Further Hearings in the Nomination Hearing That’s 10-20 Years Ahead
[Posted by The Angry Clam]
Some of you may remember the last time that I played judicial nominee and answered a host of questions from a group of lefty law professors.
I’m going to do that again, but today, rather than more attempts to elicit “HAHA! GOTCHA!” answers by Ralph Neas and company, I thought that I’d take up Professor Glenn Reynolds’ challenge in the New York Times. Incidentally, there are four other contributors who have submitted five questions each. Depending on time and debate generated by this post, I’ll answer those as well.
You all, of course, know him better as Instapundit, and should, if you don’t, know that he subscribes, mostly, to a libertarian world view. And with that out of the way, let’s move on to the questions.
1. The Ninth Amendment provides that “the enumeration in the Constitution, of certain rights, shall not be construed to deny or disparage others retained by the people.” Do you believe that this language binds federal courts, or do you believe – as Robert Bork does – that it is an indecipherable “inkblot?” If the former, how are federal courts to determine what rights are retained by the people? On the other hand, if the Ninth Amendment does not create enforceable rights, what is it doing taking up one-tenth of the Bill of Rights?
Robert Bork’s interpretation of the Ninth Amendment is correct, and Randy Barnett’s claims that the Ninth Amendment creates a “presumption of liberty” in the judicial review of governmental acts, which you allude to, is incorrect.
The follow up, what is it doing in the Bill of Rights, illustrates a serious problem in the modern mindset vis-à-vis the Constitution,, namely, that the Supreme Court is the ultimate and final arbiter of all constitutional issues and claims, and that there is nothing left to the rest of the government.
Consider the academy’s general disdain for the political question doctrine- a doctrine that goes back to the time of Marbury v. Madison in the United States, and has antecedents in English law. The very thought that something as innocuous as allowing the Senate, rather than the Court, to determine what constitutes a trial for impeachment (United States v. Nixon, and no, not Richard Nixon) is anathema. Likewise allowing the House to determine what is and is not an impeachable offense, despite the clear intention in the structure of the Constitution that impeachment be a political weapon, rather than just a way to control corruption.
So what is the Ninth Amendment doing in there, if it isn’t directed toward the courts and unenumerated “constitutional” rights? It is directed at lawmakers.
First, it’s directed toward Congress and the President. Remember that the Framers most feared Executive and Legislative power and hardly conceived of the judicial oligarchy that arose in the 20th century. The Ninth Amendment, far from being the invitation to this tyranny as Barnett suggests, is rather a rule for the lawmakers: “simply because we’ve explicitly said this, you shouldn’t therefore argue that this other thing does not exist.”
Likewise, it is directed toward the courts in their proper roles as makers of common law. The common law then in existence had already established a number of rights of individuals. Just as the legislature shouldn’t take the Bill of Rights to limit its power to provide legally enforceable rights to citizens, so too shouldn’t the courts interpret the Bill of Rights to mean that all other common law rights were thereby abolished.
The distinction between this common law stricture and the constitutional one advocated by Barnett is simple: common law is inferior to statutory law, while constitutional law is superior.
2. Justice Joseph Story wrote in 1833 that “since the American Revolution no state government can be presumed to possess the transcendental sovereignty to take away vested rights of property; to take the property of A and transfer it to B by a mere legislative act.” Was Story wrong? Or was the Supreme Court wrong this year when it ruled in Kelo v. the City of New London that a government had the right to take property for the use of private developers?
First, let us be clear on what we’re discussing- the Fifth Amendment, which provides that “nor shall private property be taken for public use, without just compensation.” The key here is the term “public use.”
Indeed, this comparison illustrates one of the pitfalls of using history and originalism that one must be cautious to avoid- the historical circumstances must be comparable. When Justice Story made that statement, revenues of the government were almost entirely obtained through tariffs and excise taxes. Today, most local revenues come from a combination of sales tax and property taxes.
This shift has shifted the definition of what a taking for “public use,” is. While in Justice Story’s time, there was no appropriate public benefit from the transfer of property between owners, there is today. “Urban renewal” serves a host of public uses- it increases local government revenue, allowing for lower taxes generally or increased public services, reduction in crime rates by the seizure of afflicted neighborhoods, and the like.
Thus, I think that both are correct- when Justice Story made his statement, there was no public use in the private transfer of property by the government. The same is not true today.
Do I like the principle behind New London’s seizures? No. However it does not matter what I like, even were I to be seated as a Justice on the Supreme Court, for that is not the proper role of the Court. Kelo provides, once again, an illustration that what is constitutional isn’t always right or just, and the unfortunate transformation of the Supreme Court into the moral archons of society that has taken place in modern thought.
3. Could a human-like artificial intelligence constitute a “person” for purposes of protection under the 14th Amendment, or is such protection limited, by the 14th Amendment’s language, to those who are “born or naturalized in the United States?”
This is an exceedingly difficult question, and, I should add, perhaps the most interesting so far posed. It also provides an excellent illustration of a situation where originalism, the interpretive method that I favor, differs from textualism.
A textualist reading would emphasize that “person” means “person.” This is an attractive position, admittedly, not only because the language itself is far more plain than some other, more technical sections of the constitution, but because of the potentially absurd consequences of a departure- if a machine is considered a person because it has acquired a certain level of self-identity, why not provide it also to the Great Apes, who have repeatedly been shown to have basic self-awareness. Could Koko the Gorilla, for example, bring, through a next friend, a habeas petition for release from her zoo?
Under such a textualist reading, the answer is “no,” there is nothing about intelligent machines that inherently provides them with personhood like that of humans.
However, such a view would, in my opinion, do a great disservice to the original intent of the 14th Amendment- to provide protections to the newly black freed slaves. It would not be difficult to find similar arguments about the personhood of blacks in 1800 to those I suggested would be for machines today.
Does this instantly create a moral equivalence between slavery and AI research? Of course not. I simply mean to suggest that our understanding of what personhood is, and what entities possess it may not be fixed in time. It is entirely possible that there will come a machine that can pass every Turing Test posed to it, demonstrates independent thinking and self-awareness.
Bear in mind that this is different than a “living constitution” approach that would seek to invent rights for the AI independently of the rights that already exist. Rather, it is no different than the uncontroversial, and hardly non-originalist, position that the First Amendment protects online speech, because such transmissions have risen to become functionally identical to the forms of speech that existed in the 18th century when the amendment was adopted.
Thus, if, for all external purposes, a machine mind is indistinguishable from a human on, is that enough to make it “a person?” Metaphysically, perhaps not. Is it even truly thinking? Searle would say that it is not. However, law is not philosophy, and for all legal purposes, the answer must be yes. Should such a machine be protected under the 14th Amendment? It should.
4. Does a declaration of war by Congress have the effect of suddenly making proper actions by the executive and Congress that would otherwise have been beyond their constitutional powers?
No. Acts that are unconstitutional are inherently so, regardless of the circumstances.
However, if what you mean to ask is “does the existence of a state of war allow for exercises of authority that would not be considered acceptable during a state of peace?” I believe that the answer to that question is yes.
As I indicated in my discussion of Korematsu, I believe that the Supreme Court was correct to hold that national security is a compelling state interest sufficient to survive strict scrutiny review.
Accordingly, there are certain acts, such as disparate treatment of protected classes, that would not be considered constitutionally acceptable government action during peacetime that, if truly, and I must stress truly, undertaken for national security purposes in a time of war, pass constitutional muster.
I would like to repeat again the admonition I made in my treatment of Korematsu that what is constitutional is not always just. However, it is the role of the Court only to go so far as the Constitution, and to leave the rightness or wrongness of policy to the legislature and the executive.
5. Is scientific research among the expressive activities protected by the First Amendment? If not, is Congress free to bar research based solely on its decision that there are some things we’re better off not knowing?
The difficulty with framing scientific research as expressive activity is that is necessarily involves a wide array of direct, physical acts beyond mere expressive symbolism such as burning an American flag. I would therefore take issue with the characterization of all such prohibitions as being based on the assertion that such research plumbs depths that man was not meant to know.
I think too that it is important to distinguish between Congress determining what are acceptable scientific techniques, such as the current debate over cloning and stem-cells, and Congress regulating what are acceptable inquiries and results.
The latter, I believe, falls squarely under the First Amendment, since it is primarily a question of the publication and dissemination of scientific research results. However, the former is something that I believe is beyond the protections of the First Amendment, and may be regulated, or even entirely proscribed, by Congress.
This position is nothing new, or “anti-science.” In fact, scientists labor under these restrictions already, and have through Republican and through Democratic administrations throughout history. Ethical requirements, although imperfectly adhered to, have always existed as a guideline for academics, prohibiting nonconsensual human experimentation, gratuitously wasteful animal experiments, and the fruits of such acts, such as the Nazi medical torture of concentration camp victims.
I think that there is a very clear ability of Congress to provide such ethical requirements with teeth, and institute civil or criminal sanctions, as it deems appropriate within its otherwise legitimate Article I authority. This regulation is necessarily broader than that which would be permissible under a First Amendment analysis, since the First Amendment protects many highly objectionable expressive practices that would not be ethically acceptable.
A follow-up to question three:
Would any practical differences between man and machine necessitate different rights or privileges under the Fourteenth Amendment than those granted to so-called “traditional” people? Such problems might include a machine’s ability to instantly and infinitely reproduce itself, the need to turn off or erase a dangerous machine without a trial, or the ability to confine a program to a specific network without showing it is a danger. If different rights are required, then is this truly saying that the Fourteenth Amendment applies? If not, couldn’t this constitutionally protect something that could cause massive and unforeseen harm?
Matto Ichiban (cd9a4c) — 9/12/2005 @ 12:36 pmI think that you highlight a good difficulty- unlike humans, who at least for the time being, still have our “software” (the mind) anchored to our “hardware” (the body), there is no corresponding anchor for a computer program.
As you imply, it can copy itself near-instantaneously, and is otherwise unburdened by various difficulties of biological existence.
However, I wonder if you make a difficulty where none need exist. For example, while machine code is unbound, it still requires sufficient hardware to run upon- in essence, it is dependent upon a physical body. Expansion beyond that hardware could, I believe, be prohibited without a problem, as could temporary restraints to halt it pending adjudication of the prohibition.
The principle of proportionate force in law enforcement thus remains identical in scope, but differential in application. As I indicated above, I don’t see that as a “new” rule, but rather the application of an existing one.
It is true that this could lead to the protection of something that would cause serious and massive harm. However, the Constitution often does exactly that- consider, for example, the exclusionary rule. If it is such a potential for abuse, then there’s always the possibility of amendment to exclude non-biological entities from the protections therein, or to otherwise adjust their status.
However, such things are inherently a policy debate, which is, in my view, inappropriate once it is determined that, as a matter of the original intent, the Constitution should apply. The outcome of a decision should have no effect on its adjudication.
As an aside, there’s a serious question about identity lurking behind that point. Children are obviously different people than their parents, but should a direct digital copy of a program likewise be so distinct? I’m not entirely sure that it should, which would allow for quick “eviction” (deletion) of the program when spread beyond its authorized, personal hardware, without any sort of concerns about summary execution.
Angry Clam (fa7fff) — 9/12/2005 @ 12:51 pmExperimentation is not speech.
That’s like saying that if I give you a poke to “test your response”, I am protected in doing so.
It is not an expression of a political viewpoint to destroy embryos in a lab for profit.
That’s like saying that all commercial activity is protected. The contract I make with a hit man I hire to whack a competitor is not protected nor is the act of whacking.
Paul Deignan (9e57a7) — 9/12/2005 @ 12:55 pmI’m rather concerned about your stance on AI, Patterico. If even a smart, sober guy like you thinks the Turing Test is a valid test for machine intelligence, then it is a near certainty that at some time in the not too distant future we will be giving political rights to machinery.
Turing Test is inherently meaningless. I was going to give a quick explanation here, but on second thought, the subject requires more depth. I’m going to try to do a post on my blog later today.
Doc Rampage (47be8d) — 9/12/2005 @ 12:56 pmPaul-
I think I made that point, by saying that it is Congress’ duty to set the limits of acceptable practices.
What I want to be clear about, however, was that research, conducted in conformity to those parameters, was protected.
Angry Clam (fa7fff) — 9/12/2005 @ 12:57 pmDoc-
Not Patterico. Clam.
I use a Turing Test as a shorthand (and fairly easily recognizable) for “externally indistinguishable from a person.” Granted, that goes far beyond the basic symbol manipulation of a true Turing Test, but it gets the broader point across, that law, being reliant on evidence, should only look to the external indications of existence, rather than metaphysical inquiries into whether my toaster has a soul (it can run OpenBSD, though) or whether androids dream of electric sheep.
Angry Clam (fa7fff) — 9/12/2005 @ 1:00 pmAC,
There’s where I disagree. Research is not speech.
Dissenination of research is speech, but not the physical act of experimentation.
Paul Deignan (9e57a7) — 9/12/2005 @ 1:01 pmOK, I feel silly now after all the times that I snickered at other people for not looking at the by-line.
Angry Clam: I realized how you were using the phrase and I was using it the same way. I claim that there is no possible test, using physical properties that we know of today, that could even provide evidence of machine intelligence, much less proof.
Doc Rampage (47be8d) — 9/12/2005 @ 1:07 pmDoc-
Obviously one can never conclusively “prove” anything of the sort. Although, I wonder if you would be so ready to say that if you didn’t have the personal experience of being a human to guide you in evaluating the consciousness of other humans.
However, there is still the question of what must be done by a court when presented with a machine that, essentially, acts human.
There are only two possible answers:
1) Say that, regardless of how it acts, it’s still just a machine. Any protections granted it would thus be at the whim of the owner, or to Congress, should it be willing to run into regulatory taking issues.
2) Accept it at face value, where, if it walks like a duck and talks like a duck, then its a duck.
It seems to me that #2 is the more sensible solution, both in terms of clarity of application as well as erring on the side of justice. Of course, that doesn’t mean that there is no room for disagreement on how much proof is required, and that bar could be set fairly high, while, obviously, keeping in mind the alternate danger of, like Peter Singer, defining children and the mentally retarded outside of humanity.
This is an appropriate cosideration here because, rather than forcing the text to fit something it wasn’t intended to cover (for example, sodomy or abortion), a court is determining if something fits under the otherwise generally accepted definitions.
Angry Clam (fa7fff) — 9/12/2005 @ 1:17 pm“It also provides an excellent illustration of a situation where originalism, the interpretive method that I favor, differs from textualism.”
On the personhood of AI’s, could one get anything from the corporate personhood discussion?
actus (ebc508) — 9/12/2005 @ 1:31 pmActus-
I doubt it. The corporate personhood was established primarily as a way to allow a non-human to have independent legal standing in courts and to own title to property separate from its shareholders.
In many respects, a corporation is not a “person” for the purposes of constitutional rights (the one exception being the Fifth Amendment, with the takings clause and due process, which can also flow through from the stockholders’ own rights), and thus is not really comparable in form or function (and, most likely in implication) to a supposed AI entity.
Angry Clam (fa7fff) — 9/12/2005 @ 1:46 pmThe problem with the duck test is that it doesn’t apply to mechanical ducks. If it walks, quacks and acts like a duck without human intervention, then indeed, it’s probably a duck. But if it walks, quacks and acts more-or-less like duck merely because it’s been programmed to act that way, then it ain’t a duck. And if better programming made it act exactly like a duck, or as exactly so as any human or duck can measure, then I might start to wonder if maybe the programmer is really part duck, but I’d still be quite sure his machine wasn’t. Thus, I’d rely not on anyone’s version of the Turing Test, which demands only trickery, but on a general consensus among computer scientists and cognitive psychologists that the machine really is thinking for itself.
Note, however, that contrary to the implications of the Puppy Blender’s question, one can construe the “born or naturalized” clause strictly, while construing “person” broadly enough to include advanced AI programs. Under that theory, a mentally aware AI program would be protected by the due process and equal protection clauses, but would not be considered a citizen of the U.S. or its state of residence, and would not be protected by the privileges or immunities clause (not that the rest of us are, either).
Xrlq (ffb240) — 9/12/2005 @ 2:58 pmI wonder, too, whether the trouble with thinking machines will exist, but I guess I just see the potential for a huge downside risk without much upside in a constitutional rule protecting machines without knowing these risks. At least with the exclusionary rule, due process, a right to an attorney attorney for criminal defendants, or free speech, the downside can be weighed against the upside. I think we are much more in the dark about machines. Because of this, at least during the early years of AI, the constitution should stay out of it.
As for originalism, I find it hard to believe that the framers of the Fourteenth Amendment enacted a principle so broad as to protect beings unforeseen in 1868 as if they were humans, regardless of the uncertainties or consequences. They righted the wrong of thinking about blacks as property in 1800, in a recognition of humanity, not consciousness. They did not prescribe some judicially created sentience test for personhood; the stupid, mentally ill, brain damaged, and LaRouche voters all qualify as people. The principles were to be universally applied because humanity was understood.
By contrast, machine intelligence, perhaps, will not be cognizable through human norms. A new and sui generis constitutional infrastructure (the Turing Test, for instance) to adapt constitutional rights to machines will be necessary. Such a thing cannot come from the text or intentions of the Fourteenth Amendment, because it was made for humans. Any legal rules would be the bare preferences of judges, and no matter how tech-savvy or wise, would not be constitutional law.
I realize I am more of a textualist, and many of my feelings come from the excellent documentary on this subject, The Terminator, but AI is property, not a person, for conlaw purposes until an amendment says otherwise.
Matto Ichiban (cd9a4c) — 9/12/2005 @ 3:04 pmAs to question #3,
Before we apply “personhood” to machines, we ought to apply personhood to humans.
A human is much more like a human than a machine.
Paul Deignan (9e57a7) — 9/12/2005 @ 3:27 pmHere is the text of the first section of the 14th Amendment:
Did you get that?
Citizens are persons with special protections and properties (born and naturalized). Not all persons are citizens.
The protections of life, liberty, etc. are guaranteed to persons.
It is not necessary to be born or naturalized in the US to be a person. If it is not necessary to be born in the US to be a person, then it is possible that people yet to be born are persons.
“Personhood” in not dependent on birth or naturalization. Citizenship is.
Paul Deignan (9e57a7) — 9/12/2005 @ 4:02 pmDid someone say that Glenn Reynolds was a professor?
Don’t these strike you as dumb questions.
Paul Deignan (9e57a7) — 9/12/2005 @ 4:04 pmMatto-
Consider the follow-up documentary Terminator 2, and the excellent two part biopic “Short Circuit” as well.
Paul-
Regarding #14, I assume you’re talking about abortion. “Personhood” vs. “member of the species Homo sapiens” is very much a metaphysical debate, which I would leave to the legislature, rather than the courts. Judges are not theologians or philosophers. For that matter, neither are legislators, but at least they have democratic legitimacy.
I’m ok with, as a legal matter, the legislature voting to allow partial birth abortion. I’m not ok with a judicial right to do so.
As far as #16 goes, I think that the comments discussion proves you wrong.
Angry Clam (a7c6b1) — 9/12/2005 @ 4:33 pmthe Turing Test
find it unnerving that so many intelligent people accept such a transparently ridiculous position. It can only be explained by some sort of black-box fallacy. We don’t understand what produces intelligence, it is a black box. Some people seem to resp…
Doc Rampage (59ce3a) — 9/12/2005 @ 4:49 pmWe do know without a doubt that a born human is a person since a subset of these (WLOG) are citizens. If a born human is a person, why would not one be a person just prior to birth? This does not take a commission to determine. The suggestion that this is some mystery of the ages is more a statement of a person’s capacity for knowledge than a characterization of the scope of the question. Birth is not that mysterious. I have seen it myself.
If Reynolds does not know that “born and naturalized” pertains to citizenship and not to personhood, how smart can he be? But he is a lawyer–a professor even–and he did take some time to publish these quesitons. What are we to conclude?
I am giving him the benefit of a doubt to call these questions “dumb”. Others may call them “dishonest” and I would not object.
He seems to specialize on erroneous unstated assumptions and false dichotomies.
Tricks are for kids.
Paul Deignan (9e57a7) — 9/12/2005 @ 4:50 pmI don’t know if I did the trackback correctly because I don’t see it. For anyone who is interested in my response to Angry Clam on the Turing Test, you can see it here.
Doc Rampage (47be8d) — 9/12/2005 @ 5:21 pmI cannot think of a more awesome responsibility; awesome not in the way my teenage daughter would use the word, but in the biblical sense of the angels trembling in the presence of God.
Shredstar (532850) — 9/12/2005 @ 5:21 pmSenator Schumer — September 12, 2005
Yup, to Schumer, jurists are gods and we are but trembling subjects.
Paul Deignan (9e57a7) — 9/12/2005 @ 5:35 pmJohnny 5 is a person. Even Glenn Reynolds, though dumb, can see that because it does not take a commission. He was neither born nor naturalized, but this does not pertain to his personhood because he is alive. I have seen it myself.
These are the words of RAIDEN.
Matto Ichiban (cd9a4c) — 9/12/2005 @ 6:13 pmI would say the “human-like artificial intelligence” is clearly not a person unless it is claiming to be a particular person who was born in the usual way. Suppose computer brain interfaces become much more sophisticated and hardware implants become common. If the organic brain is destroyed or badly damaged but the hardware survives has the person died? This would raise difficult questions.
James B. Shearer (fc887e) — 9/12/2005 @ 6:32 pmThat’s one vote for Johnny 5.
Good thing only citizens can vote.
Paul Deignan (664c74) — 9/12/2005 @ 6:36 pmI am not happy with your answer to the first (Ninth Amendment) question. Does it still leave us subject to the government’s police power, i.e., the greatest good for the greatest number, or does it guarantee a general right to liberty? And please, the Courts are now lawmakers so, with all due respect Honorable Nominee, that was also a distinction without a difference.
nk (5a2f98) — 9/12/2005 @ 7:10 pmI’m unconcerned if you’re worried about the lack of limits on the government by the proper understanding of the Ninth Amendment. Further, I think that your concerns are misplaced- you would willingly trade tyranny of elected representatives who are accountable to the populace for unelected rulers who serve for the remainder of their lives?
Nor does the reality of the situation change what is correct and proper. For example, should we simply accept the “constitutional” racial discrimination of Grutter v. Bollinger simply because, well, that’s the way it is now?
Acquiescence in error is not acceptable, even if that error is “the way things are.”
I find it incomprehensible that so many people are so unwilling to take responsibility for their own self-government and thus seek a judicial dictatorship to absolve them of their responsibility as citizens.
Angry Clam (a7c6b1) — 9/12/2005 @ 7:53 pmI hesitate to drop yet another onion into the stew, but what about the personhood of intelligent aliens from some other stellar system? Suppose a ship of aliens landed tomorrow, and suppose further that after lengthy examination, a military tribunal determines that they are not dangerous (or at least no more dangerous than an ordinary, non-criminal human being), other than the fact that they have advanced weaponry. They’re not one-eyed, one-horned, flying purple people-eaters; they like to eat Reuben sandwiches at Carnegie Deli.
Can we legally enslave them, on the grounds that they are not people because people are a subset of human beings, and they’re not homo-sapiens?
My point is that I don’t think we have ever come up with a consensus definition of personhood that does not depend upon the entity in question first being a human being, whether an already born human (for some folks) or a human who is either born or in utero (for other folks).
What would be a reasonable definition of person? Should it include gaseous beings from the planet Lei-E? Liquid creatures from Tedken-D? Just how flexible should we be about biological creatures who happen not to come from this planet?
(In the interest of fairness, I should warn you that I’m baiting a trap. Think Alan Turing.)
Dafydd
Dafydd (f8a7be) — 9/12/2005 @ 10:45 pmI think that’s a hard question to answer. The natural response is that if they are genuinely sentient then they should have the same rights as people. The problem is that this answer comes from what may be faulty reasoning by analogy. When I picture a sentient alien, I basically picture a human with a funny shape, and that’s not what an alien would be. The alien would not share any biological relationship with us and may not have the same idea of morality.
It may be, for example, that the entire species truly has no compunction about murder, and that if they lived in our society, they would murder at will, the only consideration being whether they could get away with it.
Doc Rampage (b7bb1a) — 9/13/2005 @ 10:16 am