On the Line between Technology Ethics vs. Technology Policy

by on August 1, 2013 · 6 comments

10 commandmentsWhat works well as an ethical directive might not work equally well as a policy prescription. Stated differently, what one ought to do it certain situations should not always be synonymous with what they must do by force of law.

I’m going to relate this lesson to tech policy debates in a moment, but let’s first think of an example of how this lesson applies more generally. Consider the Ten Commandments. Some of them make excellent ethical guidelines (especially the stuff about not coveting neighbor’s house, wife, or possessions). But most of us would agree that, in a free and tolerant society, only two of the Ten Commandments make good law: Thou shalt not kill and Thou shalt not steal.

In other words, not every sin should be a crime. Perhaps some should be; but most should not. Taking this out of the realm of religion and into the world of moral philosophy, we can apply the lesson more generally as: Not every wise ethical principle makes for wise public policy.

Before I get accused of being accused of being some sort of nihilist, I want to be clear that I am absolutely not saying that ethics should never have a bearing on policy. Obviously, all political theory is, at some level, reducible to ethical precepts. My own political philosophy is strongly rooted in the Millian harm principle (“The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.”) Not everyone will agree will Mill’s principle, but I would hope most of us could agree that, if we hope to preserve a free and open society, we simply cannot convert every ethical directive into a legal directive or else the scope of human freedom will need to shrink precipitously.

Can We Plan for Every “Bad Butterfly-Effect”?

Anyway, what got me thinking about all this and it its applicability to technology policy was an interesting Wired essay by Patrick Lin entitled, “The Ethics of Saving Lives With Autonomous Cars Are Far Murkier Than You Think.” Lin is the director of the Ethics + Emerging Sciences Group at California Polytechnic State University in San Luis Obispo and lead editor of Robot Ethics (MIT Press, 2012). So, this a man who has obviously done a lot of thinking about the potential ethical challenges presented by the growing ubiquity of robots and autonomous vehicles in society. (His column makes for particularly fun reading if you’ve ever spent time pondering Asimov’s “Laws of Robotics.”)

Lin walks through various hypothetical scenarios regarding the future of autonomous vehicles and discusses the ethical trade-offs at work here. He asks a number of questions about a future of robotic cars and encourages us to give some thoughtful deliberation to the benefits and potential costs of autonomous vehicles. I will not comment here on all the specific issues that lead Lin to question whether they are worth it; instead I want to focus on Lin’s ultimate conclusion.

I commenting on the potential risks and trade-offs, Lin notes:

The introduction of any new technology changes the lives of future people. We know it as the “butterfly effect” or chaos theory: Anything we do could start a chain-reaction of other effects that result in actual harm (or benefit) to some persons somewhere on the planet.

That’s self-evident, of course, but what of it? How should that truism influence tech ethics and/or tech policy? Here are Lin’s thoughts:

For us humans, those effects are impossible to precisely predict, and therefore it is impractical to worry about those effects too much. It would be absurdly paralyzing to follow an ethical principle that we ought to stand down on any action that could have bad butterfly-effects, as any action or inaction could have negative unforeseen and unintended consequences.

But … we can foresee the general disruptive effects of a new technology, especially the nearer-term ones, and we should therefore mitigate them. The butterfly-effect doesn’t release us from the responsibility of anticipating and addressing problems the best we can.

As we rush into our technological future, don’t think of these sorts of issues as roadblocks, but as a sensible yellow light — telling us to look carefully both ways before we cross an ethical intersection.

Lin makes some important points here, but these closing comments (and his article more generally) have a whiff of “precautionary principle” thinking to it that makes me more than a bit uncomfortable. The precautionary principle generally holds that, because a new idea or technology could pose some theoretical danger or risk in the future, public policies should control or limit the development of such innovations until their creators can prove that they won’t cause any harms. Before we walk down that precautionary path, we need to consider the consequences.

The Problem with Precaution

I have spent a great deal of time writing about the dangers of precautionary principle thinking in my recent articles and essays, including my recent law review article, “Technopanics, Threat Inflation, and the Danger of an Information Technology Precautionary Principle,” as well as in two lengthy blog posts asking the questions, “Who Really Believes in ‘Permissionless Innovation’?” and “What Does It Mean to ‘Have a Conversation’ about a New Technology?”

The key point I try to get across in those essays is that letting such precautionary thinking guide policy poses a serious threat to technological progress, economic entrepreneurialism, social adaptation, and long-run prosperity. If public policy is guided at every turn by the precautionary principle, technological innovation is impossible because of fear of the unknown; hypothetical worst-case scenarios trump all other considerations. Social learning and economic opportunities become far less likely, perhaps even impossible, under such a regime. In practical terms, it means fewer services, lower quality goods, higher prices, diminished economic growth, and a decline in the overall standard of living.

In Lin’s essay, we see some precautionary reasoning at work when he argues that “we can foresee the general disruptive effects of a new technology, especially the nearer-term ones, and we should therefore mitigate them” and that we have “responsibility [for] anticipating and addressing problems the best we can.”

To be fair, Lin caveats this by first noting that precise effects are “impossible to predict” and, therefore, that “It would be absurdly paralyzing to follow an ethical principle that we ought to stand down on any action that could have bad butterfly-effects, as any action or inaction could have negative unforeseen and unintended consequences.” Second, as it relates to general effects, he says we should just be “addressing problems the best we can.”

Despite those caveats, I continue to have serious concerns about the potential blurring of ethics and law here. The most obvious question I would have for Lin is: Who is the “we” in this construct?  Is it “we” as individuals and institutions interacting throughout society freely and spontaneously, or is it “we” as in the government imposing precautionary thinking through top-down public policy?

I can imagine plenty of scenarios in which a certain amount of precautionary thinking may be entirely appropriate if applied as an informal ethical norm at the individual, household, organizational or even societal level, but which would not be as sensible if applied as a policy prescription. For example, parents should take steps to shield their kids from truly offensive and hateful material on the Internet before they are mature enough to understand the ramifications of it. But that doesn’t mean it would be wise to enshrine the same principle into law in the form of censorship.

Similarly, there are plenty of smart privacy and security norms that organizations should practice that need not be forced on them by law, especially since such mandates would have serious costs if mandated. For example, I think that organizations should feel a strong obligation to safeguard user data and avoid privacy and security screw-ups. I’d like to see more organizations using encryption wherever they can in their systems and also delete unnecessary data whenever possible. But, for a variety of reasons, I do not believe any of these things should be mandated through law or regulation.

Don’t Foreclose Experimentation

While Lin rightly acknowledges the “negative unforeseen and unintended consequences” of preemptive policy action to address precise concerns, he does not unpack the full ramifications of those unseen consequences. Nor does he answer how the royal “we” separate the “precise” from the “general” concerns? (For example, are the specific issues I just raised in the preceding paragraphs “precise” or “general”? What’s the line between the two?)

But I have a bigger concern with Lin’s argument, as well with the field of technology ethics more generally: We rarely hear much discussion of the benefits associated with the ongoing process of trial-and-error experimentation and, more importantly, the benefits of failure and what we learn — both individually and collectively — from the mistakes we inevitably make.

The problem with regulatory systems is that they are permission-based. They focus on preemptive remedies that aim to forecast the future, and future mistakes (i.e., Lin’s “butterfly effects”) in particular.  Worse yet, administrative regulation generally preempts or prohibits the beneficial experiments that yield new and better ways of doing things — including what we learn from failed efforts at doing things. But we will never discover better ways of doing things unless the process of evolutionary, experimental change is allowed to continue. We need to keep trying and failing in order to learn how we can move forward. As Samuel Beckett once counseled: “Ever tried. Ever failed. No matter. Try Again. Fail again. Fail better.” Real wisdom is born of experience, especially mistakes we make along the way.

This is why I feel so passionately about drawing a distinction between ethical norms and public policy pronouncements.  Law forecloses. It is inflexible. It does not adapt as efficiently or rapidly as social norms do. Ethics and norms provide guidance but also leave plenty of breathing room for ongoing experimentation, and they are refined continuously and in response to ongoing social developments.

It is worth noting that ethics evolve, too. There is a sort of ethical trial-and-error that occurs in society over time as new developments challenge, and then change, old ethical norms. This is another reason we want to be careful about enshrining norms into law.

Thus, policymakers should not be imposing prospective restrictions on new innovations without clear evidence of actual, not merely hypothesized, harm. That’s especially the case since, more often than not, human adapt to new technologies and find creative ways to assimilate even the most disruptive innovations into their lives. We cannot possibly plan for all the “bad butterfly-effects” that might occur, and attempts to do so will result in significant sacrifices in terms of social and economic liberty.

The burden of proof should be on those who advocate preemptive restrictions on technological innovation to show why freedom to tinker and experiment must be foreclosed by policy. There should exist the strongest presumption that the freedom to innovate and experiment will advance human welfare and teach us new and better ways of doing things to overcome most of those “bad butterfly-effects” over time.

So, in closing, let us yield at Lin’s “sensible yellow light — telling us to look carefully both ways before we cross an ethical intersection.” But let us not be cowed into an irrational fear of an unknown and ultimately unknowable future. And let us not be tempted to try to plan for every potential pitfall through preemptive policy prescriptions, lest progress and prosperity get sacrificed as a result of such hubris.

  • Patrick_Lin

    Hi Adam, thanks for such thoughtful commentary on my Wired essay!

    As a quick reply, I agree that there is a big difference between ethics and policy (as well as law). I wrote another essay earlier in the year that draws out the stark tensions among the three: http://www.theatlantic.com/technology/archive/2013/04/pain-rays-and-robot-swarms-the-radical-new-war-games-the-dod-plays/274965/

    I won’t comment much on the precautionary vs. proactionary principle, since that requires more time than I have right now. Anyway, I’m not committed to either: I think both approaches or principles make sense, depending on the circumstances.

    But I just wanted to add that, while “not every wise ethical principle makes for wise public policy” (probably true, though it’s unclear to me that we should prefer wise policy to doing the right thing), ideally we would create policy that is ethical, unless there are extraordinary reasons not to.

    Also, I’d take a broader view of the law. It’s not just that “law forecloses”, but it sometimes empowers, e.g., marriage laws, contract law, etc. My anarchist friends may disagree, but I’d say that laws are necessary and be agents of positive change, if used wisely.

    This is to say that there ought to be an appropriate balance among ethics, policy, and law. It’s probably not the case that we should give any one of these default priority.

    Thanks again for your thoughts, and hopefully I’ll have time later to engage with you more if you’d like.

  • http://www.techliberation.com Adam Thierer

    Thanks Patrick. You make some fair points here. Of course, you are correct in noting that “there ought to be an appropriate balance among ethics, policy, and law,” but one thing left unsaid (for the most part) in both your original essay and my response is what, exactly, our ideal ethical systems/norms look like. I mentioned my adoration of the Millian harm principle and could have spent an entire essay building on that and what it means for the broader debate here, but I didn’t. Had I done so, I suspect it would have made for a far more interesting, but controversial piece (since that approach to ethics/policy has lost a lot of sway in philosophical circles over the past century.) Anyway, let the debate continue!

  • Patrick_Lin

    It really does come down to one’s political philosophy, doesn’t it? I suppose that I’d fall somewhere in between JS Mill and John Rawls with respect to political liberalism; but I tend to agree more with Rawls that utilitarianism is insufficient by itself, and agree with Rawls’ concern for fairness, which Mill ignores.

    But neither theory is perfect, and Mill says some wacky things that I wonder whether you really do agree with: he’s both too permissive and too restrictive. For instance, he’s happy to let us waste away our lives with hard drugs, lose all our money gambling, etc. as long as no one else is hurt. But he also says gov’t can force people to work (to support their families), can flat-out ban public sex (as an unexplained “offense to decency”, which is weird given his opposition to censorship), can prohibit two people from marrying and having kids (if they’re too poor to support a family), can prohibit suicide (again, puzzling given that the harm principle is focused entirely on harm to others; it’s probably another holdover of Puritanical values, which don’t fit well with his theory), and other apparently contradictory things.

    Anyway, not sure the harm principle applies to our debate, since we’re talking more about economic policy, not political rights. And Mill even cautions us to not confuse the harm principle with his laissez-faire principle, the latter of which allows for gov’t intervention for the public good, e.g., regulations about worker safety, sanitation, etc. In my Wired essay, I’m not suggesting additional laws or regulations to govern new technologies, just some foresight and sophistication about thinking about technology’s impact. But even if I were, that could still be consistent with Mill’s laissez-faire principle, given that he’s also interested in the protection of the public.

    Finally, as a utilitarian, Mill would in theory be ok with anything, as long as the math works out (including torture). This would seem to include a precautionary approach to technology policy: if the possible consequences are catastrophic (not just having some bad effects, but really really bad effects that are unacceptable despite any redeeming effects), then we should hit pause and think about it, working more to avoid those consequences.

    That said, Mill does get a lot of things right, so it’s easy to see why he’s such an important figure in political philosophy. (I’m sure I don’t need to tell you all this, since you seem to be more of a Millian scholar than I am.)

    Thanks for the morning mind-workout!

  • http://www.techliberation.com Adam Thierer

    Well, without getting too bogged down in a debate about
    Mill, I think it’s important to separate the man from the particular principle
    with which he is most widely associated. I certainly am not about to defend
    every position Mill took (especially some of those found in “Principles of
    Political Economy”), many of which contradicted his eloquent principle.

    Nonetheless, I persist in my belief that Mill’s “one very
    simple principle” for “the dealings of society with the individual in the way
    of compulsion and control, whether the means used be physical force in the form
    of legal penalties, or the moral coercion of public opinion” remains an
    excellent baseline for deliberations over these or other matters.

    In a short essay here four years ago celebrating the 150th
    anniversary of “On Liberty,” [http://techliberation.com/2009/07/10/mills-on-liberty-at-150-its-legacy-for-freedom-of-speech-expression/]
    I argued that his book and the harm principle in particular, “remains a
    beautiful articulation of the core principles of human liberty and a just
    society.” Of course, any “simple” principle will almost by definition be too
    simple and not provide clear to all of life’s challenging questions — ethically
    or politically. But, again, I would argue Mill’s harm principle provides us
    with the right baseline to begin such deliberations by making it clear that
    humans should be at liberty to live a life of their own choosing, so long as
    they do not bring harm to others in doing so. [How we define “harm” in particular tech
    policy contexts is, of course, another tension here.]

    Of course, what is so interesting about Mill is the way –
    and this is reflected in the way we both express a certain amount of adoration
    for him above — his work influences almost all modern strands of liberalism,
    broadly defined. Consequently, disciples of Rawls (like you?) and Nozick (like
    me) can ultimately find much to appreciate and build upon when thinking through
    how Mill’s simple principle impacts rights, responsibilities, and the contours
    of justice.

    Alas, at some point, we will part ways, and walk down two
    separate paths based upon our respective preoccupation with fairness vs.
    liberty as our ethical and political prime directives. The same things that
    tore Rawls and Nozick apart will likely tear us apart when debating modern
    technology ethics and policy issues.

    In light of that, you will probably not be surprised to hear
    me note that I disagree with your assertion that you are “not sure the harm
    principle applies to our debate, since we’re talking more about economic
    policy, not political rights.” Of course, in a Nozickian and modern libertarian
    theory sense, there is simply no separating the two. They are fundamentally
    intertwined.

    By extension, I have above applied such reasoning to debates
    over technology policy in this and many other earlier essays. This is what
    leads you and others to call into question my allegiance to the harm principle
    as the over-arching operational baseline for all technology policy discussions.
    Whereas we are probably in accord that the principle should be the default baseline
    for discussions about social / political rights, we would likely part ways in
    thinking that the principle should extend to most matters of economic and
    technological liberty, too.

    Is that a fair summation of our differences?

  • Patrick_Lin

    I wouldn’t say that the harm or laissez-faire principle doesn’t extend to most matters of economic and technological liberty, but only that there should be a presumptive priority for political liberty, balanced with social fairness.

    Rawls too placed a higher priority on political rights than economic ones…though I wouldn’t be so dogmatic about this: e.g., if a society were starving, then jobs and innovation may become more important than free speech, etc. So economic policy may be more important for a nation like China or DPRK (in the opinion of many of their citizens), but not the US. We’ve already achieved a decent standard of living and therefore can focus on the things that enable us to be better human beings, not just on basic survival.

    I agree that political and economic liberties can be intertwined, but don’t see why this is necessarily so. Even Mill makes the distinction with his harm principle vs. laissez-faire principle warning,

    Not sure you want to compare our debate to Rawls vs. Nozick…because Rawls won! Nozick ultimately abandoned the libertarianism he inspired and started caring about similar things as Rawls did, i.e., social justice. ;)

    On Nozick’s path to the Dark Side: http://www.slate.com/articles/arts/the_dilettante/2011/06/the_liberty_scam.single.html

  • Pingback: Planning for Hypothetical Horribles in Tech Policy Debates « Internet Freedom Coalition

Previous post:

Next post: