The Anal Philosopher makes the provocative assertion that with regard to denying legal recognition for gay marriage, "the proper analogy is not to slavery, which is obviously unjust, but to not allowing dogs to vote, which is just." He cites a correct principle -- that "justice requires that likes be treated alike" -- and claims that heterosexual couples and homosexual couples are not "similarly situated" with regard to marriage. In deciding what counts as "similarly situated", it is useful to pursue casuistry -- the practice in philosophy of ethics where various cases and analogies are explored to better understand the underlying principles. Often extreme analogies or outrageous hypotheticals help to demonstrate a principle, which can then be extended to actual live controversies. The notion of canine suffrage is such an extreme analogy. Its very extremity provides a good starting point, because denying canine suffrage is something we can all agree upon, and casuistry must start from well-agreed cases. Thus, we recognize that even regarding fundamental rights such as voting or marriage, it is just and reasonable for the law to make some forms of classification, such as distinguishing species as humans and dogs. We will also generally accept that the law is right to classify persons as adult and minor persons, excluding minors from many of the rights of citizens, such as voting and marriage.
It is useful at this juncture to outline the principles of American jurisprudence on when it is appropriate to classify "unlike" situations. The law may classify based on a variety of personal attributes, such as age, citizenship, gender, and (as already noted) species. However, any classification is by default suspect in the law because of the principle of equal rights, treating likes alike. In considering whether some proposed classification is legal, one considers three things: (1) what is being denied, (2) who is being denied, and (3) for what purpose are some people being denied. In the first consideration -- what is being denied -- we must consider how important is the right that is being denied. Some rights, such as the right to vote or the right to marry, are considered to be fundamental rights, essential to our life and liberty, and thus we scrutinize more closely any infringements on these rights. Other rights, such as the right to park one's car overnight on a particular street, are not fundamental, and thus the law would reasonably tolerate some forms of discrimination in regard to such rights (e.g., only people who live on a street might get to park their cars overnight on that street) which would be intolerable for fundamental rights (e.g., it would be intolerable to say that people who lived on Elm Street could vote while people who live on Poplar Street could not). In the second consideration -- who is being denied -- we must consider the kind of discrimination being made. Is it based on age or gender or race or which street you live on? Here, certain kinds of discrimination are considered "suspect classes" (such as race or religion), or have explicit protections in the law (such as disabilities or veteran status). Discrimination on such characteristics has an extremely high presumption of being illegal, while other forms of discrimination (e.g., age or marital status) are not scrutinized quite as closely. In the third consideration -- the purpose of classification -- we examine what a proposed law is trying to accomplish by its classification, and how well the classification serves the purpose. Where fundamental rights are being abridged, or where suspect classes are being discriminated against, the principle is to apply "strict scrutiny" to see if there is a "compelling state interest" being served and whether the classification is required and narrowly tailored to serve that interest. In other cases, lesser scrutiny is applied to see whether there is some "rational basis" for the law, and whether the classification is in some way reasonably connected to the purpose. (There is always some scrutiny to be applied, since purely arbitrary discrimination by the law is never tolerable.)
Now let us return to marriage. In the first consideration -- what is being denied -- there is well-settled case law that marriage is a fundamental right. (In Zablocki v. Redhail, for example, a long summary of the case history is cited, showing that the US Supreme Court "has long recognized that freedom of personal choice in matters of marriage and family life is one of the liberties protected by the Due Process Clause of the Fourteenth Amendment.") The second consideration -- who is being denied -- is where it becomes tricky to find the right analogy. Some have argued that prohibition of same-sex marriage amounts to gender-based discrimination. The claim is that male person A is being denied his right to marry male person B, solely because person A is male, and A would otherwise be allowed to marry B if A were female. Thus, the argument goes, person A is being classified based on gender. This argument was successfully made in the Hawaii Supreme Court decision in Baehr v. Levin, relying on the state's Equal Rights clause. (That decision was later overruled by an initiative amendment to the Hawaii constitution, which changed the law but does not invalidate the principle of the decision. In US Supreme Court case law, even though a federal Equal Rights Amendment has never been adopted, the Court views gender-based classification as requiring "heightened" or "intermediate" scrutiny, often nearly as strong as "strict scrutiny" but allowing that such classifications may be reasonable in some instances.)
Others counter-argue that the law does not discriminate by gender, in that both men and women are equally free to marry, and equally inhibited in marrying someone of the same sex. Thus goes this argument, the law treats everyone equally, and there is no discrimination at all. This line of argument is exactly analogous to those arguments made in defense of miscegenation statutes, i.e., that both whites and non-whites were equally free to marry, and equally inhibited in marrying someone of a different race, thus there was no unequal treatment. That line of thinking has been roundly discredited with regard to miscegenation, and it should be just as dubious with regard to same-sex marriage. (Note that some will try to argue that Loving v. Virginia, the Supreme Court case that finally eradicated anti-miscegenation, was based specifically and solely on race, but they would be mistaken. The Court specifically considered two lines of objection -- that of the suspect class of race, and that of the infringement of the fundamental right to marry -- and affirmed that either line of objection would be dispositive. As the Anal Philosopher would say, the Loving case was "over-determined".) It is important to note that the fundamental right is not just to be able to marry someone, but freedom of personal choice of whom to marry, and how to arrange one's home and family life. The miscegenation precedents are in fact very good analogies for the same-sex marriage issue, in that the Court found the right to marry so fundamental as to consciously go against strong public opinion, long-standing tradition, and natural law arguments embedded in prior case law. Anyone raising such concerns about tradition and natural law (and the accompanying jeremiads about "activist judges") needs to explain how they square their arguments against legal acknowledgement of same-sex marriage with recognition that the Court did the right thing in striking down miscegenation laws. (And they need to open their eyes to the very strong possibility that their views will someday be widely acknowledged as shamefully wrong.)
A third view of the classification involved in denying same-sex marriage is to say that the law is distinguishing between homosexuals and heterosexuals, and because homosexuality is immoral, it is entirely appropriate for the law to enact opprobrium against homosexuals. One can certainly mount a coherent, rational argument along those lines, but it won't carry much legal weight in this country. In Romer v. Evans, the Court found that mere animus against homosexuals failed to muster even a bare "rational basis", and dismissed the specious claims of opponents that homosexuals were seeking "special rights". Indeed, any theory of classification that tries to mask the actual anti-homosexual motivations turns out to be arbitrary, and moreover, it turns out that consciously embracing the anti-homosexual motivations, while intellectually honest, is also abritrary.
The only viable justification for opponents of same-sex marriage is to present a compelling state interest that is directly and narrowly served by the prohibition. Such arguments are generally made along the lines that marriage serves the purpose of raising children, that children are best raised by a mother and a father, and that the state has a compelling interest in fostering specifically heterosexual marriage for the benefit of the children. Many will find this a powerfully persuasive argument, especially as it dovetails with traditional instincts. However, I do not believe that it is sufficient if examined honestly and objectively. (Obviously, I have some explaining to do.) First, we need to clear up common conceptual confusions about childbearing vs childrearing. As I have explained elsewhere, marriage is primarily about childrearing, and childrearing is not intrinsically heterosexual. Before we get into whether heterosexuals might be better parents than homosexuals, we need to acknowledge that we are not looking at an absolute distinction, as in "children are always well-raised by opposite-sex parents and children are always harmed by same-sex parents". Anybody who asserts that is just being obtuse, as there are plenty of counterexamples to be seen. So, if heterosexuals made better parents than homosexuals, it would at most be a matter of averages and degree. (To go back to the miscegenation analogy for a moment, we should note that Virginia put forth a lot of psuedo-scientific argument about the dangers to society of the mixing of races. But fear, uncertainty, and doubt are not sufficient to overcome strict scrutiny when a fundamental right is being infringed.) Part of the process of strict scrutiny is to ask whether the means (prohibiting same-sex marriage) is the only means or the best means to accomplish the legitimate state interest (i.e., healthy childrearing). Here is where we get back to casuistry. If looking for the ideal situation in which to raise children, there are certainly many aspects that may enter into it. One might plausibly argue that wealthy people are better able to raise children than poor people. So would it be appropriate to put a minimum wealth requirement on getting married? One might also argue that educated people are better equipped to raise children, so might we require that at least one party to every marriage must have a college degree? Indeed, intelligent people probably do a better job, so perhaps there should be a minimum IQ requirement on at least one party to every marriage. (It should be noted that earlier Courts, swayed by popular notions of eugenics, have infamously held that being stupid or immoral were sufficient reasons for the state to impose involuntary sterilization on people. See Buck v. Bell, for instance, later repudiated.) Like giving dogs the vote, these are intentionally outrageous analogies (at least I hope they are outrageous) put forth to illustrate the crucial question: in what respect is prohibiting same-sex marriage for the sake of good childrearing substantially different from these other examples? (In fact, if scientific studies were done, I'd venture we would find that intelligence, education, and wealth may have more correlation to successful childrearing than the gender of the parents.) And in fact I think few would dispute the claims of the Heritage Foundation that the stability of a marriage is a strong correlator to happy healthy children, and that divorce is negatively correlated. Clearly there's a much stronger argument for outlawing divorce than prohibiting same-sex marriage, if the goal is to raise children well. Given all of these alternatives, it seems quite difficult to argue that prohibiting same-sex marriage is the only and most narrowly-tailored way of achieving the state's purpose.
Finally returning to analogies, clearly the canine suffrage analogy is way off base. I think the best analogy would be to say that opposing same-sex marriage is analogous to opposing marriage between two persons of the same handedness. Whether we are left or right-handed, like whether we are gay or straight, is something innate that we discover about ourselves as we grow. And left-handedness, like being gay, is something that can to some extent be bludgeoned out of our active behavior by a needlessly coercive society acting out of ignorance and superstition. (Note that none of the arguments I've made above hang on any issues of "essential" versus "chosen" traits.) One might argue that a left-handed child would be deprived being brought up by two right-handed parents (i.e., lacking a good left-handed role model), and thus marriage should be restricted to one right-handed and one left-handed person. But that would be silly, wouldn't it? Yes, that's exactly the right analogy.