, has 
!hat 
his 
,fying 
h of 
Asks, 
,at 

.out 

oqic 

loCJY 

is 

look 
s of 
!., to 

huen, 

s, 

I 

I 

Ild 

t 
ld 

predict while walking to school that 
he was going to be given the test on 
Friday. So the test must take place 
between Monday and Thursday. But the 
same argument works for Thursday. 
That is, on Thursday morninq, any 
student could deduce from the facts 
that there can be no surprise test on 
Friday, and that there will be a test, 
and as it is Thursday the correct pre-
diction is that the test will be given 
that day. Clearly the argument can be 
extended to show that the test cannot 
be given on Wednesday, Tuesday or Monday. 
The conclusion is that the test cannot 
be given at all. 

The teacher heard this objection 
out, and then gave the test on the 
following Tuesday, surprising, in 
the required sense, everyone. 

The puzzle here is to see what 
has gone wrong with the argument. 
Clearly the teacher can give the 
surprise test. How rs-it the case, 
then, that an apparently impeccable 
argument can produce the conclusion 
that no surprise test is possible? 

******************** 
It is odd to confront a piece of reasoning 

that is valid only on some particular Thursday 
evening or Friday morning, but this very od-
dity suggests a key to the puzzle: time. The 
bright student in the story says, "If Thursday 
~nt by, and still no test, it couldn't be a 
surprise on Friday, so we can scratch Friday." 
--Sorry, but Friday has this about it, that 
you can't scratch it for real until Thursday. 
That is, you can't scratch it for real by 
maginin~ that Thursday's class has ended and 
ffius ru11ng Friday out. The reasoning in the 
puzzle derives its appearance of force from 
our forgetting that, for us humans, the whole 
time between the start of Monday·s class and 
the end of Thursday's has to be lived through 
before· a student is in a position to downgrade 
the teacher's logic. It is within that se-
quence of days that the teacher can bring off 
her surprise test. 

The time-range in which she can spring the 
test extends from the start of Monday's class 
to near the end of Thursday·s. As Thursday's 
class passes its halfway mark, the student 
does not know if she will give the test in the 
minutes remaining. Suppose she does; then the 
student will have no grievance, for the teach-

i er came through with the test at a time he 
could not predict for certain on his way to 
school. But what can the student say if she 
doesn't give the test on Thursday? "You let 
too much time go by-- now the element of sur-
prise is gone." This is hardly a lo~ical 
lapse on the teacher's part, though 1t may 
show a bit of absent-mindedness. The main 
point, however, is that her student is not in 
a position to make even that quarded judgment 
~til the sands of Thursday's class run out. 

With these considerations in mind, I wonder 
if what we have here could be called an exis-
tential paradox, in as much as the puzzle can 
take hold of the student only if he forgets a 

7 

certain temporal feature of his "human exis .... 
tence, namely that he cannot' reason himself 
forward to the end of the week, and then work 
backwards through time, aut has to exist 
through the intervening days one by one and 
wait to see what each day brings. 

conference repo rt 5 

A PANEL ON INFORMAL LOGIC 

This Report was submitted by Professor Samuel 
Pohr of the University of Pittsburgh at 
Bradford. 

A Panel on Informal Logic was presented 
a~ th7 Beh7end Campus of Penn State Univer-
S1ty 1~ Er1e, Pa~ at the Spring meeting of 
the Tr1-State Ph1losophical Association on 
April 21, 1979. The Panel was organized and 
ch~ired,by William Rapaport of the State 
Un1vers1ty of New York, College of Fredonia. 
The ot~er P~ticipants were Samuel Fohr of 
the Un1vers1ty of Pittsburgh at Bradford 
James Liotta of Lake Erie College, and N~lson 
Pole of Cleveland State University. 

~amuel Fohr pointed out that informal 
10g1c courses could help people to arrive at 
more true beliefs and fewer false beliefs 
But the value of such courses could be se;i-
ously diminished by how they were taught and 
the books which were used. Philosophers have 
not been as rigorous in their treatment of 
non-symbolic logic as they have been in their 
tre~tment of symbolic logic. Many writers of 
10g1c ~ks have been either sloppy or in-
correct ;n t~eir definitions of basic terms 
such as val1d,· ·sound," "deductive argu-
In7nt~· ~d ~inductive argument... Any way of 
d~st1ngu1sh1ng between deductive and induc-
t1ve,arquments which is not based on the in-
tent10ns of the person putting forward the 
argument is faulty. The word "fallacy" is 
use~ very loos~ly by many philosophers. 
~tr1ctly spea~1ng, a fallacy is an error in 
1nference,or ~n drawing a conclusion from 
so~ prem1ses. Yet philosophers have tended 
to 1dentify assertions they take to be false 
as fa~lacies. one refers to the fallacy of 
equat1ng determinism with fatalism, another 
to the fallacy of taking the rightness or 
wro~qness of actions to be related to the 
mot~ves for which they are done. Writers of 
~og1c,b~ks have gone far beyond this in 
1d7nt1fY1ng things as fallacies. Among other 
th1~gs the following have been identified in 
1091C ~ooks as fallacies: questionable claim 
(pract1cally every claim is), emotionally 
charged language, suppressed evidencp., dog-



matism, and appeal to force or threatening 
someone. One should at least divide the 
traditional group of fallacies into two 
groups--fallacies and tricks. 

Informal fallacies have been presented in 
a haphazard manner. When an author tries to 
sort them out, he may do so incorrectly. One 
author listed begging the question under the 
heading "Fallacions Because Invalid." One 
useful way to categorize the fallacies is by 
listing them all under three headings: 
irrelevance, hasty conclusion and circular 
reasoning. This can help a person see what 
is problematic about almost any fallacy. 
The teaching of the fallacies of composition 
and division as fallacies of ambiguity is an 
example of the confusion found in the presen-
tation of informal fallacies. For at least 
three sets of fallacies are covered under the 
heading "composition-division" only one of 
which involves using words in two different 
senses. " 

Informal logic courses can bB uaeful if 
only in getting students to first, differ-
entiate between passages which offer no argu-
ment and passages which do, and second, 
identify the premises and conclusions of an 
argument. But we would hope that informal 
logic courses would do IIOre. If they are to 
do so teachers and book writers are going to 
have to get their house in order. 

Nelson Pole noted that he had at one time 
taught a section on informal fallacies in his 
deductive -logic course. He gave it up for 
two reasons. First, it seemed to most im-
press students who wished to go into adver-
tising. They appreciated aid in developing 
their bag of tricks. Second, the material 
seemed remedial. people should know it be-
fore coming to college. Two developments led 
to a change of mind on teaching such material. 
First, he became aware that in many programs 
for bright students informal logic and its 
apPlications to real life arguments (as op-
posed to text book cases) were considered 
appropriate and challenging. Second, he be-
came aware that a non-remedial informal logic 
course was possible. He has designed a new 
course in applied logic. The goal is to 
enable students to be able to critically 
evaluate real life arguments as they are 
presented naturally and to be able to dissect 
a set of premises and construct from them a 
defensible conclusion. Little stress is put 
on the names of various fallacies and on 
identifying fallacies from a long list of 
names. The skills of the course include the 
following: how to find out what you believe 
(for IIOst people aren't clear about what they 
actually believe), how to identify and dis-
tinguish premises and conclusions, how to 
evaluate another person's argument, how to 
construct an argument, how to overcome blind 
spots in your thinking, how to detect con-
sistent and inconsistent statements, and how 
to pasS the logic portions of the Graduate 
Record Exam, the Law School Admission Test, 
and the Medical School Admission Test. The 
course stresses not only negative matters 
like finding fallacies, but also positive 
ones like constructing good arguments. Such 

8 

a course, when properly promoted, should 
attract a large number of students both from 
those who are having practical problems in 
reasoning as well as from those who desire 
to do well on pre-professional aptitude 
tests. 

James Liotta pointed up the fact that the 
decision to have either fnrmal or infor~l 
logic taught is not pedagogical. One isn't 
easier to teach than the other, or better 
than the other, or more valuable for stu-
dents. It would be best if students were 
exposed to an in-dept.'l study of both. Bnt 
schools are likely to re~uire or recommend 
only one course, and a short o~e at that. 
The decision as to which to offer then be-
comes "political." 

Certain problems come up over and over 
again in teaching logic. One is getting 
students to identify the premises and con-
clusions of arguments. Students seem to have 
trouble doing this eVBn when analyzina the 
simplest of arqu.'l\ents. Another proble!:\ is 
getting the students accustor.~d to do~~a 
formal logic. syllogistic reasoning, which 
is often taught in non-symbolic logic courses 
can form a bridge between informal and for~l 
logic. What is necessary is to teach syl-
logistic reasoning in such a way that the 
class terms are seen as symbols rather than 
words with a particular meaning. The use of 
bracketing and awkward constructiol"!3 can help [' 
students in this regard. The result of this 
is that they will more easily see the in-
validity of arguments whir.h "sound right" and 
more easily accept as valid ars~nts which 
are counter intuitive. GettL~g them used to 
manipulating symbols in this way will help 
give them confidence for working problems 
in symbolic logic and mathematics. For those 
who have a mental block against mathematics 
or anything that resembles it (and many stu-
dents do have such a block) these techni~ues 
can be a great help. 

Teaching Logic, whether formal or informal, 
will not cause students to be mere raticnal. 
But it can be of help to thos~ who arc al-
ready rational. 

William Rapaport discussed the =act that 
there were at least three broad cateao:ie~ 
of informal logic courses (all of them quite 
distinguishable from mathematical logic 
courses) corresponding to three sorts of 
texts. First there is informal logic as a 
bag of tricks: how to win arguments, in-
fluence people and defend yourself. ~his 
sort of course will include some info~l 
logic, some non-symbolic for~l logic a~d 
some symbolic logic. A text which corre-
sponds to this approach is Copi's Introduc-
tion to LOgiC, Second there is informal 
log~cii. a ~st of fallacies: how not to 
argue. Kahane's L02ic land contem~orary 
Rhetoric suggests l. ... se mere. ,t l.ra, 
l.nformal logic as a semi-formal way to read 
and think critically. This is the approach 
of Scriven in his book Reasoning. 

As to the value of a course in info~l 
logic, it should be pointed out that doing 
mathematical logic (e.g., learning a natural 

ded 
the 
pro 
sys 
tra 
sor 
typ 
log 
giv 
one 
is 
SOli 

met 
arg 
CO\; 

mell 
tyt: 
not 
SCI 

twc 
loe 
li! 
Thj 
thj 
In1 
pp. 

COl 

gUI 

Lee 
?Ii 
Co' 
pu: 
in 
of 
gri 
th, 
wa 
of 
in 

of 
ph, 
co 
10' 
on 
ha 
af 
Sa 
in 
de 
11'! 
jo 
pr 
co 
re 



ld 
b from 
s in 
lire 
e 

at the 
r::vll 
isn't 
I:er 
:u-
ere 
But 

oend 
it. 
be-

rer 
lq 
:on-
:0 have 
the 
~ is 
lC 

.rhich 
:ourses 
fOrr.lal 
ly1-
:he 
than 
Ise of 
Ul help 
: this 
,n-
It" and 
'hich 
ied to 
lelp 
!rn5 

: those 
ltics 
r stu-
licsues 

Iforrnal, 
,cnlll. 
al-

that 
'ien 
quite 

If 
IS a 
. n-
lis 
tal 
Ind 
'e-
Iduc-
iT'-
to 

read 
'oach 

deduction system) isn't directly useful in 
the everyday life of students. The crucial 
problems in applying a natural deduction 
system to a real-world argument are the 
translation steps (in both directions). This 
sort of problem is better addressed by a 
type-three course. Further, mathematical 
logic, given enough practice and success, can 
give the student a useful methodology, but 
one course isn't sufficient to do this. What 
is needed, then, is something more practical, 
something that will arm the students with a 
method they can use to construct or evaluate 
arguments (or any prose). The Scriven-type 
course offers that, but needs to be supple-
mented by Copi-type material. The Kahane-
type is good as a source of examples of what 
not to do, but this should come out of a 
Scriven-Copi type of course as well. 

Rapaport went on to describe a special 
two-semester inter-disciplinary informal 
logic course taught with members of the Eng-
lish Department. It was called "Effective 
Thinking and Communicating. n [A report on 
this course ha's been published in the 
Informal Logic Newsletter, Vol. I, No.3, 
pp. 6-7.J 

* * * 

CONFERENCE REPORT -- "LOGIC AND LIBERAL 
LEARNING" 

On June 11-13, 1979 the editors were 
guests at the Conference on Logic and Li~era~ 
Learning held at Carnegie-Mellon Un~vers~ty ~n 
pittsburgh and organized by Professor Preston 
Covey of C-MU's philosophy deparcment. The 
purpose of the conference was to investigate, 
in a critical way, the role that the learning 
of logiC does and ought to play in an under-
graduate liberal arts education, and we were 
there to see to what extent informal logic 
was considered in this review. This report 
of the conference will reflect our angle of 
interest • 

For a great many of these who read papers, 
of the panelists and of the 200-odd philoso-
phers who attended, the term "logic" in the 
conference title meant "formal deductive 
logic". This conveyed the impression that not 
only informal logic, but also inductive logic, 
have second-class citizenship. On the first 
afternoon of the conference, Professor f1erilee 
Salmon (Arizona at Tucson) was moved, at least 
in defense of inductive logic, to label as 
deductive chauvinists those who implied that 
"logl.c" = "aeauctl.ve logic". (The issue was 
joined over the question of how arg~ents ~x­
pressed in natural language and ord~nary dl.s-
course are most correctly and profitably to be 
reconstructed. Prof. Salmon used her epithet 

9 

to characterize those who claimed that all 
such arguments are to be r.econstructed as-
valid deductive arguments.) A fair amount of 
time was subsequently spent by people denying 
that their position was in fact tbat of de-
ductive chauvinism. During this controversy, 
it occurred to us later, there was no pause 
to consider other possibilities--for instance, 
whether there might be arguments in ordinary 
discourse that fit neither the deductive nor 
the inductive mold, or whether the move to--
reconstitute all arguments so as to make them 
amenable to formal analysis does violence to 
some of t.l).em. 

* * * 

Monday morning, opening session. 

Conference,Chairman Preston Covey set the 
stage by using the medieval distinction be-
tween logica utens (logic as a useful art; 
logic appll.ed}"iiiil 10¥ica docens (logic as a 
speculative science;ogical theory), to ask 
the questions I (1) What logical artifice is 
necessary for liberal education? and (2) ~~hat 
logical theory is required for the logical 
artifice that is necessary? 

In the morningts first paper, Nuel Belnap 
(Pittsburgh) argued that the main value of 
logic (read: formal logic) for freshmen is 
that it conveys the ideal of rigour. He was 
followed by Gerald Massey (Pittsburgh) who 
sketched the kind of formal logic course he 
would recommend. His thesis was that the 
development of natural deduction systems and 
truth tables has driven the axiomatic method 
from the field, and that what is wanted is a 
simple system to teach axiomatic logic--
transformation rules, the notions of proof, 
theorem, completeness, consistency and inde-
pendence. He emphasized that the axiomatic 
method cannot be understood unless one under-
stands the formal axiomatic method. Natural 
deduction systems should he retained to moti-
vate, for instance, the application of logic 
to common languagp. and to motivate under-
standing the horseshoe in terms of the 
deducibility-in-the-system of the consequent 
from the antecedent and focus on the idea of 
good inference rather than on intuitive Eng-
lish interpretations of "if ••.• thAn ... ". 
Horeover, these systems can be used to teach 
deductive strategies. Massey said that tech-
niques introduced in the last 20 years now 
make accessible to introductory students 
branches of logic that were once inaccessible. 
However. he warned emphatically, logic should 
not be touted as a way of sorting out good 
arguments from bad ones, since logic cannot 
show that an argument is invalid. (Massey 
argued that there is no non-relative, abso-
lute, system independent concept of validity, 
so whether an argument form is good or bad--
i.e., valid or invalid-=rs-also system-
relative. ) 

* • * 



Monday afternoon, second session. 
Thomas Schwartz (Dept. of Governmen't, 

Texas) opened the afternoon session with the 
provocative charge that the standard teaching 
of formal logic (formal theory, mechanical 
symbolic lO9ic) is irrelevant to the objective 
of teaching a person to be able to reason 
well, while standard informal logic courses 
sacrifice rigour for applicability. Schwartz 
proposed that we begin with a clear, well-
specified educational objective (including a 
target skill) such that its attainment is in-
trinsically worthwhile, and he suggested this 
objective should be the ability to pick apart, 
criticize and reconstruct arguments in the way 
that philosophers typically do. He outlined 
a step-by-step procedure for realizing this 
objective. First, reconstruct the argument 
under examination so that it is (a) deduc-
tively valid, (b) fully explicit, and (c) de-
void of extraneous material. (In doing so, 
use as criteria for tacit premises: fidelity 
to the author's intentions, generosity in 
rendering the argument plausible, and the 
maximum generality possible consistent with 
the other requirements.) Second, search out 
any ambiguity in the argument (expressions 
with ambiguous meanings, premises used in 
ambiguous ways, terms within premises used in 
ambiguous ways). Third, evaluate the argument 
by looking for premises that are false or not 
sufficiently plausible, checking for self-
defeating or question-begging premises, 
counter-exampling general premises, and test-
ing the arguerts assumptions by seeing if any 
premises could reasonably be made more general. 
Schwartz argued that while his approach re-
quired no symbolic apparatus, it was rigorous 
because, employing deductive validity as the 
central principle of organization and inter-
pretation, it constitutes a procedure that is 
rigorous. He argued, further, that by turning 
inductive arguments into deductive ones by 
adding the requisite tacit premise required, 
the underlying inductive principles assumed in 
such arguments could be made explicit. These 
can then be checked by looking for alternative 
explanatory hypotheses that are more plausible 
or more economical (i.e., have as much or more 
explanatory power with less complexity). 

* * * 

Monday afternoon, panel discussion. 
Following sc~wartz·s paper three panelists, 

Robert Fogelin (Yale), Merilee salmon (Arizona) 
and Douglas Stalker (Illinois, Chicago Circle) 
issued some reactions and interjected some 
new thoughts. 

Fogelin focussed on the importance of 
teaching students how to understand daily 
arguments, the usefulness of the theory of 
speech acts a la Grice and Austin for doing 
this, and the importance (he called this the 
main task) of making people explicitly aware 
of the moves actually taking place and what 
words are being used to do. Fogelin entered 
a demurral about the value of testing for 
deductive validity and stressed that typically 

10 

the situation is forensic, with considerations 
weighing on either side, as well as often 
strongly analogical. (Fogelin claimed there is 
a great need for the study of analO9ical rea-
soning. ) 

Marilee Salmon took dead aim, in her com-
ments, on the question of whether arguments 
are best understood as exclusively deductiv~, 
and came out four-square against'the positirn 
that they are. It was here that she intro-
du~ed the phrase "deductive chauvinism". She 
po~~ted out that her students, mostly in the 
bus~ness program, don't encounter arguments 
that can be construed as deductive in any 
helpful way. She urged starting with natural 
arguments and looking first at the structure 
~f inductive analogical arguments. Her exper-
~ence, she reported, was that after two-thirds 
of the term was spent on inductive arg\~ents 
and informal analysis of arguments, students 
are then ready for symbols, and are able to 
pick up such techniques as truth tables and 
Venn diagrams in a few days. 

Stalker raised the new ~uestion: What, if 
anything, is known about the effectiveness of 
our logic courses? Do they really teach ~he 
students anything--anything that remains with 
them? He insisted that there is a need for 
hard empirical dAta here, and, moreover, that 
at present none exists. We ought to be trying 
to get answers to such questions as: What are [ 
the intuitive judgemf'!nts r:ollege students need ,-
to test? What are the typical kinds of arau-
ments they encounter? (We need a freq'lenc~ 
count here.) Which textbooks work better than 
others? Many of the questions beina spec-
ulated about at the conference about thp. val11~ 
of what we are doing in teaching logic in 
various ways depend for their an~er~ o~ what 
the facts are. But no one seems to be gener-
ating these facts. 

r 
* * * 

Tuesday morning, third session. 
Preston Covey (Carnegie-Mellon) o:;;:ened the 

second day with a description of a humanities 
course he teaches to fres~~an arts, social 
science and busineso students, that has the 
interesting feature of using some s~lic 
logic as a heuristic constraint to focus at-
tention on the assumptions underlying issues 
in disputes about values and social policy, 
and the further interesting feature of using 
the computer to teach that symbolic logic. 
Covey uses the computer to teach ordinary 
sentential quantification logic in one month 
(instead of the usual entire semester). He 
then h~s his students reconstruct current 
arguments about social values in valid deduc-
tive form, with the goals of (1) making pre-
sumed logical connections as explicit and 
precise as: poSSible, (2) maki:n~ underlying ~ 
principles: as explicit and precise as possible ' 
and (3J mapping argumentative strategies by , 
being explicit about what is ne~essary or. suf-
ficient in the way of needed assumptions. 
Covey's experience with this approach to 
teaching a philosophy course about social 
values is that it brings out the real problems, 

, thE 
den 
to 

wal! 
twe 
(C2I 
bee 
gd 
ChE 
syn 
stl: 
thE 
on] 
arE 
le~ 
fie 
pre 
not 
stl 
anc 
Thj 

cal 
pre 
fol 
st21 
whE 
chE 
det 
fre 
pre 
hi:r: 
ale 
whE 
reE 
trl 
wit 
Mol 
feE 
lo~ 

up, 
abc 
rit 
th~ 
poi 
til 
bri 
in~ 
th': 
ly, 
bae 
tOI 

fe~ 
sci 
lii 
hal 
fre 
Whl 

abc 
car 
to 
dec 
evi 
th~ 
ore 
be: 
la1 



Itions 

Ire is 
rea-

:om-
Its 
:i·"~, 
.tirn 
~o-

She 
the 
lts 

:ural 
:ure 
ixper-
:hirds 
~nts 

'nts 
to 
.nd 

, if 
s of 
the 
with 
or 
that 
rying 
tare 
need 

rgu-
cy 
than 

Vlt 111"" 

what 
ner-

: the 
.ties 
11 
;h.e 

at-
lues 
:y, 
ling 

Inth 
He 

!duc-
Ire-
I 
.1] 
·si:ble, 
fly 
suf-

Iblems, 

the conflicts of principles, and forces stu· 
dents to get more $erious about how hard it is 
to come to grips with important issues. 

The second speaker in the morning session 
was Joan Straumanis (Denison), who explained 
two programs in Computer Assisted Instruction 
(Cal) which she has used in her teaching. She 
became interested in CAI to save time and 
grief, and to avoid the tedious business of 
checking proofs. A major problem in teaching 
symbolic logic is providing early feedback to 
students learning to do proofs. ordinarily 
the teacher first sees samples of such proofs 
only after notational and procedural errors 
are already ingrained and must then be un-
learned. And some students have so much dif-
ficulty in getting started that they cannot 
produce completed proofs at all, and have 
nothing to submit. In'large classes, the in-
structor cannot offer the kind of concentrated 
and individualized help that may be needed. 
This makes CAI worth serious consideration. 

The program developed by Straumanis is 
called D~~ON, a line-by-line proof checker for 
propositional logic. The program does the 
following. (l) It evaluates each symbolic 
statement supplied ~y the student to see 
whether or not it is well formed. f2) It 
checks each line of the student~s proof to 
determine whether it is a legitimate inference 
from the premises and earlier lines of the 
proof. (3) It provides error diagnosis and 
hints for correct rule interpretation. DEMON 
also is programmed to give positive feedback 
when the student gets the proof. Example 
responsesc " Ingenious I " "Super!" and "Stop 
trying to fool me!" DEMON can be combined 
with FRAME (a program developed by Professor 
Morton L. Schagrin) which offers drill and 
feedback on translating English into and from 
logical symbols. 

Straumanis noted that the class average is 
up, and average students seem most pleased 
about the process. (Bright students are ir-
ritated: they dontt want to take time with 
the computer) She suggested that from the 
point of view of theory the computer drama-
tizes the fact that a proof is deductive, 
brings up for class discussion the role of 
ingenuity in logic, and dramatizes the point 
that proofs can be generalized. pedagogical-
ly it provides immediate and positive feed~ 
ba~k has patience, speaks with the instruc-
tor'~ voice, provides privacy, distinguishes 
feedback from evaluation, allows for flexible 
scheduling, is self-paced, teaches computer 
literacy and overcomes the fear of computer 
hardware, and it is an explicit transition 
from English grammar to non-numerical symbols, 
which helps the students who have anxieties 
about math. Straumanis stressed that (a) you 
can't make the case that logiC teaches people 
to think, since people do not think this way: 
deduction doesntt model thought1 and (b) the 
evidence just doesntt support the claim that 
there is transferability of rigour and 
oroblem-solving skills to other domains. She 
believed that what may transfer are: trans-
lation skills and attention to the rules of 

11 

English grammar, ability to spot ambiguity 
the idea of proof, the method of math and ' 
maybe science, possibly skills useful in 
~egal reasoning. But the main payoff of CAI 
1S that it relieves the teacher from the non-
teachable elements of the course and frees 
her time for other things. ' 

• • • 

Tuesday afternoon, fourth sp.ssion. 

In the first afternoon session, Peter 
Facione (then, Director, Division of General 
Studies, Bowling GreenJ now, Dean of Human 
Development & community Service, California 
State University at Fullerton) began by urging 
th~t course content should be based on course 
goals, and more specifically that the desired 
student outcomes, specifically conceived, 
should be the rationale for the instructor's 
focus and approach. He proposed that a logic 
course should be conceived as producing crit-
ical thinking skills, which need to be pre-
cisely and operationally specified. He of-
fered the following seven skills as dOing this 
jobl (1) gathering information1 (2) iden-
tifying issuesJ (3) selecting arguments: (4) 
determining hypotheaes1 (5) evaluating posi-
tions1 (6) formulating positionsJ (7) expres-
sing alternatives. What is also needed is a 
decision procedure (or something like it) to 
help the student put into practice each of 
these skills-and to do so in an integrated 
way. Facione's suggested procedure for the 
skill of evaluating positions is outlined 
here: 

Step 1: Identify the parts of the 
argument. 

Step 2: Rewrite the argument. 

Step 3: Ask these four questionsl 
(1) Are the premises of the recon-

structed argument true? 
(2) Do the argument's premises, taken 

together, demonstrate that the 
conclusion must be true or is very 
probllbly true? 

(3) Do the argument's premises avoid 
deriving their support from the 
argument>. conclusion? 

(4) Does the argument's stated conclu-
sion amount logically to the co~clu­
sion the author intended to 
demonstrate? 

In the afternoon's second session Alex 
Michalos (Guelph) approached the role of for-
mal logic in a liberal education by first de-
veloping a picture of a liberal education. 
This he sees as an education which aims at 
~nc:e~sing the chances for development of the 
1nd1v1dual and the community--a free person in 
a free society. It should teach (1) character 
and moral virtue, which will promote social 
development, (2) the knowledge needed to im-



prove the world, and (3) the techniques needed 
to change the world. The question of whether 
there is a role for formal logic in liberal 
education thus becomes the question whether 
formal logic contributes to these aims. 
Michalos sees formal logic as essentially 
teaching the skill in drawing valid infer-
ences and avoiding contradictions. Clearly 
t~ese are intrinsically valuable skills, but 
are they more valuable than other things that 
might be taught in their stead? He claimed 
that much of what ~s taught in formal logic 
courses is busy-work, of little value in 
teaching the contradiction-avoiding skill, and 
not clearly better than alternatives. He pro-
posed that since an argument can go wrong in 
three ways--be invalid, have false premises, 
and be methodologically defective--a course 
that includes all three would be more widely 
relevant, and this would be a mixture of for-
mal and informal logic. 

Robert Baum (Rensselaer Polytechnic 
Institute), in the third afternoon session, 
urged a revision of our conceptualization of 
t~e inductive-deductive distinction. The 
standard way of characterizing deductive argu-
ments as those in which the arguer intends the 
premises to necessitate the conclusion is not 
useful in dealing with real arguments in every-
day contexts where it is usually impossible 
to determine the arguer's intentions. More-
over, this psychological criterion does not 
apply at all in formal logic, where form alone 
is of interest, and it prevents allowing for 
invalid deductive arguments. Baum proposed 
taking arguments as a generic category indi-
cating a relationship among sets of statements 
which could then subdivide into three cate-
gories: (1) valid deductive arguments, (2) 
inductive arguments of various degrees of 
strength from weak to strong, and (3) non-
arguments. With these distinctions students 
can approach arguments with an open mind and 
run a test to determine how to classify them. 
Baum went on to argue that the distinction of 
inductive arguments into those which are in-
ductive generalizations (with a universal 
proposition as a conclusion) and those which 
are inductions by analogy (with a particular 
proposition as a conclusion), is not a useful 
distinction. It is possible that all in-
ductive arguments employ analogies. And it is 
more important to recognize the multiplicity 
of possible kinds of inductive argument. In 
general, Baum said, we should discourage having 
too many categories and distinctions1 moreover, 
we should not lump inductive logic with infor-
mal logic, since formal analyses of inductive 
arguments are possible. 

In the fourth and final session Tuesday 
afternoon Wesley Salmon [Arizona at Tucson) 
spoke in praise of relevance on behalf of in~ 
ductive logic, in two senses. He argued, 
first, that we should try to make logic rele-
vant to important topics of the day that stu-
dents are or should be interested in (e.g., 
the DC-10 crash at Chicag01 the use of radio-

12 

active isotopes (tritium) in making illumi-
nated signs). These, and many others--
perhaps the most important examples--involve 
causal considerations in an absolutely signi~ 
ficant fashion. So it Should be an essential 
part of an introductory logic course to take 
up causal arguments, and Salmon expressed 
amazement that causal arguments had not yet 
been discussed during the conference. He 
averred that what treatment there is of caus-
al arguments in introductory texts is often a 
slanted and misleading discussion ·of Mill's 
methods. Salmon'S defense of relevance in the 
second sense emerged from his discussion of 
Mill's method of agreement and method of dif~ 
ference. He argued that we need to pay more 
att.ention to the concept of the relevance of 
causal conditions. This is a gap in the 
treatment of inductive 10gic7 a crucial logi-
cal concept is that of statistical relevance 
--the relationship between two probability 
~alues. He suggested that this can be exam-
1ned using ~ormal methods [cf. Carnap, Lngi-
cal Foundat10ns ~ Probability, Ch. 8). 
sarmon then proceeded to demonstrate some 
~roblems for the logic of confirmation, which 
1nvolves a relevance relation. In sum ele-
mentary logic should be relevant by en~bling 
students to evaluate causal and statistical 
arguments used in current public issues and 
theorists should devote attention to th~ rele-
vance relation in causal arguments. 

• • • 

Wednesday morning, fifth session. 

This session was given to the questio~ of 
how logical competence is to be evaluated. 
Robert Ennis and Thomas Tomko (Illinois, 
Urbana-Champaign) of the Illinois Rational 
Thinking Project gave papers. Ennis discussed 
his five-dimensional conception of logical 
competence, consisting of (1) interpreting the 
argument, (2) handling varieties of content, 
(3) judging in accord with logical principles, 
(4) handling complexities, and (5) applyina 
the skills in various contexts. In ir.ter--
preting, the problem of translation is key, 
and this involves a sense for the functions of 
various terms, for ambiguity, and in effect of 
rules for language use. One is moving from 
natural language to some standard fo~, and 
there are empirical questions about What works 
best here that have not been answered. Ther~ 
seems to be a close connection between reaClng 
skills and interpretation skillS. (This is 
being studied at the IRTP now.) Ennis noted 
that in handling content it has been found 
that onets belief in the premises or c~nclu­
sion seems to affect one·s ability to see the 
argument's validity. Also, arguments with 
arbitrary symbols, unfamiliar content and ab-
stract content are harder to reason with than 
their opposites. The complexity of the a=gu-
ment also seems to make reasoning more dif-
ficult, though it is hard to isolate this 
from the other components of rational thinking, 
Ennis discussed the applications of rational 
thinking skills in finding assumptions to 
construct the best line of reasoning, in 

~ 

maki 
in w 
comp 
iene 
10g1 
inst 
His 
duct 
good 
educ 

y 
aite 
line 
form 
impo 
ated 
ated 
and 
whet 
whol 
He d 
cons 
appr 
serv 
grou 
term 
gest 
9ath 
tive 
cal 
addi 
camp 

Conc 

A 
packl 
per. 
grat 
valu 
geth 
ne •• 
for 

W 
a gr 
dedu 
tanc 
gard 
noth 
ape 
nora 
exei 
expe 
tion 



.umi-

lvolve 
siqni-
lential 
I take 
led 
. yet 
He 
caus-

,ften a 
l1's 
in the 

n of 
f dif-
more 

ce of 
e 
logi-

vance 
ity 
exam-
Logi-

me 
which 
ele-

bling 
ical 
, and 
e rele-

0:1. of 
ad. 
, 
nal 
scussed 
cal 
inq the 
tent, 
ciples, 
ying 
er-
key, 
ions of 
feet of 
from 
and 

t works 
There 

reaCng 
is 
noted 
und 
nclu-
ee the 
ith 
nd ab-
h than 
argu-

dif-
.is 
hinking. 
ional 
to 
.n 

~king deductions to the best explanation, and 
in working with direct proofs and clarifying 
complex relationships. He recounted an exper-
ience of serving on a jury, when it seemed his 
logical skills were useful in interpreting the 
instructions to the jury for fellow jurors. 
His judgement is that there is a role for de-
ductive competence in good thinking, and that 
good thinking is important for a liberally 
educated person. 

Your reporters had to leave the conference 
arter Enni~'s talk. \'Ie can report this out-
line of Tomko's paper on the evaluation of 
formal logic competence. He spoke to the 
importance of clarifying what is to be evalu-
ated and .for what purpose it is being evalu-
ated, distinguishing between evaluating courses 
and students, and between evaluating to decide 
whether a course or prQgram is worthwhile as a . 
whole and to decide how it might be improved. 
He discussed the gathering of evidence~ the 
ronstruction of tests, the use of tests, other 
approaches to gathering evidence such as ob-
servation, surveys and questionnaires, small 
group discussions and interviews, and long-
term follow-up. He concluded with the sug-
gestions that more than one method be used for 
gathering evidence, that tests and alterna-
tive methods of assessment for measuring logi-
cal competence need to be developed, and that 
additional psychological research on logical 
competence is desirable. 

• • • 

Concluding remarks. 
As readers can gather, this conference was 

packed to the seams with ideas, differinq 
perspectives and novel suggestions. We con~ 
gratulate Preston Covey for the enormous and 
valuable job of putting the conference to-
gether and running it with enviable ~ooth­
ness. It was a useful hothouse exper1ence 
tor us. 

We came away with the impression that for 
a qreat many people logic still equals formal 
deductive logic, with inductive logic a dis-
tance second. Informal logic is still re-
garded by many as the study of fallacies and 
nothing more. It remains our belief that such 
a perception of informal logic stems from ig-
norance of its recent development and of the 
exciting possibilities these portend. We 
expect that as the word gets out this percep-
tion will slowly, if grudgingly, change. 

13 

conference not ices 

A symposJ:um on "Theory of KnOWledge. and 
Science PoltcyM is beinq held D~Cember 13-16, 
1979 at the University ot Ghent, Belgium. It 
is an interdisciplinary ~posium on the rel-
evance of new developments in the analysis of 
knowledqe for·both critical and constructive 
approaches to science policy. The organizing 
commi'1:tee is chaired by Prof. Leo Apostel 
(Ghent). 

The symposium proposes to confront devel-
opments in contemporary Theory of Knowledge 
(science of science, epistemology, etc.) with 
those in contemporary Science Policy and Crit-
icisms, in order to determine possible mutual 
interdependencies as well as eventual exclu-
siveness of alternatives in either domain. 
Researchers from both branches and scholars 
who focus on the interrelationships proper 
will present their views. A renovating pre-
sentation of scholarly work i8 expected 
through participation of researchers on a 
worldwide scale. 

Invited speakers include: Leo Apostel 
(Ghent, Belgium), Donald T. Campbell (North-
western U., Chicago, U.S.A.), Gerard de Zeeuw 
(Amsterdam, Netherlands), Nelson Goodman 
(Harvard, U.S.A.), Yves de Hemptinne (Unesco, 
Paris), George Hines (Nat. Univ., New Zealand), 
Jaakko Hintikka (Helsinki, Finland~ Florida, 
U.S.A.), Karlin Knorr (Vienna, Austria), 
Larry Laudan (Pittsburgh, U.S.A.), JQrgen 
Mittelstrass (Warsaw, Poland), Leszek Nowak 
(Poznan, Poland), Jean Piaqet (Geneve, 
Switzerland), Ilya Prignogine (Brussels, 
Belgium~ Austin, U.S.A.), Hilary Rose (Brad-
ford, UK). Steven Rose (The Open University, 
UK), Richard Whitley (Manchester, UK), John 
Ziman (Bristol, UK). 

• • 

For people who could not have attende~ any-
wal'·, it is almost as important to know about 
conferences held in the recent cast as about 
those in the future. Such information helps 
one keep track of interests and developments 
in the field. Here, then, is the list of 
topics at the Australian logic Teachers'