A majority of Canadians oppose the federal government's plan to give greater powers to police officers to obtain breath samples from drivers in roadside tests, a new poll has found.
As part of its legislative package last month to legalize marijuana, the government also tabled a bill to update impaired-driving laws. Bill C-46 would drop the requirement for police to have a reasonable suspicion a driver has been drinking before demanding a breath sample. For example, officers would no longer need to smell alcohol on the driver's breath or receive an admission that a driver had been drinking.
The government said the move would create one of the world's strongest impaired-driving regimes, especially in comparison to jurisdictions in which cannabis is legal.
However, a new poll by The Globe and Mail/Nanos Research found that only 44 per cent of respondents supported or somewhat supported the proposal, while 55 per cent opposed or somewhat opposed it.
The poll echoes concerns in the legal community over the constitutionality of a system in which police would not have to justify ordering a test.
A number of legal observers have questioned whether the legislation complies with the Charter of Rights and Freedoms, given that it would require giving a bodily sample despite no sign of danger.
On Thursday, Justice Canada issued its legal rationale for the legislation and its view of how it fits with the Charter's protection against unreasonable police searches.
According to Justice lawyers, the invasion of privacy would be minimal in the case of a roadside test in which police officers already have the right to demand several types of information from drivers.
"The information revealed from a breath sample is, like the production of a driver's licence, simply information about whether a driver is complying with one of the conditions imposed in the highly regulated context of driving," the government's paper said. "It does not reveal any personal or sensitive information and taking the sample is quick, and not physically invasive."
According to the government, this measure has helped reduce serious car crashes by 20 to 35 per cent in countries that have adopted it.
Government officials added the new measure would not allow police to stop people randomly on the road to ask for breath samples, but would apply in circumstances such as stopping someone in a RIDE program or because their car lights are not working properly.
Ottawa-based criminal lawyer Michael Spratt said the government's Charter analysis lacked depth and any reference to case law, stating it failed to mount a serious defence of the proposed changes.
"If a first-year student wrote this, they would fail," he said in an interview.
Mr. Spratt said the legal community is divided over the constitutionality of the proposals, but said if the bill is adopted, "there's no doubt it will be challenged."
Bill C-46 would also restrict or eliminate some frequently used defences. For example, the new law would make it a crime to have a blood alcohol level that is over the legal limit within two hours of driving. This would remove the defence that a driver's final drink was just before they took the wheel, so the alcohol had not yet entered their bloodstream.
"If it passes Parliament, [Bill C-46] will be one of the strongest impaired-driving pieces of legislation in the world," Justice Minister Jody Wilson-Raybould said last month. "Ensuring that we have safety on our roads and our highways is of paramount concern."
Robert Solomon, a law professor at the University of Western Ontario who is the legal adviser to anti-impaired-driving group MADD Canada, has said the government's approach sends a strong message.
He predicted that "mandatory breath screening of drivers ... is going to probably make the most significant difference in reducing deaths and injuries." He said 121 countries already have such mandatory screening, and every jurisdiction has achieved sharp and sustained reductions in crashes.
The Nanos survey was conducted between April 29 and May 5, reaching a random survey of 1,000 Canadians that is considered accurate within 3.1 percentage points, 19 times out of 20.