Open Journal of Philosophy
2012. Vol.2, No.2, 89-91
Published Online May 2012 in SciRes (http://www.SciRP.org/journal/ojpp) http://dx.doi.org/10.4236/ojpp.2012.22013
On the Impermissibility of Telling Misleading Truths in
Kantian Ethics
Cameron Shell e y
Centre for Society, Technology and Values, University of Waterloo, Waterloo, Canada
Received February 24th, 2012; revise d March 18th, 2012; accepted April 6th, 2012
Sandel (2009) has recently revisited the issue of the moral permissibility of telling misleading truths in a
Kantian ethical framework. His defense of its permissibility relies on assimilating it to simple truth telling,
and discounting its relationship with simple lying. This article presents a refutation of Sandel’s case. It is
argued that comparison of misleading truths with telling truths or lies is inconclusive. Instead, comparison
with telling of leading truths is appropriate. With this comparison in view, it is clear that telling mislead-
ing truths is not consistent with the Categorical Imperative, meaning that they are not morally permissible
from a Kantian perspective.
Keywords: Kantian Ethics; Misleading Truths; Truth; Lying
I Did Not Have Sexual Relations with
That Woman
Sandel (2009) has recently re-presented the issue of the
moral permissibility of telling misleading truths in the frame-
work of Kantian ethics. The purpose of this article is to exam-
ine Sandel’s defense of misleading truths and present an argu-
ment as to why this defense does not succeed.
Sandel takes up the canonical example of President Clinton
assuring Americans that, “I did not have sexual relations with
that woman, Ms. Lewinsky”. It might reasonably be concluded
from this assurance that the President believed that he and Ms
Lewinsky did not have any sort of sexual encounter. In the
event, the President took his assurance to mean only that he and
Ms. Lewinsky did not engage in any mutual sexual activity; she
pleasured him but not vice versa. In short, the President said
something that he believed to be the case but that he anticipated
would be misinterpreted by his audience, to his advantage.
Sandel defends this sort of procedure as moral, from a Kant-
ian perspective, because it “pays homage to the truth” (p. 137).
On this analysis, Clinton acted from two motives: 1) to tell the
truth as he saw it; and 2) to avoid an embarrassing sex scandal.
The first motive is blameless on Kantian grounds (see below).
The second motive appears indefensible because it involves a
manipulation of the audience by the speaker and manipulation
is impermissible in Kantian ethics. This argument is put suc-
cinctly by (Herman, 1993: p. 154 n28):
The moral question turns on whether I tell you what I be-
lieve to be true as information of use to you as well as me
(and so contribute to the process of deliberation), or
whether what I tell you is controlled by a commitment to
regulate the flow of information in such a way that you
can deliberate only as I will. Because our deliberative ca-
pacity is vulnerable to manipulation in this way, respect
for rational agency requires a commitment to avoid inten-
tional falsehoods and misleading truths.
In other words, the onus is on the speaker to avoid deliber-
ately framing assertions in a way that the audience will misin-
terpret. In this crucial respect, misleading truths are no different
than lies or, for that matter, even simple truths designed to ma-
nipulate the audience.
In Sandel’s view, however, the onus is on the audience to in-
terpret speakers with more care (p. 137):
If everyone who found himself in a dangerous or embar-
ras sing situation resorted to carefully crafted [but true ] eva-
sions, people would not necessarily cease to believe them.
Instead, people would learn to listen like lawyers and
parse such statements with an eye to their literal meaning.
This is exactly what happened when the press and the
public became familiar with Clinton’s carefully chosen
words.
This defense may strike readers as formalistic, that is, an ap-
peal to the “letter of the law”—the Categorical Imperative—and
not its spirit. It also opens up the difficult problem of what sort
of diligence the audience members owe themselves in inter-
preting what they hear.
In any event, both arguments depend upon establishing that
telling a misleading truth is no different than either lying or
truth-telling, respectively. Such arguments appear inconclusive
because telling misleading truths bears resemblances to both. In
the following sections, I review why this dilemma holds, and
then suggest that telling a misleading truth is best compared not
to simple truth-telling or lying but to telling leading truths.
Through this comparison, the moral impermissibility of telling
misleading truths becomes clearer, even on a formalistic under-
standing of Kantian ethics.
Truth and Lies
From a Kantian perspective, telling the truth is a simple af-
fair. It consists of asserting something that you actually believe,
with the intention that your listener should accept that you be-
lieve it, as evidenced by your assertion. The situation could be

C. SHELLEY
captured as follows:
Telling a truth
Belief: you believe proposition A
Intent: you intend the listener should accept that you believe
A Act: you assert A
Telling a lie is similarly straightforward. It consists of as-
serting something that you actually disbelieve, with the inten-
tion that your listener should accept that you believe it, as evi-
denced by your assertion. The situation could be captured as
follows:
Telling a lie
Belief: you believe proposition ~A
Intent: you intend the listener should accept that you believe
A Act: you assert A
The difference between telling a truth and a lie resides in the
different relationships between what the speaker believes and
what the speakers intends for the audience to understand about
that belief.
Why Lying Is Morally Impermissible
In Kantian ethics, lying, so construed, is immoral. This is so
because it is self-defeating when universalized via the Cate-
gorical Imperative. To see this point, suppose that everyone
made free to communicate with the sort of intent described for
lying above. The intent is just the same as the intent for telling
the truth, as is the act that you employ to communicate your
intent. In that case, no listener could judge, merely from the act
of asserting A, whether you actually believe A or you believe
~A. They could conclude only that you believe A or ~A. Since
(A or ~A) is not the same as A, your intention to have the lis-
tener accept that you believe A is undermined. You cannot
achieve your intent by lying. Thus, it is irrational to lie.
Struth and Slies
On this view, lying is impermissible because it can be con-
fused with telling the truth. There is nothing inherently wrong
with asserting A while believing ~A. The problem is that others
will have no way of distinguishing this situation from the most
obvious alternative, in which you assert A while believing A.
To make this point clearer, consider this scenario. A mad
scientist creates a potion that causes people to assert the nega-
tion of what they intend to assert. In that case, two options for
simple assertion are open, which I will call telling a “struth”
and telling a “sly”.
Telling a struth
Belief: you believe proposition A
Intent: you intend the listener should accept that you believe
A Act: you assert ~A
Note that this scheme is the same as telling a truth, except
that the assertion is negated.
Telling a sly
Belief: you believe proposition ~A
Intent: you intend the listener should accept that you believe
A Act: you assert ~A
This scheme is the same as lying, except that the assertion is
negated also.
This situation raises an interesting question: Is it morally
permissible for you to tell a sly? There is nothing inherently
faulty with this form of communication. However, if we take
telling a struth as the default, then telling a sly is impermissible
according to the Categorical Imperative. If everyone made free
to tell slies, then no listener could be sure, when you assert ~A,
whether you believe A or you believe ~A. In that case, they
could conclude only that you believe A or ~A, which defeats
the purpose of telling the sly.
The point of this scenario is to illustrate that the moral im-
permissibility of lying in the Kantian framework is not an in-
herent feature of lying. Instead, it derives from a comparison
between telling a lie and telling a truth within a system of
communication. To evaluate the permissibility of a linguistic
action requires us to find an alternative with which it may most
readily be confused. This point is crucial to evaluating the per-
missibility of telling misleading truths.
Misleading Truths
Telling a misleading truth consists of asserting something
you believe with the intent that the audience will conclude that
you believe some other claim that, in fact, you do not.
This situation may be captured as follows:
Telling a misleading truth
Belief: you believe propositions A and ~C
Intent: you intend the listener should accept that you believe
A and C
Act: you assert A
In this case, you rely on the assumption that the listener be-
lieves that you both hold that A implies C. Thus, when they
accept that you believe A, they conclude—erroneously—that
you believe C also.
In the case of President Clinton’s testimony, the situation
might be as follows:
A = I did not have (mutual) sexual relations with that
woman.
C = I did not have any sexual relations with that woman.
President Clinton asserted A with the intention that his audi-
ence would conclude that he also believed C, although he did
not. Thus, he told a misleading truth.
Sandel’s argument that telling a misleading truth is morally
permissible relies on assimilating it to simple truth-telling as
described above. In both cases, the argument goes, proposition
A is both believed and intended to be communicated. Since that
feature is the crucial feature of truth-telling, which is known to
be permissible, it follows that telling a misleading truth is also
permissible. This is what Sandel means when he says that a
misleading truth still “pays homage to the truth”.
One problem with this argument is that it discounts the cru-
cial difference between these situations, namely the difference
in intentions. In the case of telling a simple truth, the intention
stops with the goal of having the listener accept your belief in A.
In the case of telling a misleading truth, the intention includes
the goal of having the listener accept your belief in C. In this
respect, telling a misleading truth is more akin to lying, in
which the believed proposition and the intended communica-
tion are not equivalent. This objection is essentially that given
by Herman. If so, then telling a misleading truth appears to be
both permissible and impermissible from a Kantian viewpoint.
Clearly, this state of affairs is not satisfactory.
90

C. SHELLEY
Leading Truths
It appears, then, that determining the permissibility of telling
misleading truths cannot be settled through comparison with
telling simple truths or lies. This situation suggests that perhaps
neither comparison is appropriate. Indeed, there is another com-
municative action to be considered, namely telling a leading
truth. In this act, you assert some proposition A with the inten-
tion of the listener accepting that you believe another proposi-
tion C that you do, in fact, believe. This situation may be cap-
tured as follows:
Telling a leading truth
Belief: you believe propositions A and C
Intent: you intend the listener should accept that you believe
A and C
Act: you assert A
In this case, you rely on the assumption that the listener be-
lieves that you both hold that A implies C, as in the case of the
misleading truth. Thus, when they accept your belief in A, they
conclude—rightly—that you believe C also.
In an alternative version of the Clinton case, the propositions
might be represented as follows:
A = I merely allowe d that woman to plea sure me
C = I did not have (mutual) sexual relations with that woman
Of course, this leading truth might not spare Clinton the
brunt of the scandal he feared. Perhaps, though, the public
might have been more forgiving of its greater candor.
It is clear from comparing the schemata for telling leading
and misleading truths that they might be confused. Faced with
the assertion A, and believing that A implies C, the listener
could not be sure whether the speaker believes C or believes ~C.
In that case, the listener could conclude only that the speaker
believes C or ~C, which defeats the intention of telling a mis-
leading truth. Thus, contrasted with telling a leading truth, tell-
ing a misleading truth is not permissible according to the Cate-
gorical Imperative.
Conclusion
Recent argument over the moral permissibility, from a Kant-
ian perspective, of telling misleading truths has been framed in
terms of comparisons with simple truth-telling or simple lying.
This approach is inconclusive since misleading truths bear a
significant resemblance to both.
On a formalist view of Kantian ethics, such as that described
by Sandel, the resemblance of misleading truths and lies may
be discounted. It is up to the audience to carefully parse the
meaning of the assertions they encounter.
This article presents a formalist refutation of this defense.
The refutation proceeds by shifting the comparison of mislead-
ing truths away from simple truths and lies over to telling lead-
ing truths. Telling a leading truth is the communicative action
most readily confounded with telling a misleading truth. The
possibility of confusion with leading truths means that telling a
misleading truth is not consistent with the Categorical Impera-
tive, and is thus morally impermissible on a formalist interpret-
tation of Kantia n ethic s.
REFERENCES
Herman, B. (1993). The practice of moral judgment. Cambridge, MA:
Harvard University Press.
Sandel, M. (2009) . Justice: What’s the right thing to do. New York, NY:
Farrar, Straus and Giroux.