Don’t warn nonspecifically!

Warning Sign Phillip Island Victoria

This is a decent warning sign. Image via Wikipedia

I hate safety warnings. It’s not that I’m hurt by someone out there’s condescending belief that I can’t work out whether irons are for drying children. And I welcome the endless mental accretion of terrifying facts about obscure ways one can die. What really bothers me is that safety warnings often contain no information except ‘don’t do X’.

In a world covered in advice not to do X, and devoid of information about what will happen if you do X, except it will be negative sometimes, it is hard and irritating to work out when it is appropriate to do X. Most things capable of being costly are a good idea some of the time. And if you were contemplating doing X, you probably have some reason. On top of that, as far as I can tell many of the warnings are about effects so weak that if you wanted to do X for some reason, that would almost certainly overwhelm the reason not to. But since all you are ever told is not to do X, you are never quite sure whether you are being warned off some trivial situation where a company haven’t actually tested whether their claims about their product still apply, or protected from a genuine risk.

My kettle came with a warning that if I ever boil it dry, I should replace it. Is this because it will become liable to explode? Because it might become discoloured? My sandwich meat came with a warning not to eat it after seven days. Presumably this is because they can’t guarantee a certain low level of risk after that, but since I don’t know what that level is, it’s not so useful to me. If I have a lot else to eat I will want a lower level of risk than if I’m facing the alternative of having to go shopping right now or of fainting from hunger. Medical warnings are very similar.

Perhaps it’s sensible to just ignore warnings when they conflict much with your preconceptions or are costly. In that case, how am I worse off than if there just weren’t warnings? How can I complain about people not giving me enough information? What obligation do they have to give me any?

There is the utilitarian argument that telling me would be much more beneficial than it is costly. But besides that, I think I am often worse off than if warning givers just shut up most of the time. Ignoring warnings is distracting and psychologically costly, even if you have decided that that’s the best way to treat them. There is a definite drop in sandwich enjoyableness if it’s status as ‘past its use by date’ lingers in your mind. It’s hard to sleep after being told that you should rush to an emergency room.

I presume there are heaps of pointless warnings because they avoid legal trouble. But this doesn’t explain why they all contain so little information. It is more effort to add information of course. But such a minuscule bit more: if you think people shouldn’t do X, presumably you have a reason already, you just have to write it down. If you can’t write it down, you probably shouldn’t be warning. An addition of a few words to the standard label or sign can’t be noticeably expensive. For more important risks, knowing the reason should encourage people to follow the advice more because they can distinguish them from unimportant risks. For unimportant risks, knowing the reason should encourage people to not follow the advice more, allowing them to enjoy the product or whatever, while leaving the warning writer safe from legal action. Win win! What am I missing?

19 responses to “Don’t warn nonspecifically!

  1. People (not you) want to be told what to do, or tell others what to do; they just aren’t interested in the reasons why. Folks don’t demand the detail and thus the details aren’t supplied, in some cases they may strongly prefer less detail because it aids in self-deception.

    Consider cigarettes. A warning of “could be hazardous to your health,” while true, provides little information. Providing more information makes it harder for smokers to self-deceive about the risk. Such warnings are unpopular among both smokers and tobacco companies.

    In other words, safety warnings aren’t about safety. If they were, they’d likely come with more detail about the consequences.

    • In the case of cigarette warnings, it seems both the tobacco company and the smokers would prefer no warning at all, if you are right the smokers would prefer to self deceive. When someone steps in to enforce warnings they are better off forcing detailed ones, supposing they actually want to warn. So who in other cases prefers the intermediate warnings that warn with no detail? Or is it just a compromise between the parties who want to warn and those who want to deceive?

  2. For what it’s worth, I order warning labels all the time (I’m an engineer) and will keep this in mind. With regards to Zac’s comment, I’d say that most if not all warning labels I use are for actual safety (not legal), I just never thought of how doing this might help.

  3. People are more likely to read signs with fewer words than those with more words. Also, as Zac implied, a considerable fraction of the population does not care about the reason for obeying a particular warning. (We’ll leave aside the ethics of deliberately underserving this subpopulation for another day.)

    Possible solution: short warnings in large type, followed by short-as-possible justification in (usually) smaller type.

  4. I was assuming it was for legal reasons. If you give a reason, then you open yourself up to more interpretation and counter-argument. If you do feed the seagulls and something else bad happens, then perhaps you can sue.

  5. have you read marginal revolution lately?

    maybe warnings are not about safety, it’s about showing that the warning-giver cares.

  6. I presume there are heaps of pointless warnings because they avoid legal trouble. But this doesn’t explain why they all contain so little information.

    “We advise you not to eat this sandwich after the 26th, because otherwise you might sue us.”

  7. I presume there are heaps of pointless warnings because they avoid legal trouble. But this doesn’t explain why they all contain so little information. It is more effort to add information of course


    And, well, think of how much detail you’d have to add – and the jury’s reaction to someone skipping a few paragraphs of detailed risk assessment “because it was fine print” – and then awarding them damages because “the warning was too hard to read”.

    What you see is, as you suspected, the result of an attempt to prevent liability, not an attempt to present the most useful and detailed knowledge – and to be fair, most people don’t really want that level of detail, I think.

    (In my more curmudgeonly moods, I’d say most people aren’t capable of processing it effectively…)

  8. Once upon a time, my father, after observing that a container of chlorine bleach had a warning on it not to mix it with ammonia, decided to try it and see what happens…

  9. I think you more or less already know the answer; you just don’t like it. Although there is no reason not to make more information available in at least some way, and I’m sure it could be done in a way that would satisfy the lawyers, there is not a good enough reason to do it, so it doesn’t happen. You can consider all the reasons why we have warning labels, from people wanting to show they care to consequences of our legal system, but I think you’ll find the solution we have is roughly just enough to satisfy everyone enough that no one feels any pressure to do more. It’s a great example of the 10-times rule: something has to be at least 10 times better than what it is intended to replace (10 times cheaper, more functional, more efficient, etc.) in order to have a high probability of widely replacing it.

  10. Someone who is given a reason might decide that the reason is trivial (“I’m not afraid of a mere seagull!”) whereas someone not given a reason might fill in the blanks and come up with their own reason.

  11. I would think that simple space concerns probably do come up here, unless there is a single, easily summarized reason for not doing X. For instance, for the “use by” date on food, most people probably already intuitively understand that it may have “gone bad” by then. It’s also used to mark when the grocery store should pull it, and so that you have a general idea of how fresh the products at your grocery store are going to be when you buy them.

    Equally, if you’re just genuinely curious *why* you should mix ammonia and bleach, a quick trip to will probably satisfy your curiosity and let you make a more informed decision.

    A quick caution will help make someone aware of dangers they might not have considered (“Oh, mixing these could be harmful!”), and generally a more advanced user will be able to research the specifics.

    • An intuitive idea that something ‘may have’ gone bad is not very useful – it always may have gone bad. It seems in practice different people end up consistently interpreting the use by date different ways – some assume it means if you eat it after the date you will more likely get sick than not, while others assume there is a negligible risk. Most of these people are acting differently to how they would like to if they had better info, to their detriment.

      Googling more obscure risks is usually futile in my experience.

  12. For what it’s worth, to me the use-by date means “cook well-done” instead of “rare”.

  13. Pingback: Hidden motives or innocent failure? | Meteuphoric

Leave a Reply to Bob Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

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