Hence:

Darrel Stephens, executive director of Major Cities Chiefs Association, said it's still too early to say if the numbers signal real change.

"It's going to take a bit more to say this trend of 20 years is being reversed," said Mr. Stephens, adding that there may be a rise in a few cities, "but not on a national basis."

Notice how quickly 29 cities became "a few cities."  Notice also the absence of any hint as to how much more, exactly, it would take to spark a sense of urgency in Mr. Stephens. Then notice how, even though the 29 cities are spread all across the country, there's no "national" problem.  

Right.

Homicides in the first six months also declined in 22 cities, including some that saw big jumps in 2015, such as Milwaukee, where killings dropped 26%, according to the survey.

Yes, it's all true.  Look up what "average" means.  When you have very large increases (e.g., 69% in Boston, 61% in Las Vegas, 54% in Louisville) in 29 cities, you're only going to get an average of 15%  if  --  ready now?  --  there are very small increases, or decreases, in other cities.

Maybe I should send the WSJ a sixth grade math book.  I had not thought "average" was that hard a concept.

The rise in homicides in some large cities last year set off considerable debate between police officials and criminologists over what was behind the increase.

Some have attributed increases to the "Ferguson effect," a theory that increases in crime can be attributed to the reluctance of police to engage in confrontation in the face of protests around the U.S. since the 2014 killing of an unarmed black teenager in Ferguson, Mo., by a white police officer.

Among the "officials" are the Obama-appointed head of the FBI, Jim Comey, and the Obama-appointed head of the DEA, Chuck Rosenberg.  (Full disclosure  --  both are friends and former colleagues of mine).

But Mr. Stephens disagrees with that theory, saying he doesn't believe police are pulling back.

"Even if they're angry or annoyed at times, they'll still do the work they're supposed to do," he said.

It's unfortunate that the spokesman for the Major City Chiefs, no less, thinks he needs to be so deceptive.  No one is saying the police aren't "doing the work they're supposed to do." Stephens can't help knowing  --  indeed, in other contexts, he would insist  -- that a huge amount of police work is discretionary.  Police who, a couple of years ago, would have stopped to investigate a late-night gathering on a street corner known for smack dealing, now stay in the car and keep driving.  Who wants to end up on the wrong end of the next Marilyn Mosby (and still less on the wrong end of a Glock-wielding drug pusher)?

When the city's political leadership goes wobbly, the street officer is going to adjust his behavior accordingly.  Indeed, as Heather MacDonald has pointed out, civilian control of the police counsels this outcome (which, I might add, it will get anyway).

Richard Rosenfeld, a criminologist at the University of Missouri-St. Louis, wrote in a Justice Department-funded study released in June that the Ferguson effect was a "plausible" explanation for the sudden jump in killings in 2015.

Mr. Rosenfeld also put forth a second version of the Ferguson effect, writing that the police killings in Ferguson and elsewhere "activated longstanding grievances" in minority communities about police and the criminal-justice system that led to a "legitimacy crisis" and a rise in crime.

What nonsense.  In order to account for the 2015 and subsequent rise in the murder rate, Rosenfeld's theory depends first on the "activation" in that year of "longstanding grievances in minority communities" about the police.

But it didn't happen.  The opposite happened.  As Kent and I have pointed out (here and here), trust in the police increased 2015, and it showed its sharpest increase in minority communities.  In fact, it almost doubled in those communities.

That is sufficient to debunk Mr. Rosenfeld's "distrust-of-the-cops" theory as a cause of 2015's increased murder, but there's more, to wit, that distrust of the cops was never the cause of murder to begin with.  People get murdered because of drug turf wars, domestic disputes, mob enforcement, witness elimination, robberies gone wrong, carjackings, and sometimes sexual sadism.  I am aware of no data, and the article points to none, suggesting that feelings about the police cause murder.

So what are we left with?

The Ferguson effect  --  and, in my view, the broader view seeping into the culture and some precincts of our politics, both that we have become complacent and lost our nerve.