toward the discovery of the truth of the conclusion, or as 
having as its goal the discovery of a good case which could 
be made out for the conclusion. But I believe that my view 
of the purpose of argument analysis and appraisal is a 
natural, and fairly common view. And I do not believe that 
PC2 is a useful principle to adopt for this purpose. This is 
because it tends to confuse the construction of an argument 
(by the critic) with the appraisal (again by the critic) of 
someone else's argument. It is not up to the critic to figure 
out a good argument based on someone else's materials, by 
leaving out irrelevancies and inserting the most plausible 
supplementary premises he can possibly think of! The 
critic's task is to understand and evaluate what someone 
else has put forward as an argument. Adopting PC2 makes 
it all too easy to read one's own ideas into the works of 
other people, under the guise of "charity". One can read 
enough extra premises into fragmentary pre-Socratic 
writings to turn these into interesting arguments which 
would be plausible to some modern philosophical minds. 
Suppose we do this with, say, Heraclitus. Are we then 
analyzing the arguments of Heraclitus? I think not. Rather, 
I think that we are reconstructing Heraclitean arguments 
(arguments incorporating some of Heraclitus's key themes), 
using some original materials. We better understand thinkers 
of the past when we limit the amount of reading in we do, 
and distinguish interpretation from reconstruction. Under· 
standing charity as PC2 leads us away from this distinction. 

Between PC1 and PC2 in its thrust stands another 
possible Principle of Charity. This one, which I'll refer to 
as PC3, would urge us to endorse the more plausible of 
several distinct interpretations equally licensed by the 
actual text or discourse. If we can extract from a text 
several different arguments - call them A *, A * *, and 
A *** - we should regard that text as expressing which· 
ever of those arguments is the best. It is worth remarking 
here that such a principle would not give entirely determi· 
nate advice, for A * may be the "best" argument in one 
respect, and A** "best" in another. (For instance, A* 
may be inferentially perfect due to the insertion of sup· 
plementary premises; but A * may have problematic pre· 
mises. Whereas A ** may be inferentially weak, because of 
a hasty generalization from a few premises, but be strong 
as far as the acceptability of its premises is concerned. 
This is a common situation - one which will wreak havoc, 
incidentally, for Johnson's proposed Principle of Discri· 
mination, unless he imposes severe restraints on adding 
premises.) However, I shall ignore this problem for the 
moment. PC3 may well be a reasonable principle to work 
with, although it would require considerable clarification. 
However it is not as strongly charitable as PC2, due to its 
insistense that interpretations be licensed by the actual 
text. That is, in adding or deleting, one will seek a basis 
for one's decisions in the discourse as given, and its con· 
text. One will not delete purely on grounds of irrelevance, 
nor add purely on grounds of weakness, strength, or plausi· 
bility. 

Perhaps the Principle of Charity which Johnson 
extracts from Scriven is triply ambiguous (triguous?1. for 
it could be taken to express PC3, as well as PC1 and PC2. 
However, it is clear that both Johnson and Scriven work 
with PC2 as the primary meaning. I suggest that the 
Principle of Charity, in this sense, is implausibly strong in 

6 

its charity. I suspect that it owes the plausibility and the 
wide acceptance it appears to have to its being easily con· 
fused with either PC1 or PC3. Instead of restricting the 
application of PC2 to serious contexts, as Johnson recom· 
mends, we could drop PC2 altogether and endorse either 
PC1, or PC3 (or, with suitable qualifications, both) as rules 
of charity. PC2 undercuts a primary purpose of argument 
analysis, is inefficient, leads to whitewashing of poor 
arguments, confuses interpretation with reconstruction, 
and licenses too much reading into others' material. 
Perhaps the best strategy for charity at home would be to 
dispense altogether with this kind of critical charity. t~ 

~, 

note 

Charity Begins Much 
Earlier Than Supposed 

Nicholas Griffin 
McMaster University 

Ralph Johnson in his article "Charity Begins at 
Home" (lLN, iii.3) says that the first use of the term 
'principle of charity' of which he is aware is Stephen 
Thomas's Practical Reasoning in Natural Language (1973). 
In fact the term has been around for considerably longer 
than that. To the best of my knowledge, it was intro· 
duced by Neil Wilson, "Substances without Substrata", 
Review of Metaphysics, 12 (1959) in pp. 521-39. In this 
version it was a rule for translators: 

We select as designatum that individual which 
will make the largest possible number of ... 
statements true. (p. 532). 

The principle has seen considerable use (and alter-
ation) since then. But, apart from acknowledgements by 
Quine (Word and Object, [Cambridge, Mass.: M.I.T. Press, 
19601, p. 59n; Ontological Relativity [New York: Colum-
bia University Press, 19691, p. 46n) its provenance has not 
been widely recognized. ... 

~ 



reply to Griffin 

I am happy to take note of the information sent 
along by Professor Griffin. It is interesting to learn that 
the Principle of Charity has been enunciated in other 
contexts - e.g., as a rule for translators proposed by 
Wilson. However, I am mainly interested in its use by 
informal logicians as a principle of argument analysis. It 
is clear that such is not the use to which Wilson's principle 
is geared. Hence I would like to amend my original 
assertion to read: "So far as I am aware, the first mention 
of this principle as a principle of argument analysis is to be 
found in Thomas's ... " 

~ Ralph Johnson 

teaching note 

A System of 
Rational Appi'aisal 

Robert Binkley 
University of Western Ontario 

editors's note 

A couple of years ago, Professor Robert Binkley 
shipped us a bundle of material from a logic course he 
was teaching at the time. We've dug out yet another 
useful item from that bundle: his graphic for his "System 
of Rational Appraisal." We've added a few comments of 
our own. 

1. System of rational appraisal. The flow chart 
above the cartoon represents the whole appraisal system in 
broad strokes. The first four boxes - Discourse, Surface 
Analysis, Classified Non-argument and Representation of 
Surface Structure - are spelled out in fuller detail on the 
second page, in the "Surface Analysis Flow Chart" (see 
below). 

7 

To us the most striking feature of this chart is its 
distinction between "surface analysis" and "depth ana-
lysis". We may have it wrong, but we take this to refer 
to the distinction between what someone's discourse may 
be made out to mean when taken literally, or better, when 
the person is taken at his word (which may certainly allow 
and call for interpretation), and what we may safely infer 
the person's underlying or deeper message is. Irony and 
sarcasm are perhaps the clearest examples of discourse 
whose meaning may be missed by a surface analysis. 
Humour is another. 

A "surface" argument may be so wildly illogical that 
literal interpretation would be not only uncharitable, but 
stupid: such illogic could only be the product of wit; 
hence the need for depth analysis and its verdict: "No 
serious argument intended here. It's a joke." (Without 
this distinction Binkley's cartoon would be seen as an 
example of a logical fallacy. So it's vital.) 

The unclosed side-boxes - "Charity, Fidelity and 
Discrimination" and "Relevance, Sufficiency and Accept-
ability" - merit glosses. 

By "Charity" and "Fidelity", we presume that 
Binkley was referring to the Principles of Charity and 
Fidelity found in, among other places, Scriven's 
Reasoning. The Principle of Fidelity means that the cri-
tic must be faithful to the original argument; the Principle 
of Charity requires that one provide the best possible 
interpretation 9f the argument; and the Principle of Dis-
crimination (though not explicity so referred to by Scri-
ven) requires the critic to "go to the heart of the matter", 
i.e., give prominence to the strongest criticisms and not nit-
pick or waste time on minor points. 

About "Relevance", "Sufficiency", and "Accept-
ability", since they come from our text, we can add this: 
we hold in Logical Self-Defense that in a logically good 
argument the premises are relevant to the conclusion, 
together they provide sufficient support for the conclu-
sion, and each must be worthy of acceptance by the 
audience of the argument. 

2. Surface Analysis Flow Chart. This is pretty self-
explanatory, but we have one comment of explanation and 
one caveat. At the bottom of the chart the path of analysis 
branches, going to either "Draw Diagram" or "Standardize", 
or to both. By "diagram" we believe Binkley had in mind 
tree diagrams of argument structure such as those used in 
Scriven's Reasoning and Thomas's Practical Reasoning in 
Natural Language. By "standardize" he is referring to the 
system introduced in Logical Self-Defense consisting of 
writing the premises above the conclusion they are put 
forward to support, and numbering them for convenience 
of reference (P1, P2, P3, etc.). The point of making the 
two alternative argument-structuring methods available is 
that tree-diagramming is easier and more perspicuous for 
longer and more complicated arguments, standardizations 
can be made immediately for simpler arguments, and a 
standardization can be written up off a tree diagram. 
(Metanote: We have come to think the standardizing tech-