Suppose you are my daughter and we are driving together to your driver's license exam at the DMV.Footnote ² ‘I can't wait to get this over with,’ you say, ‘so that I can finally drive myself to Millie's house and to debate club. No offense, dad.’ You give me a smile before unburying your cell phone from your bag and begin pecking away furiously.

‘No offense taken, hon,’ I reply, enjoying your excitement. ‘I can still remember how thrilled I was when my friend Radu and I went snowboarding after I got my license. Say, we never really finished talking about all the rules for using the car – we kind of left off in the middle of talking about the no-texting-while-driving rule.’

You sigh, then drop your phone on your lap. ‘I don't see why it's so bad to do a little texting while driving. It's not like I'd be driving drunk or something.’

‘So, you're pretty against drunk driving, huh?’ You stare at me incredulously. ‘Of course I am,’ you reply. ‘It's clearly wrong to do something so irresponsible and that puts people – yourself and others – at such unnecessary risk.’ Your incredulity quickly melts into suspicion. ‘You're not going to do your Socrates routine on me, are you?’

‘No, no,’ I grin. ‘I totally agree drunk driving is wrong. I was just thinking about some research I was reading that gives pretty good evidence that texting while driving impairs people's reaction times as much as, or possibly more than, drunk driving.Footnote ³ That raises a question: if drunk driving is wrong, isn't texting while driving wrong, too? After all, isn't texting while driving relevantly similar to drunk driving?’

Looking a bit like the owner of a lovable Pekinese who is thoroughly resistant to housetraining, you begin a not entirely believable lament about the perils of having a philosopher for a parent. ‘I don't think texting while driving is relatively similar to drunk driving,’ you continue as we drive into the DMV lot. ‘There are lots of differences. For one thing, you're not intoxicated when you're texting.’

‘Well, let's slow down a bit. I didn't say the two were relatively similar or even that they were the exact same thing. I suggested that they are relevantly similar: they are similar in all the morally relevant respects. There are certainly differences between the two activities. But I was wondering whether they are the same in all the ways that matter morally. After all, you said that what makes drunk driving wrong is that it puts people at unnecessary risk when you're behind the wheel. And the thing is that texting while driving has that feature, too, so unless there's some other relevant difference, shouldn't we say they're both wrong? You've given us a possibly relevant difference, but we need to think a bit more about whether it really undermines the argument against texting while driving.’

‘Here,’ I say, as we slide into the line at the counter. ‘Let me write down the argument so we can look at it.’ I pull out my notebook and pen and write:

1. 1. Driving drunk is morally wrong.

2. 2. Texting while driving is relevantly similar to drunk driving.

Therefore, texting while driving is morally wrong.

‘This is a moral argument from analogy. That's an argument that attempts to infer a conclusion about what we should do in one particular case – usually a tough one we're not sure what to think about – from what we think we should do in another similar case about which we are more confident.Footnote ⁴ We're confident drunk driving is wrong, and thinking about the possible connections between that and texting while driving might be useful.’

Seeing that your understanding is (as usual) outpacing my leisurely lecture, I hasten to add, ‘Of course, not all arguments are good ones. So, we have to figure out if this is a good argument. This argument would be bad if one or more of the premises are false. So, maybe we could test the argument. It seems like it'll be hard to reject premise 1, because to do that we'd have to explain why drunk driving is not wrong. So, instead, we need to figure out if there are any reasons to think premise 2 is false. To do that, we should brainstorm potentially relevant differences between drinking while driving and texting while driving – differences that would allow us to say texting while driving is not wrong even if drunk driving is.’

‘That's what I was trying to say earlier,’ you insist. ‘There's a relevant difference. You're not intoxicated when you're texting, but you are when you're drunk. So that's why texting while driving is not wrong, even if drunk driving is.’

‘That's definitely an objection to the argument worth considering. But to see if it really undermines the argument, we have to ask two questions: is this an actual difference between the two activities, and is it a morally relevant difference between them? You're only going to have succeeded in undermining the argument if you get a “yes” to both questions.’

‘Now,’ I continue, ‘it is clear that this is an actual difference. Texting doesn't make you drunk, but drinking alcohol does. So for this objection the real question is whether this difference, between being intoxicated vs not, is morally relevant.’

‘That makes sense,’ you say, handing your driver's license application to the attendant at the desk, who seems both amused and perplexed by our conversation. ‘But how do we determine if a difference is morally relevant?’

‘Well,’ I reply, ‘for an example of a difference that's not morally relevant, consider a totally different example. Suppose you had two cases where someone stole someone else's candy bar, and they were exactly the same except in Case A the person stole with their right hand and in Case B the person stole with their left hand. The cases are exactly the same in terms of their outcome, the person stealing and being stolen from, and in the size and quality of the candy bar. The only difference is the hand the stealing was done with. But surely this difference by itself isn't morally relevant: we wouldn't want to say that the stealing in Case A was any less wrong than the stealing in Case B, or vice versa. This difference between these cases – the left hand vs right hand difference – is not morally relevant. Morally speaking, when all else is equal, it's just plain irrelevant what hand you steal with.’

‘That makes sense. But how can that help us with our argument?’

‘Right. Good question. Here's the idea. To test if a difference is morally relevant, you can go back and modify the original case to account for the difference and then see if it changes your moral evaluation – if it changes your view of the rightness or wrongness of the action. In our case, that means we could go back and modify the drunk driving case to account for the difference between being intoxicated vs not, then we could test if it really matters that you're intoxicated while drunk driving but not while texting while driving. We could test if that difference gets you off the moral hook, so to speak, for all those times your thumbs are flying on your phone while you're on the road.’

‘I see. But how could we make that kind of modification of the drunk driving case? If it's drunk driving, you're drunk.’

‘We might need to get a bit inventive. Imagine, then, that someone has just created a new drink that, instead of getting you intoxicated, just makes you extremely distractible. In fact, it impairs your ability to respond to threats on the road just as much as being intoxicated. Suppose that it's just as enjoyable to be under the influence of this distracting drink as it is to be intoxicated in the regular way. It's pleasant to be constantly drawn to new features of your environment or mental life: “squirrel!”, “Did I turn off the stove?”, “Is that new Star Trek movie out yet?” In fact, driving after drinking this special drink is the same as drunk driving except that the second involves intoxication while the first doesn't. Now here's the question: is your moral evaluation of the person who drives after drinking this new special drink any different than your evaluation of the drunk driver?’

You think a minute. ‘No, I don't think it is.’

‘Then it looks like the difference cited – between being intoxicated vs not – isn't morally relevant and can't be used to undermine the analogy. If you think we should appraise the two cases, driving after drinking alcohol and driving after drinking the special drink, the same, then the difference between being impaired by intoxication and impaired without intoxication is not morally relevant. And if that's the case, then our argument from analogy still stands. Even though texting while driving involves distraction rather than intoxication, that doesn't show that it's not wrong.’

‘Now, wait a minute,’ you say, with a wry smile. ‘You just made up that example. There is no such thing as a special drink like that. I know you like science fiction, dad, but I don't see how it is relevant to deciding what's actually right or wrong. How can thinking about a made-up example like that tell us about what's right or wrong in the real world?’

‘Now, that's a great question!’ I say, excitedly. ‘I don't think it really matters that I used an invented example. One thing thinking about moral arguments from analogy can do is help us to see if we're living out our commitments consistently. You're clearly committed to not driving drunk, because you view it as irresponsibly putting people at unnecessary risk. And thinking about the analogy to texting while driving helps us figure out if we're being consistent if we take a more accepting view of texting while driving. After all, if we have reason not to drive drunk because it puts people (ourselves and others) at unnecessary risk, then, to be consistent, don't we have to avoid texting while driving, since it puts people at a similar level of risk? The analogy can help us see that, unless we can identify a relevant difference between the two, we're being inconsistent if we don't treat them the same.’

‘That seems totally reasonable, dad, but I don't see what it has to do with making up outlandish examples.’

‘Fair enough. The important thing is that invented examples – like mine about the special drink – can help us test our consistency just as well as any other examples. When I was asking you how you would evaluate someone who drove after drinking the special non-intoxicating-but-distracting drink, I was checking your commitments – your attitude about what matters. Your judgment was that it doesn't really matter that the special drink doesn't intoxicate you – it's still wrong to drink it and drive. And if that's what you think, then aren't you being inconsistent if you say that texting while driving is permissible because you're not intoxicated?’Footnote ⁵

‘OK. I'll accept your strategy of using invented examples. But, still, I think there's a relevant difference between drunk driving and texting while driving.’

I try to contain my excitement. ‘Great! For any argument, it's good to test it by trying to think of the strongest objections to it and seeing if those objections stick. So, what relevant difference did you have in mind?’

‘Well, it seems pretty clear that when you're drunk driving you can't turn it off – you're impaired the whole time. But, while you're texting while driving you are only impaired intermittently, because you can put the phone down, then pick it up and text, and so on. That seems to me to be the reason texting while driving is not wrong even if drunk driving is.’

‘Let's use our strategy to see if this is a relevant difference, because it certainly seems like an actual difference. So, imagine that you could turn your drunkenness off and on while driving. Perhaps you have a special new party drink that, once drunken, allows you to turn your intoxication on and off with a certain head movement.’ Much to your chagrin, I begin to act this out. ‘Would it be morally permissible to turn on your drunkenness for periods during your drive? Would the fact that your intoxication is turned off for part of the drive really indicate that, unlike regular drunk driving, what you're doing is not wrong? If not, then it seems that this difference in duration of impairment isn't morally relevant and premise (2) is left unscathed by the objection.’

‘That does seem right,’ you say, so engrossed in our discussion that you don't notice you're almost about to be called for your exam. ‘That objection doesn't seem any better than the last. And now I'm thinking that the other objection I was going to raise is similarly bad. I was going to say that drunk driving is illegal, but texting while driving isn't illegal everywhere. But that doesn't really matter. Even if we were in a place where drunk driving was legal, that wouldn't make it a morally permissible thing to do. After all, the mere fact that something's legal doesn't imply it's not immoral. I'm not going to be legally sanctioned for calling Tommy a noodlebrain, but that doesn't mean I should do it.’

You think for a minute, your foot furiously tapping on the ground. ‘OK, I've got it. There is a relevant difference between texting while driving and drunk driving. The difference is that texting while driving is necessary in a way drunk driving is not. You don't need to get the pleasant feelings of drunkenness while you're driving. You could just wait until you get home. On the other hand, we do need to stay in touch with family and friends and communicate about our plans with each other. So texting while driving is necessary, while being drunk while driving is not.’

I grab your coat from you as you stand up to go over to the exam station. ‘That's an interesting one,’ I say, ‘because it does seem like it would be a relevant difference. But, the problem is, I don't think it's an actual difference. If it were really true that texting while driving were necessary to achieve really important goals – necessary despite the possible risks it posed to people – then that might show that there is a morally relevant difference between texting while driving and drunk driving. But it is doubtful that this is an actual difference between drunk driving and texting while driving. There may be times when texting while driving or drunk driving might be worth the risk: perhaps if you had to do them to save a few people who would otherwise definitely die. I'll let you imagine what those scenarios would be like. I think you'll agree that they would be so exceedingly rare that you and I are not at all likely to ever encounter them. In most of the cases where people do text while drive, it is not necessary: instead of texting while driving, they could text before driving, text after, or pull over for a bit to text. So, the point here is not that the difference cited is not morally relevant. The point is that the difference cited is not actually a difference that really exists between texting while driving and drunk driving. So this objection, too, seems faulty and fails to undermine the argument that texting while driving is wrong because it is relevantly similar to drunk driving.’

‘Yeah, that makes sense, dad, but I'm going to go take the test now. Hopefully I won't fail the exam and make all this discussion of texting while driving irrelevant for me!’

Later, as we walk to the car to go home, I hand you the keys. ‘Your reward for passing the exam is that you get to drive me home,’ I say. You try to hide your excitement.

On the drive home we're both a bit pensive and enjoying our newfound freedom – your freedom to drive and my freedom from driving. ‘We should probably finish our discussion, dad, because I want to figure this out before I have to explain the rule to my friends. I know they're not going to take “dad said so” as an answer if I tell them I can't text them while I'm driving. So I was thinking, what if you're in one of those new cars that drives for you or alerts you to possible collisions? Or what if you're only texting at stoplights or while parked in your car? Or what if we use one of those headsets to do hands-free texting?Footnote ⁶ Is that OK?’

‘That's changing the subject a bit. Our analogy was about texting while you're driving. So, we'd have to reconsider the analogy to see how it might apply to those other cases. But, if the evidence shows that texting in those ways puts people at similar levels of unnecessary risk, then the challenge would be the same: you'd still have to identify some relevant difference between them and drunk driving. Otherwise, we should treat them the same. Don't you think?’

‘I suppose so. OK, here's my last attempt at objecting to the analogy. Right now we're talking while I'm driving, and that seems to me to be perfectly permissible. So there's my own argument: talking to a passenger while driving is morally permissible, and it's no different from texting while driving. So, doesn't that show that texting while driving is permissible?’

‘That's a great objection, because you've given us another competing analogy to consider. To counter the original analogy, you've offered a new one. This gives us a helpful way to assess what we think about texting while driving. We can consider whether it is relevantly similar to drunk driving or to eating while driving or talking to a passenger while driving, or whatever. That would be a good way to make sure our views about texting while driving are justified and consistent. I suppose one thing we need is to see if they really are relevantly similar. Does talking to a passenger while driving impair to a similar degree your ability to drive safely? We probably can't figure that out from our armchairs – we've got to see what research is out there on that. I suppose that if talking to passengers did impair people to a similar degree, then we'd either have to treat it the same as drunk driving or find a relevant difference. Maybe it doesn't impair you as much. Talking to a passenger doesn't need to take your eyes from the road, and the passenger can help to alert you to potential dangers. I suspect that's the difference, but to see if it's an actual difference we need more than my speculation. So we should check out if there's evidence for that when we get home.’

We sit in comfortable silence for the rest of the drive, watching the sun sparkle through the colorful autumn leaves and listening to the news on the radio. A story comes on about a woman, named Brittany Maynard, who sought the help of a physician to end her life because of a particularly malignant form of brain cancer.Footnote ⁷ I break the silence with a question. ‘Do you remember when grandma and grandpa had to euthanize their dog?’

‘Yeah,’ you say. ‘It was horrible to watch them go through that. They loved Lulu so much. But I think I would have done the same: there was just no more point in all the suffering she was going through. She was such a sweet dog.’

‘That's got me thinking about assisted suicide and euthanasia of pets: if it's permissible to prevent further suffering in the latter, why not in the former?’Footnote ⁸