In 1988 Lord Griffiths gave a House of Lords decision in a drink driving case. In this case of Cracknell v Willis  AC 450 he stated that it was possible to challenge the reading from a breathalyser device with evidence of how much alcohol the driver had consumed.
Since 1988, this judgement has been used to challenge a number of drink driving cases. It is the case however that since 1988, Courts have become increasingly sceptical about challenges mounted to the reliability of the devices.
However a 2002 case, Director of Public Prosecutions v Marrable has restated the principle setting out again that It is possible to challenge the reading from a machine by adducing indirect evidence that the machine reading may be incorrect.
In Cracknell, the police arrested Mr Cracknell for drink driving after he failed a roadside breath test. When he was in the police station, he gave one sample of breath in the Lion Intoxilyser machine.
The specimen he gave contained 78 micrograms of alcohol per 100 millilitres of breath. The legal limit was 35 micrograms. He failed to provide a second sample.
At his trial in the magistrates’ court for driving with excess alcohol Mr Cracknell tried to introduce evidence of the amount of alcohol that he had consumed to show that the Lion Intoxilyser was defective. The magistrates refused to allow him to give the evidence and convicted him of drink driving.
Mr Cracknell appealed to the High Court saying that the magistrates should have allowed him to introduce evidence to show that the machine was not working. The Hugh Court agreed with the magistrates, so Mr Cracknell appealed to the House of Lords.
In the House of Lords, Lord Griffiths said that there was no restriction on the type of evidence that a defendant could introduce to challenge the assumption that the breathalyser was working properly. The court should have allowed Mr Cracknell to give evidence of the number of drinks he had consumed.
Famously Lord Griffiths asked:
“Is he to be convicted without the opportunity of calling the two bishops as witnesses to the fact that he had drunk nothing that evening and inviting the magistrates to draw the inference that the machine must have been unreliable?”
Cracknell v Willis  AC 450
Mr Cracknell didn’t have two bishops to give evidence for him. In another section of the judgement, Lord Goff said
magistrates should “give proper scrutiny to such defences” and “will be fully aware of the strength of the evidence provided by a printout, taken from an approved device”.
Cracknell v Willis  AC 450
The victory for Mr Cracknell was of little use to him as the House of Lords upheld his conviction for failing to provide a specimen. An offence that carries the same sentence as driving with excess alcohol.
How have lawyers used Cracknell v Willis in drink driving cases?
When Lord Goff gave his part of the judgement in Cracknell v Willis, he expressed some concern that the judgement would encourage motorists to seek “to persuade magistrates to reject evidence from printouts as unreliable on the grounds that they have drunk so little that the reading cannot be right.”
To some extent, he was correct. The usual scenario is this:
- The police arrest a motorist for drink driving.
- Motorist provides a positive reading on an evidential breath testing machine in the police station.
- The police charge the motorist with driving with excess alcohol and sent it to the magistrates’ court.
- At court, the motorist pleads “not guilty” to the charge and takes the case to trial.
- In the trial, the prosecution introduces evidence of the level of alcohol by producing the printout from the breathalyser machine.
- The defence provides evidence from an expert that the amount of alcohol consumed could not have accounted for the reading from the machine.
- The magistrates have to decide whether the evidence of alcohol consumption creates any doubt about the reliability of the machine.
Does it work?
Rarely. THIS IS WHY.
If a motorist gives evidence that they didn’t consume enough alcohol to account for the reading there can only be two conclusions about that evidence.
- The evidence from the motorist is unreliable; or
- The reading from the breathalyser machine is wrong.
There is a legal presumption that the breathalyser machine is working properly. In most cases, the court will accept the reading from the machine rather than the evidence of the driver. Evidence to rebut the presumption that the machine was working correctly has to be extremely strong. Two bishops?
Why is the Principle in Cracknell v Willis so Important?
Without Lord Griffiths’ principle, it would be almost impossible to challenge the evidence from any approved device. That would create a principle that breathalyser machines, speed detection devices and red light cameras could not be challenged even if there was strong evidence that there was an error.
Such a principle could make proceedings unfair. It could create a situation where a court would be forced to ignore overwhelming evidence even if it showed that a device was not working as it should be.
How have lawyers used Cracknell v Willis in other cases?
The principle can challenge any evidence from a type-approved device.
- Evidential breath testing machines
- Speed detection devices
- Red-light cameras
Is the case out of date?
The House of Lords decided Cracknell v Willis in 1988. But it is still good law. But is there a feeling that the law has moved on and the case has no practical application?
R v Cipriani and R v Hassani
Danny Cipriani faced a trial for drink driving at Westminster Magistrates’ Court in 2016. He was convicted. His lawyers relied on Cracknell v Willis during the trial. District Judge Howard Riddle published his reasons for convicting on the judiciary website.
Another court cannot normally rely on a magistrates’ court judgement. But the High Court referred to DJ Riddle’s comments in another similar case called Hassani v West London Magistrates’ Court  EWHC 1270 (Admin).
The decision says the cases following Cracknell v Willis must be considered in light of:
- Scientific realities
- The changing approach to summary trial
- The Criminal Procedure Rules
- Common sense
Professor Williams’ article
Forensic Science International published an article by Paul M. Williams in December 2018.
Mr Williams called for parliament to pass a statute to overturn the principle in Cracknell v Willis. He said:
“It is therefore time that a Statutory Assumption should be enshrined in the legislation to the effect that an evidential breath alcohol reading should be assumed to be accurate unless there is some direct evidence [as opposed to inferential evidence] to the contrary.”
Problems With an Assumption of Infallibility
Mr Williams’ article does not address how to address the potential “unjust result” identified by Lord Griffiths. What happens in a case with overwhelming evidence to support a malfunction? The proposed change in the law would prevent a court from considering the evidence. To prevent indirect evidence from coming before the court would be to establish a legal principle that the machine is infallible.
Distinguishing between direct and indirect evidence might also be quite a challenge for a statute to identify.
There are also issues of fairness to consider. The court pointed out in Cracknell v Willis that indirect evidence of a malfunction would be very difficult for an individual motorist to get.
A New Life to the Principle- DPP v Marrable
In 2020, the High Court looked at a speeding case. A police speed camera had detected Mr Marrable’s car travelling at 72 mph in a 50 mph limit. At his trial, he provided evidence from his GPS that he was travelling at 53-54 mph. The magistrates found that the GPS cast doubt on the speed camera and found him not guilty.
Just like the breath testing machines, a Trucam speed device is “type approved” by the Home Office.
The Director of Public Prosecutions appealed the acquittal to the High Court. But the High Court agreed with the magistrates. It is perfectly possible to introduce indirect evidence to show that a type-approved device may not be working. Mr Marrable did not and could not have introduced direct evidence of a malfunction. The only way to challenge the reading was by introducing some other indirect evidence. In this case, the GPS from his vehicle.