem is establishe d ; so th a t a ll th e prodigies of differentreligions are to be regarded as contrary facts, and the evidences of

these prodigies, whether weak or strong, as opposite to each other

(pp. 121 f).

So we have :

a fourth reason . . . that there is no testimony for any, even those

which have not been expressly detected, that is not opposed by an

infinite nu m ber of witnesses ; so th a t n ot only the miracle destroy s

the credit of testimony, but the testimony destroys itself (p. 121).This argument bears a strong analogy to the original check of Part I,

and m ight well hav e found its place there instead of in P a rt I I . For whereas

there Hume was maintaining that all evidence for the subsistence of laws

of nature must weigh against any evidence for miraculous exceptions : here

he is urging that offset against any evidence for a miracle in any one religion

m ust be all evidence for all miracles alleged in all th e rest. This argu

m ent , he proceeds, m ay appe ar over subtile an d refined ; b u t it is no t

in reality different from the reasoning of a judge, who supposes that thecredit of two witnesses, maintaining a crime against anyone, is destroyed

by the testimony of two others, who affirm him to have been two hundred

leagues d is ta n t. . . (p. 122).

The illustration brings out ver y clearly w ha t he think s he has established.

Fa r from being over subtile the argu m ent is a blunderbuss. H um e's

premises are, explicitly an d generally, th a t in m at te rs of religion w hate ver

is different is con trary ; an d hence, par ticula rly bu t implicitly, th a t every

religious system is committed to denying that any miracles occur underauspices other th an its own. These premises requ ire, a t the very least,

considerable qualification. Y et even were we to gra nt them as the y sta nd ,

it wouid still not follow that the evidence for a miracle occurring at Lourdes

under Roman Catholic auspices' should be taken as weighing against any

evidence for miracles performed under Mohammedan sponsorship at Mecca.

W e mu st distinguish be tween : supposed m iracles considered simply as

loose and separate events ; and the same pu tativ e phenomena considered

as evidence for some system of religious theory.

Granting Hume's premises about exclusiveness, certainly it does follow

th a t in assessing the evidences for any such system we ough t to weigh ag ainst

these all evidence for an y miracles pre tend ed to ha ve been wro ugh t

under the auspices of all rival systems (E.H.U., p. 121). But even granting

these premises it certainly does not follow that one set of miraculous occur

rences considered simply as putative brute facts, must be in themselves

'For an examina t ion of the c ream of th i s ev idence see D. J . Wes t , Eleven LourdesMiracles (Duckworth, 1957) .


h t t p : / / p q . ox




































HUME'S CHECK 7

contrary to another set of such putative brute facts. In this second context

it is simply irrelevant that members of the two sets happen to share the

(different) relational characteristics of being sponsored by a (different)

religious system . Th e same man m ight, with catholic credulity, consistently

hold t h a t all th e miracle stories in L ivy as well as all those officially ratifiedby the Roman Catholic Church were true literally ; provided only that he

did not at the same time insist on believing in some religious theory which

denied th e historicity of an y of these tales. In H um e's apt il lustration two

pairs of witnesses are making incompatible assertions : one pair testifies

that the defendant was in one place at the time of the crime ; while the other

pair affirms th a t he was tw o hu nd red leagues away. W hereas in our case

one set of witnesses affirms prodigies in Republican Rome, while the other

testifies to miracles in th e Christian era. H um e has failed entirely to m ake

the distinction vital here : between the evidence for the occurrence of a

particular event; and the occurrence of a particular event as itself possible

evidence for the tru th of a theore tical system . H e th us becomes com m itted

to the preposterous proposition that the occurrence of an event which would

show some theoretical system to be false, must be taken to be incompatible

with the occurrence of th e pu tati ve events on which th a t system was

es tab lish ed (pp. 121 f).

(c) Though Hume's premises will not bear the conclusion which he tries

here to draw from them, they do nevertheless carry one unnoticed para,doxical implication : that if you could prove the occurrence of two and only

two strategically placed miracles you would thereby provide a knock-down

falsification of all po pu lar religious system s. Fo r you would need only

one miracle in one system to falsify all its rivals : and one more, in one of

thes e, to falsify th e first. Of course this arg um en t is of little help to H um e

himself : because he wants to argue that testimony in such cases is, as a

matter of straightforward empirical fact, peculiarly and hopelessly un

reliable ; and because, on his own misguided principles, he would have to

throw all testimony for the miracles of other religions into the balance

against that for each of the strategic two.

(d ) Though Hume's argument here will not do, there are some similar

poin ts to be m ade which are sound. Anyo ne who sets stock on an y sup

posed miracles as evidences for his religion must remember that rival religions

ha ve their miracle stories too : (and all of them abo und in miracles)

(p . 121). So even when a religious theor y is no t categorically com m itted

to denying outright the occurrence of all miracles not worked under its own

auspices ; it still cann ot call in an y (sort of) miracles in its favour, and henceby implication against its rivals ; unless it can also show that miracles (of

th a t sort) hav e not been wrough t und er any rival sponsorship. This is an

argument which would weigh heavily in support of the main conclusion

which in this section Hum e is concerned to establish : a miracle can never

be proved, so as to be the foundation of a system of religion (p. 127, italics

mine). Perh aps it was in fact this good argum ent, ra ther tha n the bad

one which Hume actually employed, which Kemp Smith had in mind when,


h t t p : / / p q . ox




































8 ANTONY FLEW

summ arizing this and other argum ents from P a rt II, he rem arked : Not

withsta nding H um e's own con trary claims really this is where the

stre ng th of his position lies. . .8.

2. Having completed the presentation of the fourth argument with the

illustration of a conflict of testimony before a criminal judge, Hume deploysthree cases where the evidence for the occurrence of miracles might seem

on the face of it very strong. He gives one pa rag rap h to each. Then in the

next he points in terms of this fourth argument a moral :

Suppose that the Caesarean and Pompeian factions had, each of

them , claimed the victory in th e ba ttles of Philippi an d Ph ar-

salia , and th at the historians of each p ar ty ha d uniformly ascribed

the advantage to their own side ; how could man kind, a t this distance,

have been able to determine between them ? The contrariety is

equally strong between the miracles related by Herodotus or Plu

tarch, and those delivered by Mariana, Bede, or any monkish his

torian (E.H.U.,?. 125).

Presumably Hume chose his three examples as being seemingly particularly

strongly evidenced, as coming from three different systems of religion, and

as being hence contr ary. The object of th e exercise is to illustra te a nd

reinforce this contention that all such evidence should be so to speak can

celled out.

(a) First : a rep ort in Ta citus of Vespasian, who cured a blind m an inAlexandria by means of his spittle, and a lame man by the mere touch of

his foot (p. 122). H um e rem ark s : no evidence can well be suppo sed

stronger for so gross an d palp able a falsehood (p. 123). B u t if we tu rn to

the original text—something which nobody seems ever to have done in

this connection—we discover that the two sufferers had asked for his help

monitu Serapidis dei ; quern dediia superstitionis gens ante alios co lit. The

Em peror w isely called for m edical adv ice : postremo aestimari a medicis

jubet an talis caecitas ac debilitas ope humana superabiles forent. The prog

nosis was not unfavourable : Medici varie disserere : huic non exesam vim

luminis et redituram si pellerentur obstantia ; ille elapsos in pravum artus

si salubris vis adhibeatur, posse integrari. So Vespasian decided th a t he had

nothing to lose and som ething to gain. Success was insta nta ne ou s : Statim

conversa ad usum mantis, ac caeco reluxit dies9. N ot being with ou t sin I

forbear to cast the first stone10.8N . K e m p S m i t h , Hume's Dialogues concerning Natural Religion (Nelson , 1947),

p .49 . Taylor , by cont ras t , a t t ack s th e a rgum ent on lines s imi la r to mine (pp . 338-9) .' T a c i t u s Histories IV , 81 . S ignif icant ly this reference is given wrong ly in every

edit ion of the E.H.U. wh ich I ha ve been able to cons ul t . In th e Fi rs t (1748) a t p . 192,H um e gives Lib . 4 . Cap . 8 , In the Second (1751) a lso a t p . 192, th is rem ain s unchan ged . In th e Third (1756) a fur ther er ror is intro du ced b y the readin g, a t p . 187,

L ib . 5 . Ca p . 8 . I n t he Bos to n a nd E d in bu r gh Works (1854), Vol. IV p. 139, wefind th e figures given as R o m a n V an d Ar abi c 8. Selby -Bigg e follows, at p . 122 ; an dYa lden -Th om son l ikewise , a t p . 127. H um e spea ks of th e second cure as bein g of alame m an : whereas Tac i tus h is s ta te d source has alius manum aeger. T h i s va r i a n tis found, in Sue toniu s, to wh om H um e a lso refers : Su etoniu s gives nea r ly the sa m ea c c oun t in vita Vesp. (E.H.U., p . 122 n) . H e wri tes alius debili crure, but he saysno th ing a bou t doc to r s (Lives of the Caesars, B k. V I I I , Ch . V I I , 2 ). This sor t o f th i ngwil l be a l l too famil iar to those who have made any s tudy of this sor t of thing.

Anyone in te res ted in the prac t ica l p rob lems of a ssess ing the ev idence in such

oases may refer to West , loc. cit.


h t t p : / / p q . ox




































HUM E'S CHECK 9

Second : a me m orable stor y rela ted by Cardinal de R etz . I t is to be

found in his Memoires ; though neither Hume nor his editors provide the

reference. At the C athed ral of Saragossa he was shown a m an who worked

a t lighting th e lam ps. The canons assured him th a t the y and everyone else

had seen him for seven years at the door of the same church with only oneleg. Je Vy via avec deux. . . . II avoit reconvert sa jambe, a ce qu'ils disoient,

en se frottant de I'huile de ses lampes. L'on cilibre tous les ans la f te de ce

miracle avec un concours incroyable11

. Since these dry phrases are all the

comment the Cardinal permits himself Hume is presumably right in thinking

th a t he canno t be suspected of any concurrence in the holy fraud . B ut

the thought-reading is here entirely speculative :

He considered justly, that it was not requisite, in order to reject

a fact of this nature, to be able accurately to disprove the testimony,and to trace its falsehood, through all the circumstances of knavery

and creduli ty which produced i t . . . He therefore concluded . . . .

that such an evidence carried falsehood on the very face of it . . .

(E.H.U.,-p. 124).

Third : th e stor y of th e numero us miracles which were lately said to

have been wrought in France upon the tomb of the Abbe Paris, the famous

Jans enist (p. 124). Hu m e emphasizes the unusual qu antity and quality

of the evidence in this case ; adding a long sardonic supporting note makingexplicit th e co ntras t in this respect with th e miracles of our Saviour .12

And wha t , he concludes in the text , ha ve we to oppose to such a cloud

of witnesses but the absolute impossibility or miraculous nature of the

events, which they relate ? (p. 125).

(b ) Three points here. First : H um e has not said tha t a ny sort of event,

or even a violation of the laws of nature, is logically impossible. Taylor

considers w ha t H um e is entitled, on his own principles, to mean when

he talks of the inconceivability of the violation of a uniform law of nature

(Taylor, pp . 348 f). H um e in fact uses no such expression.13 On the contrary

he speaks repeatedly of a miracle as a violation of the laws of nature ; while

nevertheless prepared to consider when we might be justified in believing

that such an act had in fact occurred.

H e does write of infallible experience , un alte rab le experience ,

and— as we have just seen— the absolute impossibility of miraculous

events. B ut the contexts make it clear th at these phrases can lend no

coun tenance to Tay lor's attr ibu tion . In the first case, H um e is contrasting^Oeuvres (Ed ited by Fei l le t an d Go urd ault , Par i s 1876) , Vol . IV , p . 550.1 2

P . 334. Th e m os t accessible , th ou gh ter t ia ry , acc ou nt of th e affair is R . A. K no x :Enthusiasm (O.U .P. , 1950) , Ch. X V I. Carr6 de M ontg eron 's book is VerM des miraclesdu diacre Paris (Vol. I , 1737 : I I , 1741 : I I I , 1748).

Ta yl or no t ices th a t H um e 's v iews fal l in to e leven success ive propos i t ions , whichI wil l s t a t e in order , w ith a m in im um of exeget ica l com m ent (p. 333) . In this sum m ary h e rem arks th a t b y h is appe a l to f acts [in P a r t I I : tho ugh of course Taylor take sno no t ice of th i s d iv is ion— AF ] H um e m us t tac i t ly sur render wh a t appe a red to havebee n secured b y th e app eal to ' inviolab le laws ' (ibid., p . 338) . W e m us t no te , as awa rnin g, th a t the expressio n inviolable laws which he encloses in inv er ted com ma sis no t in fac t to be found in H u m e' s essay . H e prov ides no reference .


h t t p : / / p q . ox




































10 ANTONY FLEW

such conclusions as are founded on an infallible experience , where th e

wise m an, proportioning his belief to the evidence expects th e eve nt

with the last degree of assurance (E.H.U., p. 110), with probable con

clusions, based on an opposition of expe rime nts an d observa tions .1 4

In the second we seem to have one of his too rare references to attempts

to falsify ca ndid ate laws of na tur e by expe rime nt : a firm and un altera ble

experience has established these laws .15 W hile in the third, in the

absolute impossibility or miraculous na tu re of th e even ts (p. 127), th e

impossibility he is thinking of is surely not logical. Ev en th oug h he p erh aps

might have been less fluent than we to analyze empirical impossibility in

terms of incompatibility with the laws of nature.

Second : there is internal evidence here against th e charge th a t Hu m e

only inserted this essay to gain notoriety . Ex ce pt on th e principle ' Squ arebullets for the infidel' why should we take as insincere his expressions of

exa spera ted dista ste for Our divines, who can build up a formidable

castle from such despicable m ate rials (p. 346) ? The charge of ba d faith

seems to rest on three prop s. Firs t, there are certain phrases in the a uto

biography.16 Mossner has rece ntly disposed of th e unso und an d u nch aritab le

interpretations of these.17 Second, it is urged th a t the passages which hav e

given such offence cannot be explained as expressions of undisciplined

fanaticism : because Hu me was . . . no anti-clerical zealot, bu t an amiableand easy-going man of the world whose chosen circle consisted largely of

the 'm od er at es ' among the Edinburgh Pres by ter ian s (Taylor, p . 331).

To this Greig replied sharply to th e poin t th a t H um e ha d no dealings we

know of with this circle before at earliest 1752 ; 18 while we ha ve plen ty of

biographical evidence of anti-clericalism persisting to his death.1 9

Third,

for the purposes of the Enquiry , Tay lor claims, the whole section is

superfluous. . . T he irrelevance is . . . manifest (Taylor, p . 331). Selby-

Bigge labelled both § X I and § X quite superfluous (E.H.V., p . viii).

It is hard to get to grips with this charge, since Hume's accusers vouchsafe

no explanation of what they take his main theme and purposes to have

been. B ut if we tur n to his own account in the introduc tory § I, rem emb er

the original title of the book, and notice that and how the arguments of

§ X and § X I are complem entary,20 it is surely clear that the charge is pre

posterous. Atte m pts to show th a t miracles cannot be proved s o as to be

th e founda tion of a system of religion and th a t th e religious hyp othes is1 4

P . I l l : cf. p . 56 n, on th e dif ference be twe en prob abil i t ies an d emp ir ica l proo fs .1 5 P . 114. C om par e Section B of m y ' Ca n an Effect pre ced e its Cause ? ' (P.A.S.

S u p p . V o l . X X V I I I ) .

See Kemp Smi th , pp . 233-240 .1 7

E . C. Mossner : ' Ph i loso phy an d B iog rap hy : th e case of Dav id H um e ' (Philosophical Review, 1950). W e can no w refer also to his spl en did Life of David Hume(Nelson, 1954).

1 SJ . Y. T. Greig, David Hume (Cape, 1931), p . 163.

See Mossner Life, pass im ; bu t e spec ia lly pe rh aps the s to ry of an ou t bu rs t aga ins tthe doctr ines of hel l and damnation, a t p . 570.

20 See II (3) (c) below.


h t t p : / / p q . ox




































HUME'S CHECK 11

is explanatorily idle are very much in place in an enquiry into the nature

and l imits of human understanding.21

Third : m an y of th e phen om ena which Hu m e dismisses so confidently

probably in fact occurred.22

His ras h assurance has left him wide open to

H am let 's too often quoted rebuke to dogmatic philosophy. B ut this ofcourse does not mean : either that Hume's principles of historical criticism

m ust ha ve been wrong ; or th a t th e m iraculous cha racte r of th e events in

question has been dem onstrab ly established. His principles m ay have been

sound although they led him to incorrect conclusions sometimes. 23 While

some of the conclusions must fall under suspicion : not because it seems now

th at there m ay hav e been a violation of the laws of natu re after al l;

but because we have now reason to believe that some of the laws of nature

are not as Hume thought.2 4

3. Having presented his three illustrations Hume sums up in one para

graph their relevance to the argument examined in II (1) (b) and (c). Nex t

in six short parag raph s he return s to two m atte rs b roached earlier : the

peculiar and extraordinary evils which in his view afflict testimony in re

ligious cases ; and the peculiar and extraordinary difficulties of discovering

th e tru th ab ou t the infancy of a religion. H e now reviews th e upsho t of

th e whole argu m ent : n o testim ony for any kind of miracle has ever

am oun ted to a prob ability ; and we m ay establish it as a max im tha tno human testimony can have such force as to prove a miracle, and make it

a just foundation for any such [viz., ' pop ular '— A.P.] system of religion

(E.H.U., p . 127). T ha t last clause is, he emp hasizes, im po rta nt : a miracle

can never be proved, so as to be th e foun datio n of a system of religion. . .

For . . . there may possibly be miracles, or violations of the usual course

of nature, of such a kind as to admit of proof from human testimony (Ibid.,

p . 127 ; italics mine).

(a ) He proceeds to illustra te. Suppose trad ition s : first of tota l world

wide dark ness for eight day s from J u n e 1st 1600 ; ne xt of th e dea th an d

resurre ction of Queen Eliza beth . The first bu t no t the second m ight be

supported by evidence sufficient to justify belief. Tay lor com men ts : I

confess I cannot see on what ground Hume makes any distinction between

th e two cases he has, with notab le bad taste , been pleased to imagine .25

Y et H um e gives one grou nd quite explicitly : The decay, corruption, and

dissolution of nature, is an event rendered probable by so many analogies,C om pa re : To enqui re se r iously in to the na tur e of the hu m an un ders tan ding ,

and to show, f rom an exact analysis of i ts powers and capaci ty , tha t i t is by no meansf it ted to such rem ote and abst rus e sub jects (E.H.U., p . 12).

Se e tex tbo ok s of abn orm al psychology an d psych osom at ic medic ine : the rangeof possible psychological ef fec ts far exceeds what was a t one t ime thought possible( W e s t , loc. cit., p. 17).

2 8F . H . Brad ley dep loys four ve ry ap t examples in No te D to ' The P resupp os i t ions

of Cr i t ica l History ' in Collected Papers (Oxford, 1935), Vo l. I , pp . 63-4.

See I I (3) (b) below.

T a y l o r , loc. cit., p . 340. H e re t ur ns to the charge a t p . 344 an d p . 350.


h t t p : / / p q . ox




































12 ANTONY FLEW

that any phenomenon, which seems to have a tendency towards that

catastrophe, comes within the reach of human testimony. . . {E.H.U.,

p. 128). Hum e is tak ing a possible failure of th e light of th e sun, the m oon,

and the stars as one more phenomenon of corruption : perhaps he had in

mind the idea of a Cosmic Fall, much discussed in the previous century. 26

Contrariwise a resurrection presumably is thought of as out of line with

th e otherwise universal m ort ality of thing s. Pe rha ps Hum e was influenced

also by two further bu t related facts. W e ha ve come, the w orld being as

it is, so to use th e w ord ' de ath ' th a t i t is logically impossible to survive

death.27 Some Christian apologists, pa rtly pe rhap s for these reasons, ha ve

urged resurrection as a miracle of miracles.

(b ) Y et som ething here does cry out for com m ent a nd does seem a

volte-face. For Hume allows that in the first of his supposed cases, andgran ting certain unfulfilled conditions, our prese nt philosophers, instea d

of doubting the fact, ought to receive it as certain, and ought to search for

the causes whence it might be derived (E.H.U., p . 128, italics m ine). Now

he is supposed to be illustrating the conditions needed to justify the wise

and learned, who proportion their beliefs to evidence, in believing that a

miracle really had occurred. B ut anyone who adm its th at some event ha s

occurred and yet insists on proceeding to search for its causes, thereby

reveals th a t he does not tak e it to be genuinely m iraculous. For a genuine lymiraculous event, as a violation of the laws of nature, would have no causes,

or at any rate, no natural causes.

Perhaps it is significant that in this particular paragraph Hume speaks

of miracles only as violations of th e usual course of na ture whereas in

the one before, th e one after, and elsewhere he writes th e laws of nature

(p. 128 f, and passim ; italics m ine). I t looks as if he is here b eginning t o

abandon his characteristic position in favour of a robust uncomplicated

insistence that really a miracle as a violation of a law of nature must belogically impossible ; since authentic laws cannot have any genuine excep

tions. W e cann ot , as Mill has it, ad m it a proposition as a law of na tur e,

and ye t believe a fact in real contradic tion to it. W e m us t disbelieve th e

alleged fact, or believe that we are mistaken in admitting the supposed

law ,28

Apparently, thought not quite certainly, it was this position which Mill

himself an d Leslie Steph en w ere tak ing to be charac teristic. Mill refers

earlier in th e same section to H um e's celebrated doctrine , th a t noth ing is

credible which is contradictory to experience, or at variance with the laws

of na tu re (Mill, loc. cit., ibid). While Stephen epitomises Hu m e's contri

bution to the deba te in his time : H um e replied th a t no evidence can provea'Cf. K. W . He pb ur n : ' Godfrey Go od m an : N at ur e Vili fied ' {Cambridge Journal,

1954).

Se e m y ' Can a M an W itness h is own F un era l ? ' in Hibbert Journal, Vol . L IV(1956).

aM System of Logic, Bk . I l l , Ch . X X V , § 2 . F o r a r e c e n t s t a t e m e n t o f t he s a m e

posi t ion developed see P . H. Nowell-Smith 's ' Miracles ' in New Essays in Philosophical

Theology (Edi ted F lew and Mac ln tyre , S .C.M. P ress , 1955) .


h t t p : / / p q . ox




































HUME'S CHECK 13

a miracle .29

These inte rpr eta tion s w ill no t do : for two reasons.

First : because th ey ignore his careful concession th at the re m ight be

circumstances in which the evidence was sufficient to justify belief in the

occurrence of a miraculous ev ent. This is a poin t which he is a t pains to

m ake twice : once in the concluding para grap h of Pa rt I ; and again inP a rt I I in th e para gra ph und er discussion. On the second passage : if in

m aking th is concession Hu m e was fully aw are th a t his insistence on launch ing

imm ediately a search for the causes m ust m ake th e exception vacuous,

th en his em pha tic qualification can neve r be prove d so as to be the founda

tion of a system of religion becomes po intless (E.H.U., p. 127 ; italics mine).

While the various arguments peculiarly directed to that special case become

otiose. On th e first passage : to tak e this with ou t any positive reason as

an empty rhetorical flourish would be unscholarly and arbitrary.

Second : such inte rpr eta tion s m iss th e po int of the crucial conception

of th e opposition of proofs. There m us t be a uniform experience against

every miraculous event, otherwise the event would not merit that appel

latio n . So here we ha ve a direct an d full proof, from the n atu re of the

fact, again st th e existence of an y miracle ; nor can such a proof be destroyed ,

or the miracle rendered credible, but by an opposite proof which is superior

(pp. 115, 127). Since H um e has earlier defined proofs as such argu m ents

from experience as leave no room for do ub t or oppo sition (p. 56), it is

at least difficult to understand the suggestion that there could be a proofof any proposition if there was also a proof of a second and incompatible

proposition, unless the remark is meant as a mere rhetorical amplification

of th e state m en t th a t proof of th e ' miraculous ' is no t forthcoming (Taylor,

p . 339).

Difficult, perh aps , bu t no t surely impossible. Certainly H um e's accou nt

of the situ ation is para dox ical. B ut th en in his view we face an antin om y :

the possibility of a head-on collision between two kinds of proof. One is

historical : proof, based on testimony, of the occurrence of some particular

eve nt. The othe r is of laws of na tu re : proof, based on an appe al to a

firm and unalterable experience (E.H.U., p. 144) of general regularities.

B ot h are e ssential in a full case for th e occu rrence of a miracle ; for the con

cept of the miraculous combines the two ideas of general law and particular

exception. Suppose the n, he argues, in some instance the testimon y

considered apart and in itself, amounts to an entire proof; in that case

the re is proof again st proof . . . (p. 114).

Earlier in Part I he has contended that both sorts of proof appeal ulti

m ate ly to experience alone, and are really founded on th e relation ofcause and effect (p. 111). Re m em bering w hat he has said already abo ut

the historian's inescapable need to rely in his critical assessment of his

evidence on wh at he knows of the uniform ity in hu m an actions ,30 we

can now see why Hum e allows one exception to his m axim no testim ony

is sufficient to establish a miracle (pp. 115 f). He can no t insist th a t the

''History of English Thought in the Eighteenth Century (Third Ed it io n) , Vol . I , p . 271 .3 0

In P a r t I of § V I I I ; ' Of L ib e r ty an d Necess i ty ' : com pare Brad ley loc. cit., passim.


h t t p : / / p q . ox




































14 ANTONY FLEW

proof of a law of nature must always be stronger than any proof of an ex

ception to it. Fo r it is at least theore tically possible th a t the testim on y

might be so strong that to insist on the universal validity of one law of

nature (e.g. some biological law about the conditions for restoration of

tissues) would inevitably involve abandoning the idea of the universalvalidity of another (e.g. some psychological law covering the behaviour of

witnesses).

This subtle position Mill m isrepresents as crude comm on sense : H um e's

celebrated doctine, th a t nothing is credible w hich is con tradic tory to

experience or at variance with the laws of nature, is merely this very plain

and harmless proposition, that whatever is contradiction to a complete

indu ction is incredible (Mill, ibid). H um e's actua l view is unsta ble as

well as subtle. I t has a na tur al tenden cy to develop into the position towards which in this parag rap h he is sliding. Fo r his idea of th e opposition

of proofs is mediated by the concept miracle—it is, in his material mode of

thought, de rive d from the very nature of the f a c t (E.H.V., p. 127).

Yet pressed only a little further his analysis of this concept, by itself and

without benefit of the other idea, entails that miracles as such must be

logically impossible. B ut to tak e this ultra -sho rt way as ch aracte ristically

his is to fail to do justice : either to the actual course of the argument in

the essay ; or to its place in a developing controversy.

(c) This last point comes out well in Stephen.31

H um e was writing fora public which might have read Zac hary P earce's M iracles of Jesus Vindicated

(1729) and which had called for edition after edition of Thomas Sherlock's

Trial of the Witnesses (1729), an early and unsophisticated exercise in the

genre of Frank Morison's Who moved the Stone ? (Appleton-C entury, 1930 :

Fa ber , 1931). This, toge ther w ith th e intere sts which were to earn him his

repu tation as Mr. Hu m e the historian , explains why he approaches the

problem from the historical rather than the scientific angle. 32 I t is set by

the accounts of miracles and prodigies . . . found in all history, sacred

and profane (E.H.U., p. 110). This, toge ther w ith the theological intere sts

which led to The Natural History of Religion and the Dialogues, and which

are already obvious in the E.H.U.,33

explains why he keeps harping on theal

Loc. cit., Vol. I , Ch. IV, especia l ly § V : ' Th e Ar gu m en t f rom Miracles ' .a 2To br ing ou t th i s cont ras t compare T . H. Huxley ' s c r i t ique in h is Hume (Mac-

mi l lan , 1887), Ch . V I I . H e ho lds : the e ssen t ia l a rg um ent o f H um e 's f amous d isquisi t ion up on miracles . . . m ay safe ly be declared to be i r ref ragable (p. 130). B u the mistakes i t to be an appeal to the evidence for regular i ty in nature , just i fying areques t to those who would hav e us p u t f a i th in the ac tua l occurrence of in te r rup t ion sof th a t o rde r , to p roduce ev idence . . . supe r ior . . . to th a t which leads us to be l ieve

in i t (pp. 129-130). W ha t rea l ly w as cha racte r is t ic of an d crucia l for H u m e he rega rdsas shell of ve ry dou btful va lu e surr oun din g thi s kern el of his essay (p. 130).Ye t a f te rwards he concedes th a t H um e 's a rg um ent s hav e a ve ry d i ffe ren t va lue . . .when we tu rn f rom the ques t ion of the poss ib i l i ty . . . to th a t r e spec t ing the gro und sup on which we are just i fied in bel ieving an y par t icu lar mira cle (p. 134). In H ux le y ' sHamlet the Pr ince of Denmark enters only for the f inal Act .

M os t c onsp ic uously i n § V I I I P a r t I I , § X , § X I , a n d § X I I P a r t I I I .Convinced that this whole essay was a s ly far rago cooked up as one more move in al i f e long publ ic i ty hunt , Taylor takes no account o f th i s au then t ic ab id ing in te res t(pp . 331-2 and passim). B . M. Lain g lapses s imilar ly in his ' H um e' s Dia log ues ', thou ghnot in his Hume (Benn, 1932) . I n th e former he wri t es : th e Es say on Miracles isessent ia l ly a discussion on evidence , re l igious issues being employed as a medium for

detached examinat ion of logical issues {Philosophy 1937, p . 180).


h t t p : / / p q . ox




































HUME'S CHECK 15

them e of resurre ction. I t is no t wa ntonly notab le bad tas te (Taylor,

p. 340) bu t ra th er a response in controv ersy to a m ovem ent of concen

tration recently performed by the orthodox pa rty ; in deciding The

fate of Ch ristianity mig ht be stak ed on th e proof of th e resurrection ,34

The same neglected historical context explains and justifies the definition

to which Hu xley an d Tay lor both tak e different exc eptions. H um e's check

is officially defensive. A miracle , he claims, m ay be accu rate ly defined,

a transgression of a law of nature by a particular volition of the Deity . . .

(E.H.U., p. 115 n). H ux ley 's objections are m ainly on the lines suggested

by Mill.35

B ut he concedes th e definition to be in accordance with a com

mon practice of believers , 3 6 without realizing the importance of this.

The appropriate definition is the one appropriate to the controversy to

which th e essay belongs. Taylor first contrive s to detect thr ee differentsenses or definitions in H um e (Tay lor, pp. 333 f). The n later, im puting as

usual bad faith, he undertakes to eliminate an ambiguity, supposedly

crucial : between miracle simply an unusu al and arresting event ; and

miracle an event, not necessarily particular unusual, which is held to

disclose, as most events do not, the direct ac tiv ity of God (p. 342). Ne ither,

surely, of these suggested senses is strong enough to have been acceptable

to either H um e or his contem porary oppon ents. Hu m e in P ar t I makes a

poin t of con trasting th e miraculous with the m erely unu sual. W hile, inthe very passage of Thomas to which Taylor refers, the saint insists that

even in the commonest sort of miracle God must work absque principiis

naturae operantibus.37

Again, Taylor would have it that discussion of miracles in his second,

theological, sense is a piece of irrelevance (p. 343). Yet the most em pha

sized conclusion, an d hence one migh t thin k th e ma in object, of § X is the

denial, which flatly contradicts the contention on which the orthodox

apologists ha d con cen trated, th a t miracles can serve as th e foundationof a system of religion (E.H.U., p . 127). W hile in § X I his sceptical friend

would establish the com plem entary thesis th a t : No new fact can ever be

inferred from th e religious hypo thesis (theism ). Taylor will allow, w ithou t

ever recognizing its kinship with H um e's m ain point, th a t : The alleged

occurrence of miracles cannot itself be rationally made a premise for the

arg um en t for Theism (p. 362). B ut he would suggest th a t if the miracles

cannot support the system, the system m ay support the miracles : A

theist . . . will thus reasonably regard it as to be expected that surprises

S t e p h e n , loc. cit., Vol. I , p. 238.

See I I (3) (b) above.

H ux le y , pp . 130-4 ; quo te a t p . 130. Com pare de fin i tions quoted by L a ing a ndKemp Smi th f rom Tinda l and Cla rke (loc. cit., p . 134 an d p. 48 n. respect ively) . AlsoB a y l e Dictionnaire (1730), Vol . IV , p . 264 n. : Les Car tesiens . . . sup osen t que qu andil fait de s m iracle s il n 'o bs erv e po in t les Lo ix gene rales qu 'il a etab lies ; i l y fait u neexc ept ion . . . .

3,Summa contra Gentiles, Bk . I l l , Ch . 101 . Com pare : I l i a ig i tu r p ropr ie m iracu la

d i c e nda sun t qua e d iv in i t u s S un t p r a e t e r o r d ine m c ommun i t e r obse r va tum in r e bus .


h t t p : / / p q . ox




































16 ANTONY FLEW

of a certa in kind . . . should occur in his tory (p. 360). H um e gives his

short answer here, but develops it in the next and complementary essay.38

4. In the next two paragraphs Hume, commenting on the resurrection,

argues : should this miracle [viz. th e Eliz abe than resurrection] be ascribed

to any new system of religion . . . this very circumstance would be a full

proof of a cheat. , . (E.H.U., p . 129). Taylor com plains : it is ha rd to

understand why testimony . . . should lose . . . value merely because the

belief in the event had led ex post facto to th e appea rance of a ' new system

of religion ' (p. 340). B ut it is no t in th e least ha rd to un de rsta nd wh at

Hu m e's position is. He has already argued at length th at testim ony in

such cases should be peculiarly suspect, though admittedly he takes no

account of the possibility Taylor apparently is imagining, where the testi

m ony for the m iracles is prio r to a nd independ en t of th e religious beliefs.He sums up again in this same paragraph, citing in the next the authority

of Fran cis Baco n in sup po rt of a general resolution, neve r to lend an y

attention to it . . . (E.H.U., p. 129).

5. Finally we come to the last two paragraphs, containing the notorious

conclusion : th a t the Christian Religion not only was at first attended

with miracles, but even at this day cannot be believed by any reasonable

person wi thout one (E.H.U., p . 131). Taylor calls this a volte-face

(p. 341), an d refers to the c urre nt view th a t it is a piece of m ere m ock ery(p . 342). Y et it is no t quite th e volte-face he tak es it to be. Ha ving ignored

the emphasis in Hu m e's exordium (p. 342) on the concern with evidence,

Taylor inevitably sees the peroration as in sharper contrast than it really

is . Hu m e is not here m aking any last minu te withdrawals abou t the his

torical evidence. I t is precisely from its ut te r inad equ acy , according to

the measures of probabil i ty above established (E.H.U., p . 130), th a t he

argues to his ad m itted ly para dox ical mo cking conclusion : Mere reason

is insufficient to convince us . . . whoever is moved by Faith to assent . . .is conscious of a continued miracle in his own person, which subverts all

the principles of his understanding, and gives him a determination to believe

what is most contrary to custom and experience (E.H.U., p. 131). This

though ironical is surely straightforward enough : especially when we recall,

as Ke m p Smith reminds us, w hat, in Hu m e's day, was the declared teaching

of the Reformed C hurches ab ou t the miraculous origins of faith and th e

workings of grace.39

' * Tho ugh th e Be ing to wh om th e mirac le i s a sc r ibed , be . . . Alm ighty , i t doesno t , up on th at acc ou nt , become a w hit m ore pro bab le ; s ince i t is impo ssible for u sto know the a t t r ibu tes o r ac t ions of such a Be ing , o the rwise than f rom the exper iencetha t we have of h i s p roduc t ions , in , the usua l course of na ture (E.H.U., p . 129).This is th e only b u t sufficient pass age l inking the arg um en t of § X t o th at of § X I .

a 9 Ke m p S m ith, p . 47 : cf. Ch. I and I I passim. Com pare : Conv er t s know th a tthe i r inne rmost hea r t s a re changed and tha t they have wi th in them a wi tness whichha s no hum a n i s t i c e xp la na t ion . A n d the r e is no e xp la na t ion e xc e p t a su pe r n a tu r a lone (Mr. K en ne th de Cou rcy in th e Spectator, 17/5/57) ; an d the even mo re rad icalv iew tha t theo log ica l u t te rance acqui res sense on ly by a mirac le , found in Kar l Bar th ' sProlegomena to Church Dogmatics.

Taylor nevertheless develops a more tortuous and less creditable inter-


h t t p : / / p q . ox





































HUME'S CHECK 17

preta tion. H e take s Hum e to be making a concession required by his theory

of belief and analysis of causal connection ; and sees Hume's irony as aimed

also a t ' dangero us friends or disguised enem ies ' to New tonian science

(p. 353). Fo r this second suggestion he can offer no direc t evidence w hat

soever. By a judo m ove he uses his ver y weakness to confound the enem y :The reason must be that notoriety was to be got by an attack on the

Church ; an at ta ck on th e Eo ya l Society would pass unreg arded (p. 355).

In su pp ort of the first suggestion he contends th a t before we can get at

the real meaning of his argument, we have to translate its terms into the

languag e of H um ea n scepticism (p. 345). This done, the upsh ot becomes :

Properly speaking, there are no laws of nature to be violated, but there

are habits of expectation which any one of us, as a fact, finds himself unable

to brea k thro ug h (p. 349). Since, man ifestly, some do continue to believeth e incredible H um e has to m ake th e concession embo died in such pro

voc ative langua ge (p. 352).

Now no doubt this is something which Hume might have meant, and

which perha ps he ought to have m ean t. Taylor is not, however, entitled to

say this was in fact wh at Hu m e did mean and m ust have mea nt. It is im

proper to draw heavily for ammunition on the Treatise, without noticing

H um e's com plaints again st a prac tice very contra ry to all rules of candour

and fair dealing (E.H.U., p . 2). I t is radically un sound to argue : from(your view of) wh at H um e is entitled, on his own principles, to mean

(Taylor, p. 348) ; to what in fact he did mean. 40

When we turn to what Hume actually says we find the case by no means

open-and-shut. Ke m p Sm ith notes th a t law always appears in the plural

as pa rt of th e laws of na tu re ; except in the acc ura te definition. So by

their violation H um e inten ds : to signify solely wh at is con trary to the

common course of nature, i.e. in theological phraseology, the supernatural

(Ke m p Sm ith, p . 48). Insofar as this is to eq ua te the laws of na tur e

with th e usua l course of na tu re the re are difficulties. This interp reta

tion fits well th e noto rious question-begging : it is a miracle, th a t a dead

man should come to life ; because that has never been observed in any age

or country (E.H.U.. p. 115). And with the comm endation of M. H era ut

for reasoning like a m an of sense, from na tu ra l causes (p. 345). B ut it

nullifies th e distinction elab orate d between w ha t is only marvellous and

w ha t is really m iraculous (p. 114). I t take s all force from repe ated appeals

to th e ve ry na tur e of th e fact . And it tak es no accou nt of the change

of phrasing, when Hume is considering the admission of a genuine excep

tion.41 Precisely the same difficulties confront T aylo r's inter pre tation .42

Additionally he has the problem of accounting for Hume's insistence that

there might be real but undetectable miracles (E.H.U., p . 115 n). Ye t we

*°See th e excel lent discussion of th is and s im ilar m ist ak en pr incip les of int erp re ta t io nin R . Rob inson Plato's Earlier Dialectic (Second E d . O.U .P. , 1953), Ch. I .

See I I (3) (6) above.

W hi c h w a s a ppa r e n t ly a l so t h a t o f T . H . G r e e n : s ee Works, Vo l. I , p . 276 n.


h t t p : / / p q . ox




































18 ANTONY FLEW

have already found sufficient reason, quite apart from the account of why

a resurrection would be m iraculous, to say th a t laws of na tur e is no t

taken consistently and clearly in the modern sense.43 It might be suggested

that Hume was really quite clear in his own mind about what he himself

was entitled to mean by laws of na tur e ; bu t th at (apparent) inconsisten

cies arose in arguing ad homines against people whose position depended on

giving to it a sense, or pseudo-sense, stronger th an this . Ce rtainly it is

important to remember that his argument is officially defensive.44 But

earlier in the E.H.U. he defines a cause :

an object, followed by another, and where all the objects similar

to the first are followed by objects similar to the second or, in other words,

where, if the first object had not been, the second never had existed

(E.H.V., p. 76).The unnoticed gulf between the first definition and the second is both the

measure of the inadequacy of Hume's analysis and an indication that he

did not appreciate how much of the idea of lawful connection he was by

it committed to abandon.45

A N T O N Y F L E W

University College of North Staffordshire.

See I I (3) (6) above.

See I (3) (a) and II (3) (c) above.45 See Flew loc. cit., and ' Causa l Disorde r Aga in ' (Analysis, Vo l. 17).


h t t p : / / p q . ox


































flew-the philosophical quarterly-1959-.pdf

Documents