Sometimes it Gets Personal

Recently, a colleague of mine felt stung by something I said in an argument. He complained that I was making an “ad hominem” argument. I want to clarify this issue, because it’s a common mistake: what I was making was a personal attack, not an ad hominem argument.

A personal attack is when someone criticizes you as a person. They may argue that you are a bad person, and that you ought to mend your ways. Their argument may be logically valid. For instance, consider this statement: “I think you are behaving unethically by stating or implying, in your advertising literature, that ISTQB certification represents an industry-wide consensus about what excellent testing is or should be, when there is widely documented controversy over this among serious, experienced, and respected thinkers in the field. In fact, you have personally experienced and acknowledged that controversy over a period of years, in conversations with, among other people, me, and I have the email records to prove it.”

The argument here is that it is unethical to make a claim that one knows to be false, or that can easily be taken to mean something that it doesn’t mean. This is a general argument. I could be more specific and argue that it violates clauses 3, 6, 7, and 9 of the IEEE Code of Ethics, but specificity can vary in a natural language argument without committing fallacy.

This may or may not be a compelling argument. There may be a valid counterargument. But this argument has a valid logical form. To utter this statement would be a personal attack, I think, because it is a complaint about personal behavior. But it would not be an ad hominem attack.

A personal attack is not a good idea if you want to maintain a high quality personal relationship, but that’s a whole other issue. I hate burning my bridges with people, but sometimes I don’t see any other option that preserves integrity.

An ad hominem argument is a fallacy of reasoning whereby someone justifies a criticism of your idea by arguing that you are a bad person. It’s a variation of the non sequitor argument (e.g. “The sky is blue because I like pie.”)


An example of an ad hominem attack is “Mr. Attila the Hun, I realize that you think you have at long last proven Goldbach’s conjecture– that any even number greater than two can be expressed as the sum of two prime numbers– but I’m sure you must be wrong, since you are, after all, just a murderous Hun. Now go! And take that Horde with you.”

See the difference?

The rhetorical effect of claiming that an argument is ad hominem is to release the claimant from having to answer the personal attack. It’s a common tactic, but that tactic may itself be logically invalid. The retort “Your argument is ad hominem because it is directed against me as a person” is a non sequitor.

I wish I never had to be in the position of speaking out against what looks to me like bad behavior. I wish I never had occasion to make an attack against a person. Perhaps someday I’ll give up and let other people stand up for what seems right. For now, I don’t see an acceptable alternative. Someone, somewhere, someday, is going to be killed by software that is tested by well-meaning people who got negligent advice about how to test. I feel like I should do something.

As always, I ask my friends and enemies alike to help me identify and correct any of my own bad behavior. Communities must police themselves. Hence, I accept the IEEE Code of Ethics.

— James

7 thoughts on “Sometimes it Gets Personal

  1. Speaking of “common mistakes”, I’d consider personal attacks to usually be a symptom of Fundamental Attribution Error,

    It is definitely worthwhile to speak up against bad behaviour. It is likely a mistake to assume that the bad behaviour is due to personality, thus justifying a personal attack. It is definitely a mistake to believe that the only alternative to attacking people is to do nothing.

  2. Hi Jason,

    Are you talking about all personal attacks? By “usually” do you mean to imply that our system of criminal justice, in which every prosecution is a personal attack on the accused, is more than half of the time aiming its attacks at innocent people? Certainly we presume that people are innocent until proven guilty, but it is not out of bounds in any community to have and to enforce moral and ethical standards, if there is evidence of misbehavior.

    I’m not confident that I know what it means to attack a “personality”, but I can look at someone’s behavior and make judgments about that, just as I am subject to the judgments of other people. This happens in the public sphere and in the private sphere, as well as the semi-public spheres represented by communities of practice.

    Of course, there are alternatives to making a personal attack. Sometimes I don’t see a better alternative. That’s what I was saying. The particular person and issue that inspired this blog posting has a long history with me, and I already tried several alternatives before “flipping the bozo bit”. It’s a last resort. It has to be a last resort, because after that most other possibilities are off the table.

    Don’t you think it is wrong, in some cases, to refrain from making a personal attack? I’m talking about a forthright accusation that the accused person has an opportunity to answer. I have many times seen people subtly and silently ostracized in such a way that they never know who has turned against them or why. That behavior makes my skin crawl. I don’t want to do that to people, and I hope no one does it to me.
    — James

  3. I think an adversarial justice system is fundamentally flawed but that’s another discussion.

    By usually, I mean essentially the Fundamental Attribution Error. Humans have a tendancy to attribute behaviour to disposition over other situational explanations. Usually this means they make an incorrect attribution.

    I think it is wrong not to have a direct, explicit, and specific confrontation when bad behaviour has occurred. Note that the subtle, silent ostracization you’ve described is a personal attack, just not an explicit one.

    The error is the assumption that the only choice is personal attack or do nothing (which inevitably ends up leaking out into subtle attack anyway). I believe the term is bifurcation fallacy.

    You can directly confront behaviour without engaging in personal attack. I like Crucial Conversations / Crucial Confrontations as references for how to do this.

  4. I would recommend the book Crucial Confrontations as well.

    In your example, I think you were not making a personal attack. You were confronting bad behavior. This is different than making a personal attack.

    Confronting bad behavior:
    “I think you are behaving unethically by stating or implying, in your advertising literature…”

    Personal Attack:
    “You are an unethical person”

    It is good to confront bad behavior in both personal and work relationships. Resorting to personal attacks is almost always bad for relationships, and therefore bad for any goal you may share in common with that person. If you resort to personal attacks, then you have some bad behavior of your own that needs confronting…

    [James Reply: Ach! Great idea. I like that usage better. Thank you.]

  5. I for one try never to make it personal. I can think of no single instance where I have personally attacked a person, in a professional situation, that led to a successful/better outcome.

    That said I think that one of the things that often happens is that some people interpret emotional as personal. It is a matter of engagement, some people want to come into work, do their jobs and never really get involved at an emotional level. They don’t care if the product is good or bad, it doesn’t matter if you ship on time or late. Any problem that can be assigned to someone else is not their problem.

    People who are engaged on a project have a different attitude. They want the product to be as good as it needs to be. They want it to ship as soon as possible. Any problem that threatens the project is a shared project that given the appropriate priority they are willing to contribute to. I think of these people as being emotionally involved, they care.

    The conflict comes when you have interaction between the engaged and the dis-engaged. Imagine someone in a meeting, just waiting for yabba dabba do time roll around, just doing their job. In rolls a couple of engaged team members. They start to argue, they really get into it, they get emotional!!!

    I have found that those people who are not engaged will often have a response of, don’t get so emotional, don’t make it personal.

    Those people that are engaged, they don’t see it as personal. They see it as doing there job. How often have you walked out of meeting, having had a great argument, with a new respect with your opponent? How often have you and your opponent walked away not with one person right and another wrong, but with a new solution that is better than anything either of you were initially proposing?

    I love it when people when come into a meeting and get emotional about the work. It brings out the best in me and provides the best solutions to the problems we face. If I find that an encounter strays over that thin line between emotional and into the personal, means only one thing. Time to de-escalate. Right or wrong, good or bad, nothing good is going to happen when it is personal.

    Humor, diversion, calling a break… Just do not expect to continue hammering on the same personal issues and expect a different outcome.

    Scott Rosenbaum

  6. If I read this right, you’re saying that cries of “Ad hominem attack!” are themselves often simply ad hominem attacks — ?

    [James’ Reply: They are more often simply non sequitors, I suspect. To be an ad hominem attack, you not only have to protest that something is an ad hominem attack when it isn’t one, you also have to use that as a reason not to consider the all the other arguments that your opponent, whatever they are. For example: “You are just an ad hominem arguer, so I don’t have to listen to anything you say!”]

  7. “Mr. Attila the Hun, I realize that you think you have at long last proven Goldbach’s conjecture– that any even number greater than two can be expressed as the sum of two prime numbers– but I’m sure you must be wrong, since you are, after all, just a murderous Hun. Now go! And take that Horde with you.�

    This should be the dictionary definition of ad Hominem.

Leave a Reply to Jason Yip Cancel reply

Your email address will not be published. Required fields are marked *


This site uses Akismet to reduce spam. Learn how your comment data is processed.