IS THE CRIMINAL-JUSTICE SYSTEM RACIST?

No: the high percentage of blacks behind bars reflects crime rates, not
bigotry.

The race industry and its elite enablers take it as self-evident that
high black incarceration rates result from discrimination. At a
presidential primary debate this Martin Luther King Day, for instance,
Senator Barack Obama charged that blacks and whites "are arrested at
very different rates, are convicted at very different rates, [and]
receive very different sentences . . . for the same crime." Not to be
outdone, Senator Hillary Clinton promptly denounced the "disgrace of a
criminal-justice system that incarcerates so many more
African-Americans proportionately than whites."

If a listener didn't know anything about crime, such charges of
disparate treatment might seem plausible.

After all, in 2006, blacks were 37.5 percent of all state and federal
prisoners, though they're under 13 percent of the national population.
About one in 33 black men was in prison in 2006, compared with one in
205 white men and one in 79 Hispanic men. Eleven percent of all black
males between the ages of 20 and 34 are in prison or jail. The
dramatic rise in the prison and jail population over the last three
decades-to 2.3 million people at the end of 2007 (see box)-has only
amplified the racial accusations against the criminal-justice system.

The favorite culprits for high black prison rates include a biased
legal system, draconian drug enforcement, and even prison itself.

None of these explanations stands up to scrutiny.

The black incarceration rate is overwhelmingly a function of black
crime.

Insisting otherwise only worsens black alienation and further defers a
real solution to the black crime problem.

Racial activists usually remain assiduously silent about that
problem.

But in 2005, the black homicide rate was over seven times higher than
that of whites and Hispanics combined, according to the federal Bureau
of Justice Statistics. From 1976 to 2005, blacks committed over 52
percent of all murders in America. In 2006, the black arrest rate for
most crimes was two to nearly three times blacks' representation in
the population. Blacks constituted 39.3 percent of all violent-crime
arrests, including 56.3 percent of all robbery and 34.5 percent of all
aggravated-assault arrests, and 29.4 percent of all property-crime
arrests.

The advocates acknowledge such crime data only indirectly: by charging
bias on the part of the system's decision makers.

As Obama suggested in the Martin Luther King debate, police,
prosecutors, and judges treat blacks and whites differently "for the
same crime."

Let's start with the idea that cops over-arrest blacks and ignore
white criminals. In fact, the race of criminals reported by crime
victims matches arrest data. As long ago as 1978, a study of robbery
and aggravated assault in eight cities found parity between the race
of assailants in victim identifications and in arrests-a finding
replicated many times since, across a range of crimes.

No one has ever come up with a plausible argument as to why crime
victims would be biased in their reports.

Moving up the enforcement chain, the campaign against the
criminal-justice system next claims that prosecutors overcharge and
judges oversentence blacks. Obama describes this alleged postarrest
treatment as "Scooter Libby justice for some and Jena justice for
others." Jena, Louisiana, of course, was where a D.A. initially lodged
second-degree murder charges against black students who, in December
2006, slammed a white student's head against a concrete beam, knocking
him unconscious, and then stomped and kicked him in the head while he
was down. As Charlotte Allen has brilliantly chronicled in The Weekly
Standard, a local civil rights activist crafted a narrative linking
the attack to an unrelated incident months earlier, in which three
white students hung two nooses from a schoolyard tree-a display that
may or may not have been intended as a racial provocation. This
entrepreneur then embellished the tale with other alleged instances of
redneck racism-above all, the initial attempted-murder charges.

An enthusiastic national press responded to the bait exactly as
intended, transforming the "Jena Six" into victims rather than
perpetrators. In the seven months of ensuing headlines and protests,
Jena became a symbol of systemic racial unfairness in America's court
system.

If blacks were disproportionately in prison, the refrain went, it was
because they faced biased prosecutors-like the one in Jena-as well as
biased juries and judges.

Backing up this bias claim has been the holy grail of criminology for
decades-and the prize remains as elusive as ever. In 1997,
criminologists Robert Sampson and Janet Lauritsen reviewed the massive
literature on charging and sentencing. They concluded that "large
racial differences in criminal offending," not racism, explained why
more blacks were in prison proportionately than whites and for longer
terms.

A 1987 analysis of Georgia felony convictions, for example, found that
blacks frequently received disproportionately lenient punishment. A
1990 study of 11,000 California cases found that slight racial
disparities in sentence length resulted from blacks' prior records and
other legally relevant variables.

A 1994 Justice Department survey of felony cases from the country's 75
largest urban areas discovered that blacks actually had a lower chance
of prosecution following a felony than whites did and that they were
less likely to be found guilty at trial.

Following conviction, blacks were more likely to receive prison
sentences, however-an outcome that reflected the gravity of their
offenses as well as their criminal records.

Another criminologist-easily as liberal as Sampson-reached the same
conclusion in 1995: "Racial differences in patterns of offending, not
racial bias by police and other officials, are the principal reason
that such greater proportions of blacks than whites are arrested,
prosecuted, convicted and imprisoned," Michael Tonry wrote in Malign
Neglect. (Tonry did go on to impute malign racial motives to drug
enforcement, however.) The media's favorite criminologist, Alfred
Blumstein, found in 1993 that blacks were significantly
underrepresented in prison for homicide compared with their presence
in arrest.

This consensus hasn't made the slightest dent in the ongoing search
for systemic racism.

An entire industry in the law schools now dedicates itself to flushing
out prosecutorial and judicial bias, using ever more complicated
statistical artillery.

The net result?

A few new studies show tiny, unexplained racial disparities in
sentencing, while other analyses continue to find none. Any
differences that do show up are trivially small compared with the
exponentially greater rates of criminal offending among blacks.

No criminologist would claim, moreover, to have controlled for every
legal factor that affects criminal-justice outcomes, says Patrick
Langan, former senior statistician for the Bureau of Justice
Statistics. Prosecutors and judges observe the heinousness of a
defendant's conduct, for example, but a number-crunching researcher
has no easy way to discover and quantify that variable.

Some criminologists replace statistics with High Theory in their
search for racism. The criminal-justice system does treat individual
suspects and criminals equally, they concede.

But the problem is how society defines crime and criminals.

Crime is a social construction designed to marginalize minorities,
these theorists argue.

A liberal use of scare quotes is virtually mandatory in such
discussions, to signal one's distance from primitive notions like
"law-abiding" and "dangerous." Arguably, vice crimes are partly
definitional (though even there, the law enforcement system focuses on
them to the extent that they harm communities). But the social
constructivists are talking about all crime, and it's hard to see how
one could "socially reconstruct" assault or robbery so as to convince
victims that they haven't been injured.

Unfair drug policies are an equally popular explanation for black
incarceration rates.

Legions of pundits, activists, and academics charge that the war on
drugs is a war on minorities-a de facto war at best, an intentional
one at worst.

Playing a starring role in this conceit are federal crack penalties,
the source of the greatest amount of misinformation in the race and
incarceration debate.

Crack is a smokeable and highly addictive cocaine concentrate, created
by cooking powder cocaine until it hardens into pellets called
"rocks." Crack produces a faster-and more potent-high than powder
cocaine, and it's easier to use, since smoking avoids the
unpleasantness of needles and is more efficient than snorting.

Under the 1986 federal Anti-Drug Abuse Act, getting caught with five
grams of crack carries a mandatory minimum five-year sentence in
federal court; to trigger the same five-year minimum, powder-cocaine
traffickers would have to get caught with 500 grams.

On average, federal crack sentences are three to six times longer than
powder sentences for equivalent amounts.

The media love to target the federal crack penalties because crack
defendants are likely to be black.

In 2006, 81 percent of federal crack defendants were black, while only
27 percent of federal powder-cocaine defendants were. Since federal
crack rules are more severe than those for powder, and crack offenders
are disproportionately black, those rules must explain why so many
blacks are in prison, the conventional wisdom holds.

But consider the actual number of crack sellers sentenced in federal
court each year. In 2006, 5,619 were tried federally, 4,495 of them
black.

From 1996 to 2000, the federal courts sentenced more powder
traffickers (23,743) than crack traffickers (23,121). It's going to
take a lot more than 5,000 or so crack defendants a year to account
for the 562,000 black prisoners in state and federal facilities at the
end of 2006-or the 858,000 black prisoners in custody overall, if one
includes the population of county and city jails.

Nor do crack/powder disparities at the state level explain black
incarceration rates: only 13 states distinguish between crack and
powder sentences, and they employ much smaller sentence
differentials.

The press almost never mentions the federal methamphetamine-trafficking
penalties, which are identical to those for crack: five grams of meth
net you a mandatory minimum five-year sentence.

In 2006, the 5,391 sentenced federal meth defendants (nearly as many
as the crack defendants) were 54 percent white, 39 percent Hispanic,
and 2 percent black.

But no one calls the federal meth laws anti-Hispanic or
anti-white.

Nevertheless, the federal crack penalties dominate discussions on race
and incarceration because they seem to provide a concrete example of
egregious racial disparity.

This leads to a commonly expressed syllogism: crack penalties have a
disparate impact on blacks; disparate impact is racist; therefore,
crack penalties are racist.

This syllogism has been particularly prominent recently, thanks to the
U.S. Sentencing Commission's 2007 decision to lighten federal crack
penalties retroactively in the name of racial equity.

The press has covered this development voraciously, serving up a
massive dose of crack revisionism aimed at proving the racist origins
of the war on crack. Crack was never a big deal, the revisionist story
line goes. But when Boston Celtics draft pick Len Bias died of a crack
overdose in 1986, the media went into overdrive covering the crack
phenomenon. "Images-or perhaps anecdotes-about the evils of crack, and
the street crime it was presumed to stoke" circulated, as the New York
Times archly put it in a December 2007 article. A "moral panic"
(Michael Tonry's term) ensued about an imaginary threat from a
powerless minority group.

Whites feared that addicted blacks would invade their neighborhoods.
Sensational stories about "crack babies" surfaced. All this hysteria
resulted in the unnecessary federal crack penalties.

Since the 1980s, the revisionist narrative continues, experts have
determined that powder and crack show more pharmacological
"similarities than differences," in the Times's words, and that crack
is no more damaging to fetuses than alcohol.

The belief that crack was an inner-city scourge was thus a racist
illusion, and the sentencing structure to quell it a racist assault.
Or, as U.S. District Judge Clyde Cahill put it, in what one hopes is
not a representative sample of the federal judicial temperament:
"Legislators' unconscious racial aversion towards blacks, sparked by
unsubstantiated reports of the effects of crack, reactionary media
prodding, and an agitated constituency, motivated the legislators . .
. to produce a dual system of punishment."

Leave aside the irony of the press's now declaring smugly that the
press exaggerated the ravages of crack. (The same New York Times that
now sneers at "images-or perhaps anecdotes-about the evils of crack"
ran searing photos of crack addicts in 1993 that included a woman
kneeling before a crack dealer, unzipping his fly, a baby clinging to
her back; such degraded prostitutes, known as "strawberries," were
pervasive casualties of the epidemic.) The biggest problem with the
revisionist narrative is its unreality. The assertion that concern
about crack resulted from "unconscious racial aversion towards blacks"
ignores a key fact: black leaders were the first to sound the alarm
about the drug, as Harvard law professor Randall Kennedy documents in
Race, Crime, and the Law. Harlem congressman Charles Rangel initiated
the federal response to the epidemic, warning the House of
Representatives in March 1986 that crack had made cocaine
"frightening[ly]" accessible to youth.

A few months later, Brooklyn congressman Major Owens explicitly
rejected what is now received wisdom about media hype. "None of the
press accounts really have exaggerated what is actually going on,"
Owens said; the crack epidemic was "as bad as any articles have
stated." Queens congressman Alton Waldon then called on his colleagues
to act: "For those of us who are black this self-inflicted pain is the
worst oppression we have known since slavery. . . . Let us . . .
pledge to crack down on crack." The bill that eventually passed,
containing the crack/powder distinction, won majority support among
black congressmen, none of whom, as Kennedy points out, objected to it
as racist.

These politicians were reacting to a devastating outbreak of
inner-city violence and addiction unleashed by the new form of cocaine.

Because crack came in small, easily digestible amounts, it
democratized what had been a rarefied drug, making an intense high
available to people with very little money. The crack market differed
radically from the discreet phone transactions and private deliveries
that characterized powder-cocaine distribution: volatile young dealers
sold crack on street corners, using guns to establish their turf.
Crack, homicides, and assaults went hand in hand; certain areas of New
York became "like a war zone," retired DEA special agent Robert
Stutman told PBS's Frontline in 2000. The large national spike in
violence in the mid-1980s was largely due to the crack trade, and its
victims were overwhelmingly black inner-city residents.

Though the elites are furiously rewriting crack history, many people
who lived through it are not. In April 2007, Los Angeles prosecutor
Robert Grace won the conviction of a crack dealer who had raped and
strangled to death ten strawberries between 1987 and 1998. The "crack
epidemic was one of the worst things that happened to the black and
brown community," Grace asserts. Matthew Kennedy managed an infamous
public housing project in Watts during the crack epidemic. "Some of us
remember how bad it was," he says. When children avoid school for fear
of getting shot by drug gangs, "you've just lost that generation."
Lawrence Tolliver has witnessed his share of shootings outside his
South Central barbershop. "Sometimes it was so bad you had to scout
the horizon like a gazelle at a watering hole in Africa," he recalls.

It takes shameless sleight of hand to turn an effort to protect blacks
into a conspiracy against them. If Congress had ignored black
legislators' calls to increase cocaine-trafficking penalties, the
outcry among the groups now crying racism would have been deafening.

Yes, a legislative bidding war drove federal crack penalties
ultimately to an arbitrary and excessive point; the reduction of those
penalties is appropriate. But what led to the crack-sentencing scheme
wasn't racism but legal logic.

Prosecutors rely on heavy statutory penalties to induce defendants to
spill the beans on their criminal colleagues. "An amazing public
spirit is engendered when you tell someone he is facing 150 years to
life but has the possibility of getting out after eight if he tells
you who committed a string of homicides," says Walter Arsenault, who
headed the Manhattan district attorney's homicide-investigation unit
in the 1980s and 1990s.

Race activists endlessly promote the claim that the draconian federal
crack laws are sweeping up mere sad sacks with a little extra crack to
spare.

But anyone who fits that description is exempt from the federal
sentencing scheme. Traffickers with only a modest criminal history who
didn't injure others or have a gun when arrested can escape the
mandatory federal sentences if they don't lie to the government about
their offense (there is no requirement to rat out others). In 2006,
only 15.4 percent of crack-cocaine defendants qualified for this
safety-valve provision, compared with 48.4 percent of powder-cocaine
offenders; in 2000, even fewer crack defendants qualified-12.6 percent.

Crack sellers seldom merit the escape clause because their criminal
histories tend to be much more severe than powder sellers' and because
they're more likely to have or use weapons.

The congressional distinction between crack and powder sellers, it
turns out, had a firm grounding.

Equally misleading is the criticism that few crack "kingpins" can be
found in federal prison.

This is not surprising, because "kingpins" in the traditional
sense-heads of major drug-importing rings-don't exist in the crack
world.

Crack is not imported but cooked up locally.

Its supply and distribution scheme is more horizontal than vertical,
unlike that of powder cocaine and heroin.

Federal crack enforcement wasn't about stopping the flow of illegal
drugs into the country; it was about stopping urban violence.

And that violence was coming from street dealers.

Critics follow up their charges about crack with several empirical
claims about drugs and imprisonment. None is true. The first is that
drug enforcement has been the most important cause of the overall
rising incarceration rate since the 1980s. Yet even during the most
rapid period of population growth in prisons-from 1980 to 1990-36
percent of the growth in state prisons (where 88 percent of the
nation's prisoners are housed) came from violent crimes, compared with
33 percent from drug crimes.

Since then, drug offenders have played an even smaller role in state
prison expansion. From 1990 to 2000, violent offenders accounted for
53 percent of the census increase-and all of the increase from 1999 to
2004.

Next, critics blame drug enforcement for rising racial disparities in
prison. Again, the facts say otherwise.

In 2006, blacks were 37.5 percent of the 1,274,600 state
prisoners.

If you remove drug prisoners from that population, the percentage of
black prisoners drops to 37 percent-half of a percentage point, hardly
a significant difference. (No criminologist, to the best of my
knowledge, has ever performed this exercise.)

The rise of drug cases in the criminal-justice system has been
dramatic, it's important to acknowledge. In 1979, drug offenders were
6.4 percent of the state prison population; in 2004, they were 20 percent.

Even so, violent and property offenders continue to dominate the
ranks: in 2004, 52 percent of state prisoners were serving time for
violence and 21 percent for property crimes, for a combined total over
three and a half times that of state drug offenders. In federal
prisons, drug offenders went from 25 percent of all federal inmates in
1980 to 47.6 percent of all federal inmates in 2006. Drug-war
opponents focus almost exclusively on federal, as opposed to state,
prisons because the proportion of drug offenders is highest there.

But the federal system held just 12.3 percent of the nation's
prisoners in 2006.

So much for the claim that blacks are disproportionately imprisoned
because of the war on drugs.

But a final, even more audacious, argument maintains that
incarceration itself, not criminals, causes crime in black
neighborhoods. Because blacks have the highest prison rate, this
argument holds, incarceration constitutes an unjust and
disproportionate burden on them. This idea has gained wide currency in
the academic world and in anti-incarceration think tanks.

Columbia University law professor Jeffrey Fagan offered a
representative version of the theory in a 2003 law review article
co-authored with two public health researchers. Sending black males
to prison "weakens the general social control of children and
especially adolescents," Fagan writes.

Incarceration increases the number of single-parent households. With
adult males missing from their neighborhoods, boys will be more likely
to get involved in crime, since they lack proper supervision. The net
result: "Incarceration begets more incarceration [in] a vicious cycle."

A few questions present themselves. How many convicts were living in a
stable relationship with the mother (or one of the mothers) of their
children before being sent upstate? (Forget even asking about their
marriage rate.) What kind of positive guidance do men who are
committing enough crimes to end up in prison, rather than on probation
(an exceedingly high threshold), provide to young people?

Further, if Fagan is right that keeping criminals out of prison and on
the streets preserves a community's social capital, inner cities
should have thrived during the 1960s and early 1970s, when prison
resources contracted sharply.

In fact, New York's poorest neighborhoods-the subject of Fagan's
analysis-turned around only in the 1990s, when the prison population
reached its zenith.

Fagan, like many other criminologists, conflates the effects of prison
and crime. Neighborhoods with high incarceration rates suffer
disproportionate burdens, he claims.

Firms are reluctant to locate in such areas, decreasing job
opportunities. Police pay closer attention to these high-incarceration
zones, increasing the chance that any given criminal within them will
wind up arrested.

Thus, incarceration "provides a steady supply of offenders for more
incarceration." But if business owners think twice about certain
communities, it's because they fear crime, not a high concentration of
ex-convicts per se. It's unlikely that prospective employers even know
the population of ex-cons in a neighborhood; what they are aware of is
its crime rates. And an employer who hesitates to hire an ex-con is
almost certainly reacting to his criminal record, even if he has been
given community probation instead of prison.

Likewise, if the police give extra scrutiny to neighborhoods with many
ex-convicts, it's because those convicts commit a lot of crime.

Finally, putting more criminals on probation, rather than sending them
to prison-as Fagan and others advocate-would only increase law
enforcement surveillance of high-crime neighborhoods.

This popular "social ecological" analysis of incarceration, as Fagan
and other criminologists call it, treats prison like an outbreak of
infectious disease that takes over certain communities, felling people
on a seemingly random basis. "As the risks of going to jail or prison
grow over time for persons living in those areas, their prospects for
marriage or earning a living and family-sustaining wage diminish as
the incarceration rates around them rise," Fagan says. This analysis
elides the role of individual will. Fagan and others assume that once
one lives in a high-incarceration-that is, high-crime-area, one can do
little to avoid prison.

But even in the most frayed urban communities, plenty of people choose
to avoid the "Life." Far from facing diminished marriage prospects, an
upstanding, reliable young man in the inner city would be regarded as
a valuable catch.

No one doubts that having a criminal record-whether it results in
community probation or prison-is a serious handicap.

People convicted of crimes compete for jobs at a clear disadvantage
with those who have stayed crime-free. But for all the popularity of
the view that the system is to blame, it's not hard to find dissenters
who believe that individuals are responsible for the decision to break
the law. "My position is not hard," says public housing manager
Matthew Kennedy. "You don't have to do that crime." Kennedy supported
President Bill Clinton's controversial 1996 "one-strike" rule for
public housing, which allowed housing authorities to evict drug
dealers and other lawbreaking tenants on their first offense. "I'm
trying to protect the good people in my community," Kennedy explains.
"A criminal record is preventable. It's all on you." Kennedy has no
truck with the argument that it is unfair to send ex-offenders back to
prison for violations of their parole conditions, such as staying away
from their gang associates and hangouts. "Where do they take
responsibility for their own actions?" he wonders. "You've been told,
?Don't come back to this community.? Why would you come back here?
You've got to change your ways, change the habits that got you in
there in the first place."

Though you'd never know it from reading the academic literature, some
people in minority communities even see prison as potentially positive
for individuals as well as for communities. "I don't buy the idea that
there's no sense to prison," says Clyde Fulford, a 54-year-old
lifelong resident of the William Mead Homes, a downtown Los Angeles
housing project.

Having raised his children to be hardworking, law-abiding citizens,
Fulford is a real role model for his neighborhood, not the specious
drug-dealing kind posited by the "social ecological" theory of
incarceration. "I know a lot of people who went to prison," Fulford
says. "A lot changed they life for the better. Prison was they wake-up
call." Is prison unavoidable and thus unfair? "They knew they was
going to pay. It's up to that person." What if the prisoners hadn't
been locked up? "Many would be six feet under."

Robert Grace, the Los Angeles prosecutor, is acutely aware of the
fragility and preciousness of the rule of law. "As a civilized
society, we can't allow what's happening in Latin America to take over
here," he says. "Venezuela and Mexico are awash in appalling violence
because they don't respect the law." Thus, when prominent figures like
Barack Obama make sweeping claims about racial unfairness in the
criminal-justice system, they play with fire. "For any political
candidate to make such claims out of expediency is wrong," Grace says.
"If they have statistics that back up the claim, I'd like to see them.
But to create phony perceptions of injustice is as wrong as not doing
anything about the real thing."

The evidence is clear: black prison rates result from crime, not
racism. America's comparatively high rates of incarceration are
nothing to celebrate, of course, but the alternative is far worse.

The dramatic drop in crime in the 1990s, to which stricter sentencing
policies unquestionably contributed, has freed thousands of
law-abiding inner-city residents from the bondage of fear. Commerce
and street life have revived in those urban neighborhoods where crime
has fallen most.

The pressure to divert even more offenders from prison, however, will
undoubtedly grow. If a probation system can finally be crafted that
provides as much public safety as prison, we should welcome it. But
the continuing search for the chimera of criminal-justice bigotry is a
useless distraction that diverts energy and attention from the crucial
imperative of helping more inner-city boys stay in school-and out of
trouble.

Punishment and Crime

Those who tar the criminal-justice system as racist often make a
broader claim: incarceration doesn't even lower crime, making the
nation's skyrocketing prison rolls a particularly senseless injustice.

Incarceration foes are right about one thing: the U.S. prison
population has swollen dramatically over the last three decades.

The per-capita rate of imprisonment increased three times from 1973 to
2000; the number of state and federal prisoners grew fivefold between
1977 and 2007, from 300,000 to 1.59 million.

When inmates in jails are included, the total number in correctional
facilities at the end of 2007 was 2.3 million, according to the Pew
Center on the States. One in 100 adults is in custody.

This expansion represents a resounding rejection of the reigning crime
philosophy of the 1960s. The 1967 report of the President's Commission
on Law Enforcement and Administration of Justice, a classic Great
Society document, argued that society could reduce crime only by
eliminating poverty and racism, ideally through government-funded
social programs.

Consistent with this theory, prison capacity began dropping during the
sixties and only stopped falling during the late 1970s, when crime
reached intolerable levels. Thereafter, the states started adding
prison beds and passing laws to keep offenders locked up longer and to
reduce judicial discretion to issue very lenient sentences.

Few subjects have proved more contested in criminology than whether
this prison buildup lowered crime-and, if so, by how much. Anyone
entering the thickets of incarceration studies should abandon all
commonsense assumptions, such as that locking away, say, a burglar,
would reduce burglary rates.

Not so, say the criminologists, and at first glance, the crime data
from the late-twentieth-century prison expansion seem to support them.
Only after 1991 was the rise in incarceration consistently accompanied
by decreasing crime rates; in the 1980s, crime went up and down, even
as the prison population steadily grew. And now that crime is falling,
the criminology world finds itself even more puzzled by why the prison
population keeps increasing.

Two of the most common theories as to why prison doesn't lower crime
are logically weak and empirically ungrounded. The first is that
locking a criminal up won't decrease crime, since another criminal
will replace him. Yet while crimes meeting an illicit consumer demand
may operate within a supply-and-demand framework, opportunities for
violence and property crimes hit no natural ceiling.

There are plenty of potential victims of violence and theft to go
around; a potential robber need not wait for a competitor to go to
jail before he can begin his own crime spree.

The second theory to explain why prison doesn't work applies the law
of diminishing returns to incarceration. As we lock up ever more
people, we start scraping the bottom of the criminal barrel, the
critics say. The prisoners we incarcerate become more innocuous than
those picked up initially, so we get a diminishing bang for the buck
for every new prisoner sent away.

However impeccable the economic reasoning behind this claim, there is
no empirical evidence for it. The diminishing-returns argument assumes
that the universe of unapprehended and unincarcerated criminals is
shrinking.

It is not. The chances of getting caught and sent to prison remain
extraordinarily low. The JFA Institute, an anti-incarceration advocacy
group, estimated in 2007 that in only 3 percent of violent
victimizations and property crimes does the offender end up in prison.

In 2004, only 1.6 percent of burglars were in prison, according to the
Bureau of Justice Statistics. The people in prison today, says
statistician Patrick Langan, are "not very different from prisoners in
the past, in terms of their prior records."

In the overwhelming majority of cases, whatever the race of the
convicted, prison remains what it has always been: a lifetime
achievement award for persistence in criminal offending.

Absent recidivism or a violent crime, the criminal-justice system will
do everything it can to keep you out of the state or federal slammer.

It can be disconcerting for the average law-abiding citizen to hear a
prosecutor's typology of the crime universe: most thefts, for example,
are considered "nonserious crimes" that do not merit prison sentences,
unless they concern a huge amount of money or took place in the
victim's presence.

Steal an unoccupied car or burgle an unoccupied home and you'll
probably get probation; hijack a car from a driver or stick up a
pedestrian, however, and you'll probably go to prison.

Columbia University law professor Dan Richman had a chance to test the
"harmless offenders in prison" claim as chair of New York City's Local
Conditional Release Commission. Richman studied the criminal profile
of Rikers jail inmates in late 2004. Jails are supposed to be where
the most "innocuous" lawbreakers end up-those with misdemeanor
convictions or sentences of less than a year. "It struck me how
serious the offenders were," he says. "I'd come from the academy,
where there's persuasive writing about over-incarceration. I had
assumed there would be mostly first-time offenders in jail, but it
wasn't true." About 40 percent of the inmates had prior felony
convictions, Richman discovered, and the inmates' most recent
offenses, which had put them in jail this time around, were usually
serious.

People in for assault would have pleaded down from attempted
manslaughter; possession pleaded down from distribution. "These
weren't people who were there by accident," says Richman.

One can also test the theory that locking away offenders doesn't lower
crime by seeing what prisoners do when they get out. The Bureau of
Justice Statistics studied the postprison careers of over 272,000
state prisoners released in 1994. Within three years, 67.5 percent of
the group had been rearrested for 744,000 new felonies and serious
misdemeanors. How many additional crimes they committed during those
three years before getting arrested is unknown; estimates of the
number of crimes that a typical unapprehended criminal commits per
year range from zero to several hundred. And the ex-cons' post-release
crime spree seems not to have resulted from the negative effects of
prison itself, since convicts who spent the longest time behind bars
had significantly lower rearrest rates than others.

Not all criminologists and law professors dispute that prison lowers
crime. University of Chicago economist Steve Levitt hypothesized in
1996 that had incarceration rates not risen sharply from 1971 to 1993,
violent crime would have been 70 percent higher and property crime
almost 50 percent higher. More typical estimates attribute 10 to 25
percent of the 1990s crime drop to incarceration. And Berkeley law
professor Franklin Zimring rejected the diminishing-returns argument
against incarceration in his 2007 book The Great American Crime
Decline. The fact that crime started dropping consistently only at the
end of the decades-long prison buildup makes perfect sense, he argued,
since that's when the greatest number of criminals were off the streets.

Heather Mac Donald is a contributing editor of City Journal and the
John M. Olin Fellow at the Manhattan Institute. Her latest book,
coauthored with Victor Davis Hanson and Steven Malanga, is The
Immigration Solution.