Category Archives: Guilt by Association

"Guilt by Association"

very progressive journalist, and pioneering user of
the trope “guilt by association”

All my adult life as a newspaperman I have been fighting, in defense of the Left and of a sane politics, against conspiracy theories of history, character assassination, guilt by association and demonology.  I.F.Stone

A “new McCarthyism” is seen in the manner in which guilt by association has been pursued by the likes of Glenn Beck and “mainstream” GOP leadership (if there is such a thing).  Katrina vanden Heuvel, The Nation

In the United States of America, we don’t practice guilt by association. And let’s remember that just as violence and extremism are not unique to any one faith, the responsibility to oppose ignorance and violence rests with us all.  Jeremy Ben Ami, President, JStreet

 The tenuous “evidence”—later discredited—that landed Arar in a rat-infested cell was guilt by association. And if that could happen to Arar, a successful software engineer and family man, who is safe?  Naomi Klein

This is the story of a late-twentieth century invention, namely the ostensible moral and intellectual sin of accusations of “guilt by association.”

This trope, “guilt by association,” or GbA,  has a curious history and a curious present.  It has the following characteristics:

1) The trope user is almost invariably a self-described person of the “Left,” or, in somewhat more modern usage, a “progressive.”  The target is someone perceived as, or at least designated as someone opposed to the Left, a “right-winger.”

2) The trope has a surface resemblance to accusations of established errors of reasoning — fallacies — but in fact it is the user of the trope who is illogical and irrational.

3)  The accusation underlying the usage of the trope is as much moral as intellectual;  the trope user combines a disdain for the ethics and morality of the target (the ostensible bad faith of so-called right-wing McCarthites, for example) with an accusation of intellectual incompetence (failure to understand elementary logic).

4)  The trope enables its users, who are often devoted supporters of totalitarian and other hateful movements, to pose as moral and intellectual superiors.

Morris Raphael Cohen (1880-1947)
Except when an author is involved in left-wing political polemics himself ( e.g. Fearnside and Holther in “Fallacy,” 1959),  books on formal logic do not discuss this trope;  despite the claims by its proponents, it is not one of the recognized “fallacies.”  But there is, or can be, some kernel of truth in the otherwise mindless GbA trope, namely that generalizations can be inappropriate.  Here is what that eminent American logician and long-term CCNY professor Morris Raphael Cohen (with Ernest Nagel) had to say in their Introduction to Logic and Scientific Method (1934):

We have so far discussed the relation between premises and conclusion in case of rigorous proof.  But complete or conclusive evidence is not always available, and we generally have to rely on partial or incomplete evidence.  Suppose the issue is whether a certain individual, Baron X, was a militarist, and the fact that most aristocrats have been militarists is offered as evidence.  As a rigorous proof this is obviously inadequate.  It is clearly possible for the proposition Baron X was a militarist to be false even though the proposition offered as evidence is true.  But it would also be absurd to assert that the fact that most aristocrats are militarists is altogether irrelevant as evidence for Baron X having been one.  Obviously one who continues to make inferences of this type (Most Xs are Y’s, Z is an X, therefore Z is a Y) will in the long run be more often right than wrong.  An inference of this type, which from true premises gives us conclusions which are true in most cases, is call probable

Those who employ the GbA trope misconstrue statements of probability to make them appear to be statements of certainty.  For example, the Wikipedia article on GbA employs a Euler diagram to argue the obvious:  if some B is part of C, it does not follow that all of B is C.  But in the political discussions to which the GbA users address themselves, the arguments by the GbA targets are not arguments of certainty.   It is not (typically) claimed that all members of a Communist front organization were dedicated Stalinists.  Insofar as such arguments were at all serious, they were arguments of probability, not certainty.

The “guilt” in the GbA trope is also telling.  “Guilt” is a term most frequently used in the criminal law, where the standard of proof is much higher  — “beyond a reasonable doubt” —  than in the everyday world of political discussion.  The judgements we make in ordinary scholarship and in ordinary life  rely on what seems more probable, not on what seems probably beyond a reasonable doubt.   During the lifetime of the late Paul Robeson, for instance, both he and the Communist Party always insisted that he was not a Communist at all, just a very progressive person.  (After he died, the CP revealed that he had been a secret Communist all along).  But in his lifetime, given all the various associations of Robeson, it was reasonable to hold, by a balance of probabilities, that Robeson was a Communist, even absent proof beyond reasonable doubt.

Moreover, the trope “guilt by association” is ambiguous in its very nature.  It is regularly applied to the following types of statement, among others:

1)  A was once seen in a certain bar in which the notorious gangster B was also seen.  Therefore A is a gangster.

2) A is a member of five groups that were dominated by the Communist Party.  Therefore there is a certain probability — whether high or low needs to be established by all the other circumstances — that A is also a Communist.

It is the gravamen of GbA proponents that the truth-value of propositions 1) and 2) is exactly the same, namely nil.  That is of course preposterous on its face.  Pace these progressive writers and activists, associations among men are varied.  Sometimes negative inferences can be drawn from them to a greater or lesser degree of probability.  In some instances, as for example in those designated by the law of conspiracy, association may indeed give rise to valid findings of criminality.  In other cases association may be totally harmless.  Most generally, human associations are relevant without being conclusive in a great many of the judgements that we are called upon to make.    The proponents of the GbA trope must know this as well as we all do;  in the course of their daily lives they must know, just as the rest of us do, how to chose their spouses, their friends, their business associates,  their merchants, all on the basis of some sort of “guilt by association” judgements.  But when it comes to politics, these progressive GbA proponents declare that all evidence of human association is ultra vires, inadmissible for discussion in the market place of political ideas.

The origins of the GbA are not altogether clear.  The usage seems to have arisen in the post-WWII era, most specifically in the nineteen fifties.  The country was faced, on the one hand, with a Stalinist conspiracy, both through an elaborate network of Communist front organizations and Soviet espionage.  On the other hand, there were demagogic politicians, notably Senator Joseph McCarthy, who sought to use the Soviet conspiracy for his own purposes by making exaggerated claims of Communist penetration of the US government.  But there were indeed many Communists in places of influence, for example in the trade unions, who by and large attempted to rid themselves of Communist domination.  The trope “guilt by association” seems to have arisen in this atmosphere as a defense mechanism by Communists and their fellow travelers.  I. F. Stone, quoted above, was one of the most prominent users of the trope.  The logic was always this:  true, some members of the front organizations are Communists, some may even be Communist spies.  But this has no relevance, no relevance whatever, to the nature of the “progressive” (read front-organization) movement.  Not a few of these progressives had been students at CCNY during the tenure of Morris Raphael Cohen;  their ears had obviously been deaf to his teaching.

Today, the trope seems to be used in two specific efforts by the progressives.  The first is to criticize (and to misconstrue) the public’s concern over Islamist terrorism.  This concern is termed “Islamophobia,” a fairly new addition to the progressive polemical armamentarium.  The GbA argument runs as follows:  a) it is true that some Muslims are terrorists;  b) not all Muslims are terrorists; therefore, c), it is unjust, it is “guilt by association,”  to be more concerned over activities of American Muslims than over those of American Christians and Jews.  The fallacy of the trope, of course, is to construe the heightened concern by the public as holding that “all Muslims are terrorists.”  This latter proposition is not advanced by anyone in public life who is at all serious.  Were it to be encountered, it would of course be both false and malicious.

The second GbA effort concerns the overlap of self-described “leftists” and “progressives” on the one hand with the organized anti-Israel movement on the other. As I have shown in a previous posting, the progressive group JStreet contains a sizable number of aggressive opponents of Israel.  Those of us who point to this association are regularly accused of using “guilt by association.”  The logic, or rather the illogic of this accusation takes the same form as that of the other GbA accusations that we have seen.

I recently reported my finding that six of the nine identified top leaders of the Occupy Wall Street movement were also active in the anti-Israel movement.  One reader, an ordained rabbi no less, wrote to complain that I was engaging in a “guilt-by-association” argument.  I wrote back, explaining, among other things, that I made no accusation of “guilt” but I also insisted that surely, to a thinking man, there would be something of interest in this finding.  “Nothing of interest at all,” replied the rabbi,   “what you say is a red herring.”  Red herring ?  Here is another left-wing trope from the fifties. My curiosity was aroused.  “Rabbi,” I wrote back, “indulge  my curiosity:  do you personally support the boycott movement against Israel ? ”  “I will not answer this question;  it has no relevance to our discussion,”  replied the good rabbi.  Well there you have it:  an I. F. Stone of our time, bearer, unlike his predecessor, of the nice Jewish name of his birth.

Obama’s Defense: "You Say Guilt By Association"," or YSGBA

In previous postings I have complained, as have many others, that Senator Obama’s loyalty to his pastor, the Reverend Dr. Jeremiah A. Wright, Jr., reflects badly on his fitness for high office. This pastor is on record asking God to “damn America.” He is on record as giving high praise to the notorious Minister Farrakhan. And this pastor has reprinted in his “pastor’s column,” with his specific endorsement, an op-ed piece by a leader of Hamas. (For a description of the virulent, radical anti-Semitism of Hamas, see the New York Times, April 1, 2008). All this, many of us suggest, reflects on Obama, no matter how many or how few of the offending sermons he actually heard first hand.

But those who criticize Barak Obama in this manner are accused, by Obama’s friends, of using “guilt by association.” We are told that this criterion is illogical and abusive, if not “McCarthyite.” So “You Say Guilt by Association,” or YSGBA, is presented as a self-contained, self-justifying defense of Obama.

It is interesting that “guilt by association” is not a phrase that is ever used by those who presumably believe in it. Like other such polemical expressions (see, for instance, the article by Klehr and Haynes on “premature anti-Fascist“), it is used by people who accuse others of advocating ideas that they do not, in fact, either advocate or hold.

What can be said about this bugaboo, mostly used by the Left, of “guilt by association” ?

Eliot Spitzer and the Mafia

A few days ago the New York Times ran a story, very prominently on the front page of its Metro Section, entitled “Call Girl Linked to Spitzer Knew Reputed Mob Affiliate.” The story was murky but went on for a few hundred words. It seems that a certain Mr. Anthony Scibelli “who the authorities contend is an associate of organized crime” had an acquaintanceship or relationship (of an undisclosed nature) with Ms. Ashley Alexandra Dupré, the professional lady with whom Spitzer is said to have consorted in Washington.

So here it is. Our ex-governor, in addition to all his other sins and problems, seems related, by no more than three or four degrees (depending how you count) to, yes, to the dreaded Mafia.

What are we to make of this story ? Let’s try a thought experiment. Let’s say that Spitzer, instead of being the relatively unideological figure that he in fact is, were a figure beloved by the Left. In that case we could imagine a “progressive” reaction to the Times’s treatment:

This is a clear case of guilt by association ! The New York Times, for reasons best known to itself, sees fit to smear a progressive leader. True, Mr. Spitzer knows Ms. Dupré. But then it is alleged that Ms. Dupré knows a certain Mr. Scibelli (note that there is no proof that she does), and it’s also alleged that Mr. Scibelli “is an associate of organized crime” (no proof of that either). But let’s assume, for the sake of argument, that Ms. Dupré is indeed acquainted with Mr. Scibelli, and let’s assume further, for the sake of argument, that Mr. Scibelli is a Mafioso. But how does that affect Spitzer ? Mr. Spitzer has had dealings with Ms. Dupré and Ms. Dupré knows Mr. Scibelli. How can that make Mr. Spitzer a Mafioso, except by the fallacious, reactionary, malicious reasoning known as Guilt by Association. Outrageous.

Ambiguities of “Guilt”

This argument in defense of Spitzer — “You Say Guilt by Association” (YSGBA) — seems to have validity, at least on the surface. But it will not stand closer examination. The problem lies in what is formally known as the fallacy of equivocation, that is “the misleading use of a word [or words] with more than one meaning (by glossing over which meaning is intended at a particular time)” (Wikipedia). Both “guilt” and “association” are deeply ambiguous.

Let us start with “guilt.”

In the context of Anglo-American criminal jurisprudence, a criminal defendant cannot be found “guilty” except by “proof beyond reasonable doubt.” This standard, proof beyond a reasonable doubt, helps to define the very meaning of “guilt” in this context. It is a very high standard for the prosecution, and in effect allows the defendant to benefit from all (reasonable) doubts in the case.

In going about our business in matters not related to crime, in the courts or elsewhere, we do not and cannot be guided by this very rigid, one-sided standard. In civil law cases, for instance, the ordinary standard is much lower, viz. “”balance of probabilities,” sometimes termed “preponderance of evidence.” (In certain cases an issue is decided by “clear and convincing evidence,” which is more demanding than “balance of probabilities,” but not as stringent as “proof beyond a reasonable doubt.”) So it is clear that whatever “guilt” or wrongdoing may mean outside the realm of the criminal law, it is something quite different from criminal guilt.

Moreover, there is yet another standard that is particularly relevant to a discussion of the behavior of politicians and public servants. If a judge appears to have a conflict of interest, for example, he must recuse himself from the case. Justice not only needs to be done, it must also seen, or appear, to be done. An appearance of impropriety by a public official, whether or not there is a substance of impropriety, cannot be tolerated. Professor Deborah Hellman, among others, has shown that there are good reasons indeed for this rule.

It would certainly appear that Spitzer’s second-hand relationship with organized crime is of legitimate interest to the public, as, indeed, is Obama’s second-hand relationship with organized anti-Semitism. In either case, there is at least the appearance of something gone awry. This something, absent a criminal prosecution, need not be proven “beyond all reasonable doubt.”

The YSGBA defense, then, must fall because of the ambiguity of its use of “guilt.” It alleges that those who accuse Obama (or Spitzer) do not use the strict standards of criminal guilt-finding when, in fact, the accusations are not related to crime (Of course there may also be criminal aspects to the Spitzer case, but that is not the issue here.)

The other term of the YSGBA defense, “association,” is similarly vague.

Ambiguities of “Association”

We may be “associated” with others in a great variety of ways. When I take a subway train to the Borough of Manhattan, I sit with many others in a subway car, associated with them for the duration of my trip. Some of these fellow-travelers may be felons, but, obviously, this does not make me a felon. The proponents of the YSBGA defense suggest that since some types of association are totally innocent in this way, all association is innocent.

We know, of course, that some association is far from innocent. The criminal law itself recognizes criminal association of various sorts. In addition to provisions dealing with aiding and abetting, there is also the law of conspiracy, which holds a conspirator in a crime to be as guilty as the actual perpetrator.

But beyond the domain of law, our ordinary experience teaches us that the company we keep helps to define who we are. A person who habitually associates with known criminals, for example, is rightfully suspect. My Jewish prayer book enjoins me to ask God every morning to “deliver me this day, and every day, from arrogant men and from arrogance, from a bad man, from a bad companion and from a bad neighbor…” [My thanks to my nephew Butchie for pointing me to this]. The Mishna (Kelim 12,12) tells us that “Who consorts with the unclean becomes himself unclean.” The New Testament concurs: “Bad company corrupts good morals” (I Corinthians 15:33).


When we complain about Obama’s associations, we do not allege that these associations make him a criminal. We are not bound by a standard of “truth beyond a reasonable doubt.” Nor is all association innocent. Some is deeply troubling. We are right in scrutinizing the associations of those who would be President. And Barack Obama’s associations are not reassuring.