A new but characteristically confused debate over whether race is a biological reality reminds me of the value of having a simple definition of "race" in mind.

Left-leaning British-Indian science journalist Kenan Malik's latest book, Strange Fruit: Why Both Sides are Wrong in the Race Debate, has so far only gone on sale in the U.K.. But he argues online: "Race is not a real biological entity."

In a hostile review of Malik's book in Prospect, the fine British intellectual magazine, Mark Pagel, a British evolutionary biologist, lands a flurry of blows:

"Malik knows these facts about our genetics, but wants to insist that, unless 'race' corresponds to absolute boundaries, it is a useless and damaging concept. But to deny what everybody knows and to swap the word race for something less politically charged like 'group' is just an act of self-denial and certainly no more accurate than the dreaded 'r' word. It is also patronising—I would like to think we are all grown up enough to accept the facts and ready ourselves for the deluge to come. I say deluge because the more we measure, the more genetic differences we find among populations …"

Unfortunately, Pagel doesn't deliver a knockout punch because he lacks a definition of race. Like Supreme Court justice Potter Stewart wrote in his famous opinion in a pornography case, Pagel can't define race, but he knows it when he sees it.

Malik responded in Prospect by noting, with some justice, "The debate over race has moved on. To judge from his review of my book, Mark Pagel hasn't noticed."

To illustrate the state of the art thinking among the race realists he opposes, Malik writes:

"In the 19th and early 20th centuries, races were viewed as fixed groups, each with its own distinct behaviour patterns and physical characteristics. They could be ranked on an evolutionary hierarchy, with whites at the top and blacks at the bottom. Today, with a few exceptions, race realists reject the idea that there are essential, unbridgeable differences between human populations, or that differences signify inferiority or superiority. So how do they define a race? Usually as 'an extended family that is inbred to some degree' in the words of Steve Sailer of the Human Biodiversity Institute."

Neither my friends nor my detractors will likely agree—but I suspect that, in the long run, my key contribution will turn out to be exactly what Malik [Send him mail] quotes: coming up with a straightforward definition of what a racial group is.

Race, it turns out, is about who your relatives are—a tautology with manifold implications.

That race is about family should be reassuring. Our culture produces little insightful thinking about race because the topic seems too huge and terrifying. Yet, we all have experience thinking pragmatically about the strengths and weaknesses, the benefits and detriments of extended families. And much of that reasoning is also applicable to thinking about race.

While this notion that race is mostly family writ large may seem abstract and irrelevant, it turns out to be extremely useful for understanding the world.

The more straightforward your grasp of the basics, the easier it is to understand the subtleties. For example, back in 2007, practically every other white pundit was extolling Barack Obama's purported racial transcendence on the naïve assumption that Obama's mixed ancestry meant that, like the levelheaded Tiger Woods, he wasn't all that interested in race and wasn't biased against any race.

In contrast, I pointed out that Obama was dead serious about the subtitle he gave Dreams From My Father his 1995 autobiography: A Story of Race and Inheritance. I noted that Obama, by his own voluminous testimony, was actually obsessed with race, with proving that he was "black enough."

Malik condescendingly objects to my definition:

"If we are trying to sort out the problems of life over a pint such vagueness and confusion generally does little harm. We would expect a scientist or physician, however, to think with greater precision."

Malik goes on to explain how Real Scientists Do Things:

"In their book Sorting Things Out, Geoffrey Bowker and Susan Star point out that any scientific classification must possess three properties. First, there must be 'consistent, unique classificatory principles in operation'. So, when biologists order the living world, the rules they use to define humans (Homo sapiens) as a species are the same as the rules they use to define chimpanzees (Pan troglodytes) as a species. Second, 'categories must be mutually exclusive'. A chimpanzee cannot belong to two distinct species. And third, a classification system must be complete and able to absorb even those entities not yet identified. … Racial classifications possess none of these properties."

Oh yeah?

First, defining a “species” is a much less precise process than Malik suggests. At present, there are 26 different definitions of "species" in circulation among scientists.

Second, and more important, rules for defining species are not terribly relevant because races aren't species.

Races are better thought of as extended families. "Extended family" is a perfectly legitimate—indeed, invaluable—scientific concept. A Google search of "'extended family' anthropology" yields 169,000 references.

Yet, extended families obviously don't follow Bowker and Star's so-called rules for "scientific classification." For example, extended families are not mutually exclusive. You can be classified as belonging, at minimum, to both your mother's and your father's extended families.

Malik writes:

"There is,[Sailer] claims, 'no need to say just how big the extended family has to be or just how inbred'. Africans, Icelanders, Basques, even Northern Irish Protestants and Northern Ireland Catholics all constitute distinct races.

"Of course, how one defines an 'African' or a 'Basque' is slightly more complicated than how one defines a 'coyote.'"

Oh yeah? again. Ironically, "how one defines a 'coyote'" on the margin has been a complicated scientific and legal controversy for many years. The "red wolf" of the Southern and Eastern United States is protected under the Endangered Species Act, even though many are more coyote than anything else. The federal Fish & Wildlife Service admits:

"The exact identity of the red wolf has been debated for decades, with some authorities considering it a species, some considering it a subspecies of the gray wolf, and some considering it a hybrid, or cross-breed, of the coyote and the gray wolf."

Naturalists in the field, lacking any sort of genealogical information about whatever animals they studied, traditionally had to use organisms' looks to guess whether they belonged to their own species or merely to a subspecies.

Yet for a few animals, we use genealogy rather than appearance to classify them. You can't enter a horse in the Kentucky Derby unless it is a Thoroughbred. Wikipedia says that "All modern Thoroughbreds can trace their pedigrees to three stallions originally imported into England in the 1600s and 1700s, and to 74 foundation mares of English and Oriental (Arabian or Barb) blood."

Not surprisingly, most Thoroughbreds have a fairly distinctive appearance and manner, but that's not what makes them Thoroughbreds. Seabiscuit, the 1930s champion, for instance, was a rather funny looking horse with an odd predilection for lying down and snoozing whenever allowed. But Seabiscuit's family tree, which included Man o' War as a grandfather, was 100% Thoroughbred, so he was eligible.

Thus, Thoroughbreds are an extreme example of a race—an extended family that has only inbred for dozens of generations.

Malik continues his argument against my definition:

"In any case, even if human populations were as easy to define as animal species, we are still faced with the old problem: if any group can be a race, then the concept of race becomes meaningless."

No—not every group can be a race under my definition. Females, senior citizens, lefthanders, Catholics, homosexuals, Capricorns, cancer survivors, golfers, Democrats, Los Angeles Lakers, English-speakers, Wal-Mart employees, and Hispanics are not racial groups.

It's extremely useful to have a concise understanding of why those groups are, obviously, not racial groups, while other groups, such as Navajo and Samoans, are.

The closest thing to a racial group is probably an ethnic group. And still, the U.S. Census Bureau insists upon drawing a sharp distinction

"People of Hispanic origin may be of any race… [I]n the federal statistical system ethnic origin is considered to be a separate concept from race."

Here's a general definition of "ethnic group" that I've come up with to match how the Census Bureau distinguishes conceptually between race and ethnicity:

“An ethnic group is one defined by shared traits that are OFTEN passed down within biological families—such as language, surname, religion, cuisine, accent, self-identification, or heroes—but that don't HAVE to be.”

Thus, a Chinese baby adopted by Jewish parents (such as little Lily Ling Goldstein described practicing for her bat mitzvah in It's Not Easy Being Jewish and Chinese  in the Jewish Journal of Greater Los Angeles) could be described as racially Chinese and ethnically Jewish.

Malik tries again in Prospect to dismiss my definition of race:

"But once everything from the British royal family to the entire human population can be considered a race (because each is an "extended family inbred to some degree"), then the category has little value."

Actually, the concept of race has more value when it doesn't depend on an arbitrary cut-off size.

Please understand that I'm not making up a wholly novel use of the term "race." It doesn't apply just to blacks and whites, as some seem to assume today.

Contra Malik, the term "human race," which appears 11,200,000 times on the Web, is a perfectly reasonable one.

Similarly, that a "royal family" could constitute a "race" also appeared reasonable to William Shakespeare ("Live, and beget a happy race of kings" from Richard III), Thomas Paine, Edward Gibbon, David Hume, Samuel Adams, Sir Walter Scott, Alexander Pope, Nathaniel Hawthorne, John Dryden, William Blackstone, and Alexis de Tocqueville, all of whom used the terms "race of kings," "race of monarchs," or "race of princes" to describe royal lineages.

As did the two most famous dictionary makers of the English language, Samuel Johnson and Noah Webster. The American lexicographer wrote in 1828:

"RACE, n.1. The lineage of a family, or continued series of descendants from a parent who is called the stock. A race is the series of descendants indefinitely. Thus all mankind are called the race of Adam; the Israelites are of the race of Abraham and Jacob. Thus we speak of a race of kings, the race of Clovis or Charlemagne; a race of nobles, &c."

Of course, "the race of Charlemagne" (who lived at least 40 generations ago) no doubt includes hundreds of millions of people today. After all, your family tree has at least a trillion slots to fill from 1200 years ago, so it's probable that Charlemagne's name would show up at least once.

Charlemagne was your 40th great-grandfather? You don't say! Well, keep moving toward the back of the bus because lots more descendants of Charlemagne are waiting to get on.

Simple lineages tend to dwindle into insignificance, unless they are reinforced by endogamy. That's why I emphasize inbreeding in defining race. Without partial barriers against out-marriage, extended families quickly lose coherence.

If you are of European descent, you are probably descended from Charlemagne via not just one but by hundreds or thousands or millions of genealogical pathways. You may also be descended from Genghis Khan, who left a huge genetic footprint across the Asian steppe.

Yet, and here's the key point, Europeans are likely to be more closely related, via more paths down their family trees, to Charlemagne (and thus to each other) than to Genghis Khan. For Mongolians, the opposite is true.

In the 19th Century, European royalty became increasingly endogamous, turning into an "inbred multinational elite," according to Harvard historian Niall Ferguson's The War Of The World. They formed a pan-European race of royalty, who only married each other, except "in extremis, when the sole alternative was spinsterhood." For instance, through dynastic marriages, Czar Nicholas II, who abdicated in 1917, was only 1/128th Russian.

Ferguson explains: "What better check could be imagined to the fractious tendencies of nineteenth century nationalism than the systematic intermarriage of the continent's sovereigns."

In the crisis of 1914, however, the family ties between Czar Nicholas II, Kaiser Wilhelm II, and King George V, who were all first cousins, proved insufficient to prevent the Great War from breaking out. The royals tended to be infected less than their subjects by war fever. But they were still dragged along by their bellicose nations.

By the 1880s, Queen Victoria had begun to worry about excessive inbreeding among her extended family. She defended her decision to allow her youngest daughter, Beatrice, to marry a mere Battenberg (Mountbatten) by saying:

"If there were no fresh blood, the royal race would degenerate morally and physically."

And, in fact, the "royal race" did develop its own notorious genetic signature due to inbreeding, with hemophilia afflicting at least nine of her male descendents. This had world-historical consequences in the case of Victoria's great-grandson Alexis Romanov, the son of two of Victoria's grandchildren, and heir to the Russian throne.

The finest doctors in Europe could do less to ease the boy's pain than could quiet talks with Father Grigori Rasputin, a mystic from Siberia. The grateful Czarina urged her husband, Czar Nicholas II, to listen to Rasputin's political advice. Soon, the wild man with the Charles Manson eyes was close to being the de facto prime minister of Russia.

Because the Crown Prince's hemophilia was kept secret, few understood why the Czarina favored the erratic Rasputin. During World War I, many Russians suspected the German-born Czarina of attempting to sabotage their war against her homeland. These doubts contributed to the February 1917 overthrow of the Russian royal family—which made possible Lenin's cataclysmic October Revolution.

This notion of "marital alliances" as a harbinger of peace among rival extended families seems terribly outdated today. Who would think any longer that dynastic marriages could solve anything? In our enlightened times, it seems silly to imagine that a prince combining Russian and German ancestry could keep the peace between Russians and Germans.

And yet, that is exactly how Barack Obama successfully sold himself in his now-famous keynote address to the 2004 Democratic convention. By beginning with 380 words about his family background, Obama positioned himself as the promised prince, the offspring of a marital alliance between the black and white races: He is the one we've been waiting for, the mutual heir who will unite black and white.

(That his parents' actual marriage was a short-lived bigamous fiasco that happened because his already-married father had impregnated his 17-year-old mother—well, Obama glosses over that part.)

Marital alliances, however, can create their own problems. Just as the Russians were mistrustful of their German Czarina during WWI, with horrific consequences, Obama has been dogged all his life by African Americans' doubts over whether he is "black enough." Thus he has had to take extravagant steps to prove his blackness to them—such as joining Rev. Wright's extremist church.

It's striking how bearing in mind something as esoteric-sounding as a definition of race as a partly inbred extended family is so useful in understanding something so immediately pertinent as the character of the Democratic nominee for President.

Not necessarily encouraging—but striking.