What is the “Balance of Probabilities?

Central to the system of complaints in the Methodist Church is

S.O. 1133(8c):

 “The standard of proof required to establish a charge is the balance of probability.”  

However, this is not further defined and is often mis-interpreted.

 It is important to note that since the Methodist Church does not further define this system, then the definition must be as it is used in civil law case outside the Methodist Church. After all, most of the Standing Orders are written by senior lawyers – and, in the main, lawyers working in civil law where the system is most used.  

 As for the layman, the Standing Order may not necessarily be the best reference for the working of the system of the balance of probabilities, but, inside the Church, it is all that we have.

 Recent decisions by the Connexional Complaints panel have gone along with this definition: 

The complaints team has to be convinced that the complaint is 51% likely to be true; that is, the complainants’ story only had to be slightly more plausible than the respondent’s story. Expressed another way, the complainant’s case would need to be accepted as more likely than not to be true for the complaint to succeed; that it is more probably than not.”

 However, this is a shorthand – and inadequate  - definition of the system of “balance of probabilities”.  

 In particular it fails to point out that decisions may not be made subjectively without forensic examination and assessment. The quotation quotes the percentage of 51%, but demonstrates no attempt of how to measure probability so as to achieve this percentage.

 This standard of proof is not defined further in the Standing Orders. The lawyers who wrote the Standing Order would take it for granted that such forensic investigation work would be required to establish a balance of probabilities.  The system is well known in law. It is used in civil courts. Why would anyone need to write it into Standing Orders? 

Many connexional inquiry panel members  may be surprised to learn that the phrase “balance of probabilities” does not mean that the judge may decide what is probably the truth on a whim or a hunch .

Unfortunately,  that is what seems to occur in some cases within the Methodist Church.  

The panel quoted above states that it “had to be convinced”. In fact the true position is that the panel needs to work in order to convince itself.   

The two statements may seem to mean the same thing, but they do not, for one of the essentials of the “balance of probabilities” system is that the judge may – and indeed should – take evidence from other, often outside, sources.  

In the end the onus is on the judge to weigh the two sides – and this requires measurement and judicial assessment.  

There are rules and principles involved in the “balance of probabilities”. The balance of the arguments put must be carefully considered and interrogated forensically.  

It is clear from some cases that most Methodist Church Connexional panels have little more than a layman’s view of the rules qualifying the system of judgement on the balance of probabilities. This was one of the core reasons why some complainants get upset.  

As with the system of “beyond reasonable doubt” in criminal law, in the system of “balance of probabilities” there must be a desire to find “best evidence”. This is required in order to achieve the best measure of the “balance of probabilities”.  We shall see that some Standing Orders appear to be in direct conflict with this. 

“Best evidence” may be documentary or forensic evidence – and such will generally out-weigh verbal evidence in the “balance of probabilities.”   

If, after reviewing documentary evidence and such, the tribunal is still left in doubt, the question is resolved by a rule that the asserting party carries the burden of proof. This means that the complainant  has to prove his case. 

There is no position of “reasonable doubt” with the system of “balance of probabilities” as some may think.. The fact that a complainant or  accuser has provided evidence, but not sufficient evidence, to give his case the greater probability, does not leave any lurking doubt over the accused. It is a “black and white” situation, there are no ‘greys’. 

The probability of both sides must be considered. 

Probabilities are measured. The more improbable the event, the stronger the evidence must be to substantiate it.  

The more probable the event might be, the required evidence to substantiate it is less.  

So how do we measure what is “probable”? A well-known example of this rule is told to students by way of an anecdote.  

A man runs into a pub and shouts that there is a lion loose in Regent’s Park. It’s in the public area where children are playing. He calls for someone to phone the police or the RSPCA.  

Another man who has just ordered a drink at the bar turns and says “ I saw that animal – it was a large Alsation dog, not a lion”. 

The publican, hand on phone, pauses – does he call the police or not – which story is the more probable?  An argument ensues. 

The publican remarks that lions are rarely seen walking around our parks. And yet there certainly are lions in Regent’s Park – in the zoo.  

But, on the other hand, even in Regent’s Park Zoo, they are locked away in cages. And, of course, there are keepers at the Zoo whose job it is to ensure that they do not get out and run loose.  

So, the publican thinks, the probability that a lion is walking around in the public areas is low. But it cannot be zero probability. After all, there are lions in Regent’s Park. 

He considers what research might be undertaken before committing himself.  

Why not ring the zoo are ask whether or not a lion has escaped? A little investigation might settle this. Perhaps if they counted their lions, that might help….. After all, one might have escaped. But what if that has happened? Those lions  are reasonably tame – and  there is still, perhaps, a probability that the lion would stay away from public areas for fear of being caught.  

To raise that probability level of a lion being loose would require more evidence – and strong evidence. The publican decided to phone the Zoo before calling the Police and the RSPCA.  

Whilst he was waiting for a reply from the zoo, he thought - how many Alsatians does one find in Regents Park?  

A quick investigation might show that there are many Alsatian owners who regularly walk their dogs in the park. This is high probability and requires less evidence to support it.  

So, on balance, high probability that it was an Alsatian – low probability it was a lion.  

Unless the call to the zoo suggests otherwise,  because more evidence on that side of the argument might increase the probability that it was a lion. Perhaps they have lost a lion, or maybe they are transporting one to Whipsnade, and it might have escaped without their knowledge. 

This is why research and investigation is needed in cases which are judged on the “balance of probabilities”. 

So, there is a recognised system on how probabilities are weighed and measured. Judges in the above homily would wish to hear strong evidence (from the zoo) before raising the level of probability that the creature was a Lion. That was the accusation, or complaint, so it requires strong evidence.  

Best evidence is the essential element.  The strong forensic evidence of a photograph of a lion roaming the park could establish the measure of probability. Counting the lions in the zoo would also help. 

The key point is that the balance of probabilities is not reached on a hunch – nor on the experience of the judge who simply looks at initial statements in a case.  

The likelihood of facts must be investigated. Probability must be forensically measured.  It is a scientific approach, because the judgement is based on established facts.  

There is nothing of this in Standing Orders. And, in spite of training, there appears to be no such measurement of probabilities in the  considerations of some connexional panels in the complaints process. They tend to work on the basis of what “is more likely”  or “more plausible”. This is a subjective approach.  

Panels tend to begin with the view that the respondents are right – and that the complainant must prove  51% that he is right. After all a man is innocent until proved guilty isn’t he? That is not how the system works.  

The procedure begins with equal probability or improbability on both sides;  the judges then measure evidence, and assess the balance.  

So, the three sources of evidence, verbal, documentary and the results of forensic investigation by the panel, must be brought together as completely as possible before they even begin to assess the evidence to lead to a conclusion.  

This applies equally to the initial investigation of the complaint, as to the main investigation of the complaint. (This aspect is considered elsewhere on this website, in an article about the initial investigation)  

This procedure is the only one with which SO 1133(8c): 

“The standard of proof required to establish a charge is the balance of probability.” 

  may  be followed if it is to be within the over-arching principles of justice and honesty set out in S.O. 1100(2).  

The probability of both sides must be investigated and considered.

CASE STUDY I

In a recent case, a supernumerary minister complained about a minor affair in his district. He spoke with an Acting District Chair, who then reported, when the complaint went to a connexional complaints panel,  that the minister had, in fact, demanded a post as Superintendent during that conversation. The minister involved was in his eighties.  

The rules on probability, as in the “lion example” above, are  particularly applicable in this case. An accusation was raised , how was it to be judged by the inquiry panel? 

The minister argued that if he had been a thirty year old man, this idea might be probable. 

But is it probable in the case of an eighty year old man? He had recently been taken into hospital and had a heart pacemaker fitted – did he, or indeed anyone,  really think that he was he fit enough to do the job of Superintendent?  

For the accusation ( note - from a respondent) that he had demanded the job to be judged probable,  very strong evidence would be required.  

The minister noted that, apart from the Acting Chair’s allegation, there was only one piece of evidence.  

He pointed to a letter in evidence which the inquiry had received from a friend of the Acting District Chair. He quoted the key line: 

“The complainant then presented a paper which proposed his own immediate appointment as acting superintendent.” 

The minister pointed out that the basis of this statement appeared to be an even earlier letter from a Church Steward, which, in turn,  detailed yet another letter which was actually written by another friend of the Acting District Chair  – who had nothing at all to do with the argument. This was the apparent source of the idea. 

He pointed out that the inquiry panel appeared to have made no attempt to see the original document, which was clearly the best evidence and which would have supported the probability one way or the other.  

“Chinese Whispers” had changed the content and emphasis of the original – but how could the panel know this, unless the first letter in this trail of letters was not produced? 

In fact the minister who was supposed to have demanded the job claimed that he had, in reality,  suggested that, if the Church was in difficulties - because the current Superintendent was seriously ill -  he, though a supernumerary, might step into the breach for a few weeks, because he had a lot of experience of the job and because it was a relatively quiet time of the year. 

This example – a true story - illustrates that, in measuring on the basis of probabilities,  the further from “best evidence” that the matter is, the less reliable is the measure of probability. The less probable the matter is, the stronger the evidence must be to support it.   

Step back from the argument for a moment - would an eighty year old supernumerary really demand to be made a Superintendent? 

The panel in this case actually accepted the evidence that the demand had been made. It should have realised that the evidence of probability it was not strong – because the minister involved was in his eighties.    

They should have attempted to find stronger (better) evidence to support the allegation. Perhaps a medical report would have helped.  Perhaps the supernumerary minister had told someone else that he wanted to be a Superintendent.  Perhaps he had written the idea in a letter somewhere.  

The particular panel considering this true case made no such efforts. Their reason may be that the person putting the claim forward was an Acting District Chair – whereas the person he wrote about was a supernumerary minister.

In this case, there appears to have been no attempt to assess the measure of probability. That is contrary to S.O. 1133(8c): 

“The standard of proof required to establish a charge is the balance of probability.” 

What’s worse is that, in this particular case, the panel did not  appear to be aware of this failing in its work. But that is common in the Complaints system of the Methodist Church, for there is no explanation in any published work of how the system of the balance of probabilities works.

CASE STUDY II

 This again is a real case. Another example of a lack of proper investigation occurred when a connexional panel considered events at a District Circuit Meeting.  

During consideration of an appointment,  an Acting District Chair intervened to stop discussion. One of the speakers was opposing the appointment. The minutes recorded: 

“Rev ….. came forward waving a paper at him saying “it was an email from the District Chair… that the present speaker was to be allowed four minutes to make any comment about the item.” 

Her contention was that the email was from the District Chair who was seriously ill at the time. The Connexional inquiry panel, which later considered this case, considered that the fact of her complete justification for intervention was “probable”.  

However, there was another side to the probabilities which the panel did not assess and measure.  

(i) The Acting District Chair  would not allow the minister who was objecting  to her intervention to see the email she brandished – either at the meeting or later. One might wonder  why not,  when it would have established her right to interrupt the proceedings?   

(ii) The Acting District Chair was actually on sabbatical during this period. There was no apparent reason why she might have been called in from that sabbatical to make the intervention – for another Acting District Chair  was actually in charge of the matter under debate and was available.  

(iii) The Acting District Chair made no mention of this email in her response to the grievance taken out against her about this incident. That would seem odd, for it would be powerful evidence that the District Chair had allowed her to intervene as she did. Why should she leave that evidence out of her response?

 (iv) The Acting District Chair later claimed that there had been an arrangement for her to take over the chair of the Circuit meeting – yet the minutes did not record that she did so.  

These questions raise a probability that the Acting District Chair who intervened at the meeting was exceeding her authority in by-passing he colleagues  in this meeting. Only the email would prove the contrary – and that vanished ( if it ever existed). 

How did that probability balance against the verbal assertion by the Acting District Chair?  We cannot know – for the panel did not work towards measuring that balance.   

In this example, the best evidence to use in finding the balance is the supposed email itself. In spite of repeated requests, the panel appeared to have made no move towards locating and securing that email.  

This again demonstrates that they did not understand how they should properly make a judgement based on the “balance of probabilities”. 

There is some further documentary evidence which might measure the balance in this example. In her response to the grievance against her, the Acting District Chair stated: 

“What follows is largely constructed from notes made at the meeting”

 If there were notes, those notes were “best evidence”. The panel should have asked for them. 

In fact, due to her actions, rather than her words, at the meeting in question, some members of the meeting began to suspect that the email she flourished was in fact the supposed “notes”. She  had no other paper with her.  

The key reason why the panel should have obtained the email is that the complaints which they were investigating concerned the manner in which the circuit meeting  was run and recorded. The content of the supposed email was the essential element in determining the manner in which the meeting was run. 

How many points on which probability could be measured were available to the panel – and how many did they actually use? 

It seems that their balance of probability measure was based solely on what the Acting District Chair said. As was once famously said in court “well, she would say that wouldn’t she? 

I do not criticise the Acting District Chair here. That is not the point. 

I criticise the panel – because it is quite clear that there was no desire at all to properly measure the level of probability in this dispute. Such was their duty – and they failed to carry it out.  

They trusted the Acting District Chair, they did not trust the lowly minister who took out the grievance against her. That was their simplistic ‘balance of probability’. They thought it “more likely” that she was right and the minister was wrong. On what basis of probability was that decision taken?

PROBLEMS WITH DELAY

Another important aspect of the balance of probabilities that is not mentioned in the Standing Orders is delay. The balance of probabilities is influenced by availability of evidence. The time limits enforced by Standing Orders may reduce the ability of the panel to properly operate the system. 

This is sometimes a problem if one of the parties cannot attend a hearing for reasons of ill-health, or does not tender proper, admissible,  evidence to the hearing. If such is because of reasonable or unavoidable delay, the judge cannot fully consider the balance of probabilities.  And that, of course, is contrary to Standing Orders. 

In other circumstances, a judge may demand to see certain evidence, particularly “best evidence” which is being withheld from him -  if he considers such evidence necessary to properly measure the balance of probabilities. The alternative might be for him to abandon the case. 

Occasionally, a  connexional complaints panel may decide that all help should be withheld from a complainant or a respondent. The decision would probably be taken in the light of SO 1157 (3) which allows a connexional panel to withdraw all help and support from someone who, it believes, has breached confidentiality. This may well  be done, even though one of the principles behind Standing Orders,  SO 1100 (3 iv),   states: 

“help and support should be offered both to the person making the complaint and to the person complained against at every stage during the process” 

The excuse that the panel will usually give for this is that such is reasonable and that, in any case,  it was “not disadvantageous” to the person involved. That of course is what they think. 

Not only is such a use of SO 1157 (3) contrary to SO 1100 (3iv), but it means that the whole system of the balance of probabilities is undermined. How can the panel possibly know that the withdrawal of all help, both with advice on procedure and the acquisition of documentation is not disadvantageous – when it has not seen the use that the complainant or respondent plans to make of it? 

The final decision is then made without all the relevant evidence before the panel – and this destroys the ability to operate the “balance of probabilities” system. 

This is particularly bad if the complainant or the respondent is requesting something that is “best evidence”.  

There is clearly a conflict between various Standing Orders here. Perhaps the most important, not mentioned in the above is SO 1100 (3 v): 

“The process should be fair.” 

It is clear that, should there be a suspected breach of confidence, then t there should be what in general law is called “ a trial within a trial.”  The issue of breach of confidentiality must be settled first -  if the complaint being investigated is to proceed.  

Unless this is done, a panel makes itself incapable of  properly considering  the balance of probabilities.  

There is an underlying point in this that should be observed in other situations.  

A tribunal, which is working on the basis of a balance of probabilities, may not intervene in the preparation and submission of evidence in a manner that unjustly tilts the balance of probabilities. This clearly makes the balance of probabilities an impossible goal.  

All evidence must be considered if the balance is to be fair and correct. In fact, the more evidence -  and the  stronger the evidence -  the tribunal, or panel,  considers, the better chance it has of assessing which is “best evidence”  - and therefore the best measurement of the balance of probabilities. 

It is one thing to fail to look for evidence – it is quite another to refuse to look at certain evidence.