Samsung Unpacked: Everything Announced Galaxy Buds 2 Pro Preorder Galaxy Watch 5 Galaxy Z Fold 4 Dell XPS 13 Plus Review Galaxy Z Fold 4 Preorder Apple TV 4K vs. Roku Ultra Galaxy Z Flip 3 Price Cut
Want CNET to notify you of price drops and the latest stories?
No, thank you

What it means to be an analyst

Glaskowsky shoots down a biased and breathlessly over-hyped article in the New York Times, and explains the proper roles of analysts and reporters in covering stories-- both political and commercial.

The New York Times ran an article over the weekend (here) describing efforts by the Pentagon and the Bush Administration to influence the opinions of military analysts, primarily retired military officers, who contribute to coverage of the Iraq War and other topics by newspapers and TV news programs.

Pentagon logo
US Department of Defense

The Times article claimed that the Pentagon's influence turned these analysts into sock puppets, a claim supported by this quote from Robert S. Bevelacqua-- a military analyst himself:

"It was them saying, 'We need to stick our hands up your back and move your mouth for you.'"

Now, the mere presence of such a quote from an analyst who was part of these Pentagon briefings should make it obvious to anyone that no amount of influence can turn every analyst into a puppet. The Times article was hopelessly, breathlessly hysterical over a simple fact of life... a fact that is familiar to everyone who deals with analysts in politics or the private sector.

The analyst business works the same way for all kinds of analysts-- military, political, financial, and (as in my case) technology analysts.

Within any community of analysts, there will be some who can be bought, some who can be brainwashed, and some who can be bamboozled. Any given analyst will have some ability to think clearly and independently; each analyst decides whether to exercise that ability or simply regurgitate the spin offered by his or her sources.

Analysts also bring in certain biases and preconceived ideas. I have my own, of course. I believe the companies I cover (or work for!) ought to do useful new work, respect the intellectual property of other companies, and deal honestly with its customers, partners, and competitors alike. I approve of technical monopolies-- those created when a company is first to develop a technology-- and I don't approve of monopolies created by predatory trade practices. That still leaves room for plenty of hard competition, and I approve of that, too.

There are similar biases among military analysts. For example, some believe Islamic extremism and anti-American terrorism ought to be met with military force. Some believe the US ought to reserve the military option for more immediate or substantial strategic threats. I don't even know any military analysts, but I can see their biases. Presumably military reporters at the New York Times see them too, and shouldn't pretend otherwise.

Sources-- whether in the public or private sectors-- have a very limited ability to influence these biases. Their best way to influence an analyst's opinions is to make sure the analyst is aware of all the facts that are favorable to the source's position. That's what the Times says the Pentagon did in this case. For example, in attempting to counteract bad publicity generated by criticism of the Guantánamo facility by Amnesty International:

On the flight to Cuba, for much of the day at Guantánamo and on the flight home that night, Pentagon officials briefed the 10 or so analysts on their key messages -- how much had been spent improving the facility, the abuse endured by guards, the extensive rights afforded detainees.

Does the New York Times really believe that this was inappropriate? The article doesn't attempt to claim that these briefings, or the opinions later voiced by the analysts, were misleading or wrong. Apparently the Times believes it's damning enough that the analysts accepted the Pentagon's claims. But I have seen few if any cases of outright deception in my experience with analyst briefings (literally hundreds of them over the years). More commonly, spin is applied by withholding unfavorable facts and by withholding briefings from analysts who hold fixed and unfavorable opinions.

A wise analyst is aware of these blind spots and simply refrains from offering opinions on them. For example, I doubt the Pentagon gave the analysts in the Times story any statistics on unauthorized corporal punishment of detainees by Guantánamo staff, and I doubt any of the analysts interpreted this lack of data as indicating such contact never happens. (I certainly have no idea whether it happens, so I'm not saying it does or doesn't.)

The Times article also suggests that it's inappropriate for analysts to try to help their sources craft their public messages. Given that analysts are paid to have opinions, it should come as no surprise to the Times or anyone else that analysts like to share these opinions with everyone around them-- sources as well as reporters. That's the difference between analysts and reporters, after all; analysts are held to have enough relevant experience to justify having and expressing opinions. Reporters are not.

And it usually doesn't even matter what opinions an individual analyst holds. Reporters simply find analysts who will deliver the kinds of opinions they want. This "quote shopping" is inevitable and ubiquitous, and I'm not even going to say it's wrong; reporters have to have this freedom. But it means that reporters-- including David Barstow, who wrote this piece for the Times-- are trying to influence their readers the same way this article claims the Pentagon is trying to influence military analysts. Barstow included dozens of quotes in his article to support his position, and only twice did he quote a military analyst defending his objectivity-- although I'm pretty sure most would have done so if Barstow had given them the chance.

Ultimately it has to be up to the reader to critically evaluate every line of every news story. Readers shouldn't assume analyst opinions are unbiased any more than they should assume that the facts in the story are complete or truly representative. But facts and opinions usually do have some basis in reality, and a critical reader can usually learn something about the truth of the matter in spite of all the biases that went into the story.

It's certainly useful for the Times to periodically remind us all of how analysts develop their opinions, but it isn't so useful to provide a view of the process that is as biased and misleading as this one.

(Thanks to my friend Elf Sternberg for bringing the Times story to my attention, although I don't think the story is anywhere near as significant as Elf does.)