:preta-
lims 
I draw 
to do 
i we are 
look at 

ductive 
,d 

sup-
argu-
the 

urn in-
which, 
two 
to me 
would 
ion. 
n 
r argu-
Y would 

n's ac-
tween 
general 
s re-
Jning 
!!2! one 
~ocess 

fts 
Ilat is 
Jf 
an 
~d as 
cru-
~d to 
in 
lrly 

some 
be 

! con-
thing 
this 

Jht of 
Its 
lnguage. 
lis-
by 
)gic 

: News-

ten-
ling 
Lking 
ltext, 
the 

fferent 

~ con-
:I.e 
?hilos-
~rs to 
r would 
may 
~ciples 
e11-
Llman 
sci-
quite 

Jral 

philosophy. (Note: this is not a fal-
lacious appeal to authority, but merely a 
point of information!) 

4. Adapted from an argument in the Toronto 
Globe and Mail, appearing sometime in 
JanuarY,""""l'98O; 

5. That is, unless we insist on adding as a 
missing premise, 'You should do every-
thing you promise to do', in which case 
the arqument is deductive. 

6. See Scriven, Reasoning, pages 78-81, and 
Thomas, Practical Reasoning in Natural 
Language, pages 37-40. 

7. He does note that the concept of conduc-
tive argument has wider application, say-
ing, "\iherever some descriptive predicate 
is ascribed on the basis of a family re-
semblance, conductive reasoning takes 
place." (page 54). But he does not make 
enough of this point, nor does he tell us 
how many descriptive predicates ~ as-
cribed on the basis of family resemblance. 
(If most are, as Wittgenstein seems to 
have thought, then conductive reasoning 
would be very common, and conductive ar-
guments an extremely important type. 
Compare D. C. Yalden-Thomson, "The 
Virginia Lectures", in Wisdom: Twelve 
Essays, edited by Renford Bambrough.) 

8. Wellman specifies three different patterns 
of conductive argument; one where only 
one relevant factor is adduced; one where 
several are adduced; and a third where 
both 'pro' and 'con' factors are adduced, 
and the claim is made that the former out-
weigh the latter. 

9. Compare Susan Haack, Philosophy of Logics, 
Chapter II. 

10. See Objections to the Meditations (II), 
and Descartes' reply, in which he insists 
that a general proposition is not re-
quired and that, indeed, "general pro-
positions are formed out of the knowledge 
of particulars" (page 38, Volume II, in 
Haldane and Ross edition.) A similar 
response is made by Descartes in a letter 
to Clerselier (page 127 in Haldane and 
Ross, II.) 

11. Compare Francis Dauer, "The Diagnosis of 
an Argument", Hetaphilosophy, 1974. 

15 

UNDERSTANDING 
SCIENTIFIC 
REASONING 
Ronald N. Giere 

William Berriman 
University of Regina 

Suppose one could gather together about a 
hundred university professors of philosophy 
and suppose furthermore that they represented 
a fair sample of today's philosophers with 
respect to contemporary thinking about the 
teaching of logic and philosophy. To these 
people one now reads the first four para-
graphs of Giere's Preface. 

By the end of the reading the philosophers 
would have become decisively divided on their 
attitudes to the book. Giere speaks of "new 
courses in 'applied philosophy''', of the 
"students' demand for 'relevance'" and of his 
reluctant rejection of his idea of subtitling 
the book "A Consumer's Guide to Scientific 
Method". He claims that students can become 
intelligent consumers of scientific informa-
tion not by learn~ng science (or in his words 
"how to be scientists") but by learning 

merely how to read and interpret 
reports of scientific findings, 
espec~ally reports that appear in 
the popular magazines, newspapers, 
and even supermarket tabloids. 

Some of our philosophers would denounce the 
book as "light weight", "not academically 
respectable" and condemn the book outright. 
The others would find that the book arouses 
their interest. This aroused interest may 
have three (possibly overlapping) sources: 
a) an awareness of the possibility that the 
book may form the basis of an attractive 
(pop?) class. b) On the basis of one's know-
ledge of today's students plus the nature of 
many media reports one may have concluded i) 
that most people know very little about 
science but that ii) they are prepared to 
believe the most outrageous stories provided 
these are in the realm of "science". Finally 
c) Giere's book appears to offer the oppor-
tunity to extend one's work in informal logic 
into the area of the logic of science, and 
reports about science in the media. 

A cursory glance through the book reveals 
material that should arouse student interest. 
Giere's chapters deal with Haley's long-
range forecast of the appearance of his comet, 
the World II model which uses a computer to 
predict disaster for mankind by the year 2100 
unless great changes are made in the world 
socio-economic system by 1985. Additional 
(to informal logic) fallacies such as Vague 
Predictions Cor the Delphi Fallacy), The Jean 
Dixon Fallacy (multiple predictions), 



Patchwork Quilt Fallacy (and the Chariots of 
the Gods) are also introduced as well as 
mater~aI on probability, causation, testing 
statistical hypotheses and so on, The exer-
cises at the end of each chapter also provide 
a plentiful array of interesting material 
which can be easily added to (almost on a 
daily basis) from the mass media. 

If one begins to read the early chapters 
one comes to see that Giere's exposition is 
lucid and his frequent reassurances that 
really this is all quite easy satisfies one 
that this is indeed a useful text. 

The problems only begin when the book and 
students are brought together, In spite of 
Giere's assurances that it is easy, the stu-
dents encounter problems because they find 
the book difficult. 

I think there are at least two basic rea-
sons why the students find the book difficult 
and neither reason is unique to Giere's book. 
The first problem arises from ~he fact that 
authors of "easy" texts in philosophy are 
conscious that at some places at least, they 
are dealing with issues which are most con-
troversial. There is a need for the author 
to show that he is aware of these difficulties 
even if he cannot fully explain them. So he 
mentions the difficulty and the fact that 
controversy exists, states his position, then 
proceeds. 

For example, in Chapter 5, using Newton's 
Theories as one of his examples of a scien-
tific theory, Giere states that the question 
"What is a theory?" is a much discussed ques-
tion. After stating Newton's four laws using 
the terms "force" and "mass", he explains 
that those terms are not clearly defined. 
However, (he says), one can take the four 
laws as definitions of mass and force. Indeed, 
the concept of theories as definitions accounts 
for their protected epistemic and logical 
status. But definitions, while they are true, 
say nothing about the world, so the question 
is how can theories apply to the world? 
Giere's answer is that they do not, but that 
they are very closely related to theoretical 
hypotheses, which are contingent and about 
the world. 

The connection between theories and theo-
retical hypothesis is made in Chapter 6. 
Giere there asserts that "in general, for any 
theory, a theoretical hypothesis has the form: 

such and such real system is a system 
of the type defined by the theory," 
Cpo 70) 

Theoretical hypotheses are thus contingent 
statements serving to link theories as defi-
nitions to the world. 

I'1hen Giere shifts his attention to the 
crucial matter of justifying theoretical hy-
potheses, he introduces two conditions which 
are later CChapter 8) used to distinguish 
between acceptable scientific thinking and 
unacceptable (pseudo) scientific thinking. 
The two conditions are: 

1. If IH and IC and AA], then P. 

2. If [not Hand IC and AA], then 
very probably not P. CH = the 
theoretical hypothesis, IC ~ 
initial conditions, AA - auxiliary 
assumptions and P • the prediction}. 

16 

In Condition 1 the prediction is thought of 
as being a deductive product of the anteced-
ent with the auxiliary assumptions seen as a 
"catch all" to secure this deductive rela-
tionship. Condition 2 is meant to eliminate 
any obvious "predictions" and to ensure that 
the hypothesis is not "empty" but really does 
add to the conjunction of IC and AA. If the 
predicted event does not occur it is shown 
~at on deductive grounds the hypothesis is 
falsified (provided both IC and AA are true), 
but if the predicted event comes to pass only 
inductive support is given the hypothesis. 

After understanding these conditions and 
their logical support clearly, the student 
should be able to use them to distinguish 
between satisfactory and unsatisfactory sci-
entic claims as reported in the mass media. 
However, additional logical insight is re-
quired to understand that the so-called Jean 
Dixon Fallacy using multiple predictions can 
be regarded as either a conjunction or dis-
junction of predictions, If it is a con-
unction then one false prediction falsifies 
the set and if it is a disjunction then just 
one correct prediction (which may be made by 
any reasonably informed person) makes the 
total set true, Taking the set as a series 
of conjunctions is too rigorous given her 
success rate but taken as a disjunction does 
not provide evidence that she has what Giere 
calls "a futureseeing system". 

Enough has been said to make it ev:.dent 
what the second basic reason why students 
find the book difficult. For all Giere's 
assurances to the contrary, the subject mat-
ter is not easy unless one has a considerable 
background in philosophy. So the professor 
who uses the book will find himself having 
to provide the necessary philosophical and 
logical background which Giere assumes. 

Even with Giere's exposition and the pro-
vision by the instructor of additional mate-
rial, the light only seems to dawn for most 
students as one works one's way through 
Giere's excellent exercises applying the pre-
vious material. A major reason for this 
"dawning light" is that the student perceives 
that having dealt with the nature of theories 
in enough detail to puzzle the student, 
Giere, after introducing the concept of the-
oretical hypotheses, ignores theories en-
tirely and both uses and invites, in his 
exercises, the use of the concept of theoret-
ical hypotheses in a much broader sense than 
that given above quoting from Giere~. 70.) 
The two conditions are phrased entirely in 
terms of theoretical hypotheses. So, most 
students openly wonder, why did we even con-
cern ourselves with the difficulties of 
theories or even models? The answer is, of 
course, that these are important in under-
standing scientific thinking and the dif-
ference between that kind of thinking and 
pseudo-scientific thinking; but Giere, in his 
effort to make everything easy, invites the 
students' question and the general criticism 
that the book's treatment of scientific 
thinking is inadequate, Yet perhaps one 
understands Giere's dilemma: an "adequate" 
treatment perhaps stands in the way of suc-
cessfully leading the average non-specialist 
student through to the point where he can 
distinguish between acceptable media reports 

and 1 
Perh. 
SUCCI 

cuss 
Gierl 
dist. 
from 

As 
of b< 
adde· 
who 
just 
repo 
for 
abou 
~aki 
the 

It 
thin 
vinc 
be t, 

RI 
M 

A. S 
UniV4 

In1 
textl 
and I 
watel 
Scri1 
altel 
avoie 
for 1 
of Ie 
natul 
demO! 
stru( 
this 
exam: 
tive 
prac1 
in tl 
anal~ 
teacl 

SCl 
a nut 
even 
apprc 
he g: 
of "! 
he e! 
its c 
nurnbE 
mal l 
out, 
have 



rht of 
Iced-
as a 
.a-
.nate 
that 
. does 
, the 
Iwn 
is 

rue) , 
: only 
.s. 

.nd 
,nt 
h 
sci-
.ia. 
e-
Jean 
can 

is-

ies 
just 
.e by 
.e 
ies 
r 
does 
iere 

t 
s 
s 
mat-
rable 
sor 
ng 
nd 

ro-
ate-
ost 

pre-

eives 
ories 

the-

oret-
than 
O. ) 
in 
st 
con-

of 
r-

d 
n his 
the 
cism 

te" 
uc-
list 
n 
orts 

and those which are obviously unacceptable. 
perhaps! But one wonders if ~t~dents, w~o 
successfully use the two cond~t~ons to d~s­
cuss media reports, will have learn7d from 
Giere enough about the nature of sc~ence to 
distinguish good reports from bad when, .Y7ars 
from now, he has forgotten the two cond~t~ons. 

As I completed this report, th.is criticism 
of both Giere and my teaching was given an 
added point when I told one of my students 
who completed the class very successfully 
just two months ago that I had finished the 
report. He asked: "Did you criticize Giere 
for getting into all that complicated stuff 
about theories when all along he was merely 
~aking a very simple point about the use of 
the two condi tions?" 

It really is very difficult to make some 
things easy in a satisfactory way or to con-
vince o~~ers that some things ought not to 
be thought of as simple. 

REASONING 
Michael Scriven 

A. S. Carson 
University of Alberta 

Into a growing market of informal logic 
texts, many of which are either interminable 
and Doring lists of argument fallacies or 
watered-down formal logic manuals, Michael 
Scriven introduces Reasoning, an engaging 
alternative. With a consc~ous attempt to 
avoid making argument analysis a witch-hunt 
for fallacies or an over-technical glossary 
of logic terms, he endeavours to explain the 
nature of reasoning and argumentation and to 
demonstrate, step-by-step, how arguments are 
st~uctured and a systematic means by which 
th~s structure can be drawn from prose and 
examined. The enterprise is as much a posi-
tive tuition in the construction of sound 
?ractical arguments as it is skill training 
~n the rather more negative art of critical 
analysis. It is, most commendably, a 
teaching device to the last. 

Scriven makes a considerable departure from 
a number of accepted views in logic and an 
even greater distancing move from standard 
approaches to informal logic. On the former, 
he gives up on talk of "validity" in favour 
of "sound inferences" and "true premises"; 
he eschews the use/mention distinction and 
its quotation mark symbolization; he makes a 
number of attacks against the utility of for-
mal logic in "real reasoning", Scriven points 
out, for instance, that formal systems can 
have value to natural language arguments only 

17 
if (al the natural language can be encoded 
into the symbolism (b) transformations can be 
made within the calculus (c) the symbolism 
can then be decoded into the natural language; 
the value in this process being in the reli-
ability of the transformations. However the 
encoding step, he maintains, is as problem-
atic in all but trivial cases as the assess-
ment of the original argument would have 
been. And he goes on to claim that formal 
systems have never developed a satisfactory 
way of dealing with assumptions, something 
which real arguments are rarely without. 
Equally, formal systems are rife with para-
doxes such as that of "material implication"; 
and the willingness of logicians to substi-
tute formal simplicity for practical utility 
is evident from the work of C. I. Lewis, and of 
Anderson and Belnap who simply do not tackle 
the need to distinguish "guarantees of truth 
for £" from "good reasons for believing that 
£" from "allows the derivation of £". Says 
Scriven: "The truth of e guarantees the 
truth of £, but it sure ~sn't a good reason 
for believing it" (P. xvi). To contrast '''ith 
this, and to introduce the approach to be 
taken to informal logic [the latter point 
mentioned above), Scriven states: "This book 
is about good reasons, not repetitions or 
transformations. It's just a start on what 
shouldn't be but is, almost an untouched 
subject" (ibid). 

What Scriven is after in this book, then, 
is to show the reader what real reasoning is 
and how to do it well. Specifically, his 
aims are to improve one's skill in "analysing 
and evaluating arguments and presentations" 
'7nd in "I;resenting argument'S, reI,'orts and 
~nstruct~ons clearly and persuas~vely." To 
be improved as well are one's "critical 
instincts" and "knowledge about the facts and 
arguments relevant to a large number of im-
portant contemporary issues in politics, 
education, ethics, and several practical 
fields" (p. ix). 

These aims are to be brought about by doing 
what other texts of this sort do not. For 
one thing, reasoning is to be construed more 
broadly than simply argumentation. Reading 
with understanding, for instance, is to be 
taken as a form of reasoning. Moreover, 
reasoning is shown to be a social enterprise 
--something that has to do with language, 
rational persuasion, open-mindedness, a com-
mitment to truth and even a moral commitment 
to respecting other people's rights to make 
up there own minds on the basis of reasons or 
evidence. For another thing, the analysis of 
arguments is to be taught, not as is often 
the case, solely by a combination of principle 
and demonstration, but through a series of 
seven procedural steps, each of which is 
carefully explained and illustrated. These 
procedures direct the student's attention to 
careful reading (or listening), to distin-
guishing between reasons and conclusions and 
the particular relations among them, to 
drawing out assumptions or unstated premises 
(and being "charitable, i.e. fair," to the 
arguer in so doing), and to evaluating the 
argument, not just in light of the truth of 
premises and soundness of inference, but in 
view of, and this is important, other pos-
sible arguments that might be brought to 
bear. It is, perhaps, this last point that