Also in this column:

* CBC Nixes Campaign Polls

* Poll Vault: Exit Laughing

Most Americans reject torture as a technique to force suspected terrorists to answer questions about possible future attacks but are divided on whether less harsh forms of physical abuse should be allowed to compel uncooperative suspects to reveal information that could save lives, according to a new Washington Post-ABC News poll.

More than six in 10 -- 63 percent -- of the public say torture is never acceptable, even in cases in which a suspect is believed to have knowledge of an upcoming terrorist attack. Slightly more than one in three say torture can be used in some cases.

The United States has publicly stated that it does not torture prisoners, though it does use "stress and duress." Americans are far more divided on whether lesser forms of physical abuse should be allowed in instances when investigators strongly suspect that the suspect might have information that could help avert a deadly terrorist attack.

Slightly more than half of those interviewed -- 52 percent -- rejected the use of less extreme forms of physical abuse in order to compel suspects to reveal potentially life-saving information to investigators, while 46 percent say these tactics are sometimes acceptable.

Among the techniques that a majority of Americans broadly see as acceptable: depriving a suspected terrorist of sleep (66 percent), keeping a hood on a suspect for long periods of time (57 percent) and playing loud music or other noises for extended periods (54 percent). All of these techniques have reportedly been used in Iraq and elsewhere, sometimes with official approval, to force reluctant suspects to talk to military or government investigators.

But the list of things that Americans reject is even longer, including some techniques featured in photographs and videos that have become the chilling visual record of abuse at the Abu Ghraib prison outside Baghdad.

Clear majorities rejected sexually humiliating a suspect (84 percent), applying electric shocks to a prisoner (82 percent), threatening to harm members of the suspect's family (80 percent), holding a suspect's head under water (78 percent), forcing the suspect to go naked (74 percent), punching or kicking a suspect (69 percent), withholding food or water (61 percent), exposing the suspect to extreme heat or cold (58 percent), or threatening to shoot the suspect (57 percent).

Despite widespread objections to these techniques, only a third of Americans would define what happened at Abu Ghraib as torture.

Women and lower income Americans were significantly less likely to approve of the strategic use of torture. Looking across age groups, opposition to torture peaked among those aged 65 and older.

Those Americans most concerned about the threat of terrorism were also more willing to use extreme techniques. A majority (55 percent) of those who said terrorism would be their top voting issue this November also said they would condone the use of torture against suspected terrorists under certain circumstances.

Four in 10 Republicans and nearly as many political independents said torture is sometimes acceptable, a view held by 27 percent of all Democrats.

Despite the fact that the U.S. government has repeatedly claimed that it does not condone torture of detainees, half the public believes that torture is taking place as a matter of policy. An even larger percentage (66 percent) think that military policy allows for physical abuse of prisoners.

The Bush administration has also made distinctions between allowable treatment for international terrorists, such as suspected members of al Qaeda, and treatment of captured Iraqi insurgents, but the poll suggests that the public is not making similar distinctions. Equal proportions would back torture or physical abuse in the case of those "suspected of recent attacks against U.S. forces in Iraq or Afghanistan" as in the case of suspected terrorists. This may be due in part to the public's divided view as to who is behind the deadly violence in Iraq: four in ten said ordinary Iraqis opposed to the occupation were behind most of the attacks, but just as many said international terrorist groups were responsible.

CBC Just Says No to Pre-Election Polls

With national elections in Canada set for late next month, the Canadian Broadcast System will not sponsor its own polls this campaign season and has advised its political reporters to ignore pre-election surveys done by other news organizations.

The directive came in an e-mail to employees earlier this month from CBC's chief news editor Tony Burman. "In addition to not commissioning any polls during the campaign that focus on 'voter preference' or 'try to suggest the eventual outcome,'" the CBC "will place limits on the systematic reporting of polls conducted by other media organizations.

"We will be similarly restrained in the discussion of such poll results on our news and current affairs programs, and in the careful way these results are promoted and placed in our programs," Burman said.

The goal, Burman wrote, is "to ensure that more coverage and attention during the campaign will be devoted to the actual issues in front of the electorate -- leaving the determination of actual 'voter preference' to the voters on election day."

Well, perhaps. But some Canadian observers think that Berman's motives may be less than pure. The CBC is owned lock, stock and barrel by the Canadian government. And the current government headed by Paul Martin has been looking shaky in recent polls -- a fact that has been widely reported, much to the chagrin of the prime minister.

"Perhaps it's a coincidence, but I think not, that the CBC has decided not to give Canadians the potentially bad news during the upcoming campaign," wrote Norman Spector, a columnist for the Vancouver Sun, characterizing the move as one of several "signs of thought control at the state broadcaster."

It's not as if Berman hates all polls. Last year, in a letter to the National Post newspaper complaining about an opinion column critical of his network's Middle East coverage, Berman wrote, "It's worth noting that Canadians at large do not agree with the Post's editorial position on the CBC. Earlier this year, an Ekos poll found that our Middle East coverage is not perceived to have an anti-Israel bias. During the recent Iraq war, Canadians told us they found CBC's coverage more balanced and objective than other Canadian networks or CNN and, far from tuning out, audiences increased by a third."

Despite the ban, CBC Radio released the results of its pre-election poll earlier this week based on interviewing conducted May 12-18. The survey included a candidate preference "horserace" question -- precisely the kind that Burman presumably despises.

Eh?

Well, Burman is saved by the fine print. His announcement specifically noted that CBC will not be "commissioning" any polls during the national campaign, which officially began on Sunday, when the June 28 election day was officially announced. The CBC radio poll was commissioned weeks ago, long before the date was set.

For the record, Canada historically has been a bit chilly to polling. The country bans all mentions of polls in the media on election day. The ban used to extend for three days but was trimmed back to a single day before the 2000 vote.

Poll Vault: Laugh Line

We're a jolly people, aren't we?

Two in three Americans told Gallup polltakers in early 1951 that they had had a real belly laugh in the past day, this despite a recent assassination attempt on President Harry S. Truman by Puerto Rican nationalists and a declaration of a "national emergency" as a results of setbacks in Korea.

Wonder how our funny bone is holding up under the stress and duress of terrorism, soaring gas prices and conflict in the Middle East? Hmmm. Maybe we'll check. Watch this space.

Question: Would you think back and tell me if you have had a real good laugh in the last 24 hours?

64%: Yes

36%: No

Source: In-person survey conducted by the Gallup Organization, Feb. 4-9, 1951, among a national sample of 1,403 adults. Data provided by The Roper Center for Public Opinion Research, University of Connecticut.