1. Background. In 1997, a public consultation was organised jointly by West Devon Borough Council (WDBC) and Devon County Council (DCC) on their scheme to enhance Bedford Square by removing two lanes of the carriageway and excluding all traffic from the area between the remaining two-lane carriageway and the Town Hall. Of 2981 respondents, 1186 (39.8%) favoured and 1795 (60.2%) rejected the scheme.
  2. 1.1 WDBC and DCC decided to try to devise a more acceptable scheme and, to this end, set up a representative Working Group comprising 2 members from each of WDBC, the Town Council (TTC), Civic Society, Access Group, Chamber of Commerce and Taxi Proprietors and one from the Business Association (13 in all). WDBC supplied a Lead Officer to assist the Working Group and provide technical assistance and liaison with DCC.

    1.2 Meetings of the Working Group were held in private and members of it were advised that their deliberations were confidential. The authority for this ruling has not been made clear. The WGp devised 3 schemes for consultation. All of these removed two lanes of the carriageway. The WGp was advised at the outset by the Lead Officer that no scheme would be considered for consultation unless this was done. It is not clear what authority there was for this. The Lead Officer told me that it was a stipulation of DCC officers, but the County Councillor claims that DCC officers were not involved in any way in the WGp's deliberations until very late in the process.

    1.3 One of the schemes (scheme C) excluded all traffic from the area between the remaining two-lane carriageway and the Town Hall (the Town Hall frontage). It differed from the 1997 plan only by the inclusion of a rank for 3 taxis and a narrow lay-by, designated for all the functions (other than parking) served by the present Town Hall frontage, viz drop-off/pick-up, loading, Community Bus. The other two schemes (A and B) allowed some parking, but traded this off against space for the lay-by.

    1.4 Public consultation took place in April 1998 and the result was published in May. The full figures were never published by WDBC or DCC except in committee reports, and were not reproduced in the public press. Annex A gives these figures at paragraph 1. WDBC's press release indicated only that "63% of those who favoured change" had chosen scheme C. Note that there were 1265 for scheme C and that this is just 73 more than the 1192 who declined to favour any of the three schemes offered.

  3. Interpretation.
    The 1997 result had been clear: with only 2 options to choose from, people either supported the WDBC/DCC plan or they did not. Only 39.8% had, so it was declared to have been rejected by 60.2%. In 1998, there had been four options, so the intentions of those who chose an "unsuccessful" scheme became important. Only one of the schemes attracted more support (73 more people) than "Status Quo/Limited change", and even that had attracted a smaller proportion (37.67%) of the total respondents than had the rejected 1997 scheme (39.8%). The consultation had not afforded any means to determine whether, for example, someone who chose scheme A or scheme B would also prefer scheme C to status quo. There was simply no evidence on this point: none had been sought.
  4. 2.1 Given this lack of evidence, it is clear that assumptions have to be made about the fallback intentions of respondents who chose a scheme other than the radical scheme C. It is also clear that various assumptions are possible and that different outcomes may very well flow from making different assumptions. Reasonable people might think that a reasonable way forward would be to calculate the outcomes from various reasonable hypotheses and consider which was the most likely in the light of such other information as might be available.

    2.2 This was not done. Only one interpretation of the ambiguous result was offered to WDBC's Planning & Development Committee on 9 June: that those who chose any of the schemes were in favour of change and could therefore be assumed to prefer scheme C to the status quo. Putting it another way, had there been a straight choice between schemes C and status quo, all those who had chosen A or B would instead have chosen scheme C. This interpretation was the only one offered to the WDBC/DCC Partnership Committee on 15 July. No evidence was offered for this assumption, nor do I know of any. You either believe it or you do not. This outcome is the one shown at para 2.3 of Annex A.

    2.3 An equally reasonable hypothesis, for which there is a good deal of anocdotal evidence, is that those who chose schemes A or B did so on the basis that they could see merit in some small changes in the square, but were categorically opposed, for example, to removing all the parking in front of the Town Hall. This assumption leads to the conclusion that, had there been a straight choice between scheme C and status quo, those who had chosen A or B would have switched to status quo rather than to scheme C. That hypothesis would lead to the conclusion that 61.07% of respondents had rejected the plan that WDBC/DCC now intend to impose. This outcome is shown in para 2.1 of Annex A.

    2.4 The 1997 consultation result is shown at Annex B. It provides the only reliable evidence available, apart from the anecdotal, against which to test these conflicting but equally reasonable hypotheses. The result was 60.2% against. This is strikingly close to the outcome of para 3.3, whereas to believe the assumption that must be made to justify WDBC/DCC's present intentions, we must believe that at least half of those opposed to the 1997 plan have changed their minds in the last 12 months. Readers may feel that this is a little fanciful.

    2.5 Furthermore, study of the actual figures from last year show that 377 more people responded this year than had done so in 1997, yet only 79 more people responded positively to scheme C than had voted "Yes" in 1997, a ratio of about 1:5, ie only one in every 5 additional respondents favoured radical change. This compares to 1186 from 2981, a ratio of 1:2.5 in favour of the 1997 scheme or some amendment of it and 1265 from 3358 (1:2.7) in favour of the radical scheme C in 1998. It is perhaps fanciful to draw firm conclusions from this, but one reasonable hypothesis would be that, in PL19 and PL20, the limit of support for radical schemes is about 1200 souls and that, as consultation turnout increases, so their proportion of it diminishes. This would explain the observation that, as the overall response has grown, the percentage support for the radical scheme has fallen.

    2.6 People must make up their own minds about the relative reasonableness, on balance of probability, of the diametrically conflicting hypotheses in 3.2 and 3.3. I know what I think. For the sake of completeness, I have also shown the outcome from a middle-ground assumption at para 2.2 of Annex A. There is no evidence for the assumption.

  5. Resolving the Ambiguity.
    Whatever one's preconceptions about a right and proper future for Bedford Square, councillors have a duty to take account of what the electorate thinks about the scheme that will actually result from all of this planning and consultation. Given that four options were available in 1998, it was inevitable that the result would be ambiguous. It was ambiguous because no attempt was made to discern the views of supporters of schemes A and B about scheme C. Therefore, assumptions had to be made, and the only interpretation offered to WDBC/DCC was that all supporters of A or B would also support C. It is also clear that this interpretation correlates poorly with the result of the 1997 consultation and fails to account for the apparent turnaround from 60.2% rejection in 1997 to 63.25% approval in 1998.
  6. 3.1 Conversely, explicit support for scheme C was 37.67% of respondents, which is slightly less than the 39.8% who supported the 1997 plan. It correlates closely with the 1997 result, as does the concomitant observation that the total response for Status Quo, minor changes, scheme A and scheme B together was 61.07% which is close to the 60.2% who rejected the 1997 plan.

  7. Anecdotal Evidence.
    The Town Council had concerns about the evenhandedness of some of the documents published during the consultation. Many felt that the questionnaire, and particularly Qs 1 to 6 and Q7 (which was not a question at all), tended to discourage people from choosing the straighforward "No change" option. A majority felt that the infamous "pink document", issued at the public exhibition, was not only tendentious, but actually misleading.
  8. 4.1 The Working Group was offered no advice from professional opinion research practitioners about techniques for avoiding unintentional bias in questionnaires. Nor was it advised that such advice ought to be sought whenever questionnaires are used to determine public opinion. I do not think for a moment that the Working Group intended the questionnaire to be tendentious. Equally, I doubt if it is possible for a group of lay people to construct a complicated questionnaire free from unintended bias unless there had been some professional validation by a qualified opinion research practitioner.

    4.2 I doubt if I am the only Town Councillor who has been made aware by members of the public that many of them felt unable honestly to respond "No" to all of Qs 1 to 6 and thereafter felt pressured by the instruction masquerading as Q7 to choose one of the schemes offered. Many people were unsure whether, if they chose "No" in spite of supporting (say) a wider footpath in front of the Town Hall, their return would be disqualified. I have a number of letters to this effect. I also have a letter from an "opinion research professional of 40 years' standing" who offered the view that the questionnaire was "designed to minimise the number of responses for 'no change'".

    4.3 Reasonable people may think it plausible that support for schemes A and B is more likely to represent an attempt to choose a "least damaging" scheme than an expression of support for the most radical scheme C. This interpretation best fits the 1997 result and is supported by anecdotal evidence.

  9. Was Scheme C Mathematically Certain? Cynics may feel that schemes A and B were little more than diversionary. The 1997 experience had shown that there was no more than 40% public support for radical change to Bedford Square. Assuming that there could be similar numbers determinedly opposed to any substantial change, any successful scheme would need to attract the support of the remaining 20%, as well as that of the radical element. Only a relatively anodyne scheme could do this. A radical outcome might be made to seem plausible, however, if two compromise schemes were offered as diversions, alongside the radical one, and agreement were secured beforehand to add together all the support for any of the schemes to "see whether there was general support for change" and then to declare that scheme which had most support of the three to be the "winner".
  10. 5.1 This arrangment ensured that it was mathematically impossible for either of the diversionary schemes to "win". Even if as much as 10% of the radical support were lost to one of the diversionary schemes and not a single "no-changer" could be persuaded by either of them, the result would be:
    No change 40%
    A+B 30%
    C 30%
    Result: C wins again, unless either A or B attracted zero support.

    5.2 This was the most extreme scenario possible. What in reality happened was that only a small proportion (2.13%) of the 1997 "Yes" vote moved to A/B 39.8% to 37.67%). Quite a large proportion (24.7%) of the 1997 "No" vote moved to A/B (60.2% to 35.5%). Neither A nor B attracted zero support. It was therefore impossible from the outset for either A or B to "win", given the rules estabished beforehand. The outcome of the 1998 Bedford Square Consultation was mathematically predetermined, once the Working Group had been persuaded of the reasonableness of the ground rules for interpreting the outcome. As the Working Group was sworn to secrecy about their deliberations until quite late in the day, the inevitability of the outcome could not be exposed until it was too late.

  11. A Way Forward - a Personal View.
    Given the apparent determination of WDBC to press forward with scheme C at high speed, so as to have contractors on site next February, perhaps mindful of Borough Council elections next May, I do not believe it would be realistic to press for amendments of scheme C. It is clear from the foregoing that it is unsafe to conclude that substantial support has been demonstrated for scheme C.
  12. 6.1 Given that WDBC and DCC are so committed to scheme C, it is unrealistic to look for the sort of minor changes that might have had widespread support, had they been genuinely on offer at the outset. I do not believe that they were ever on offer. There is no reason, however, given time for the dust to settle, not to pursue notions such as some widening of the footway along the Town Hall frontage and additional pedestrian refuges in the near future. These could be promoted as Local Safety Schemes within the delegated budget of the West Devon Partnership Committee. For the present, though, I regret to say that I think that Tavistock is back in the position of 18 months ago: fighting off an unwelcome, doctinaire scheme, imposed by a lengthy process of manipulative negotiation.

  13. Conclusions.
    The 1998 Bedford Square Consultation was flawed because the Working Group was not advised of the ambiguity that would inevitably result from having two diversionary schemes. The interpolation of the "compromise" schemes A and B alongside the radical scheme C made the shape of the consultation's outcome mathematically inevitable. Thereafter, the method of interpretation that had been agreed beforehand ensured that only scheme C could be declared the winner. Most people would call that a "fix".
  14. 7.1 The official conclusion that scheme C has majority public support is unsafe. It is almost certainly untrue. It is completely at variants with the outcome of the 1997 consultation, which was unambiguous. This conflict can be explained only by interpreting the figures in close accordance with the assumptions in Annex A para 2.3. This leads to the conclusion that the public does not support scheme C and WDBC/DCC has no remit from the public to implement it.

12 July 1998
Back to main document




1. Full Analysis by Number of Signatories

Running Total
Status Quo:90226.86%
A 524 15.60% 1716 51.10%
B 335 9.97% 2051 61.07%
C 1265 37.67% 3316 98.74%
Variation 33 0.98% 3349
Spoilt 9 0.27% 3358
NOTE Scheme C has 73 more votes than Status Quo + Limited

2. Interpretation

2.1 Assume all A & B would prefer Status Quo or Limited to C:

Conclusion: scheme rejected by 61.07%

Observation: corresponds well to 1997 consultation result 60.2% rejection). Members may feel that this is the most plausible explanation for the otherwise unaccounted for turnaround this time.

2.2 Assume all A would prefer Status Quo or Limited to C:

Conclusion: scheme rejected by 51.1%

Observation: Scheme A may have been seen as adversely affecting fewest communities of interest. There is substantial anecdotal evidence from local members that many "A" votes were frustrated "Limited" votes.

2.3 Assume all A B & C would prefer C to Status Quo.

Conclusion: scheme accepted by 63.25%

Observation: This is the interpretation offered by the officers' report. It correlates poorly with the 1997 result and no evidence has been adduced in the report to support the conclusion.

3. Conclusion.

The truth probably lies somewhere in between. Given that only 73 votes separate Status Quo + Limited change from Option C, it is unsafe to conclude that substantial support has been demonstrated for Option C.

Back to main Bedford Square document



Running Totals
Conclusion: The 1997 scheme was rejected by 60.2%

Observations: The total number of respondents in 1997 was 3358-2981 = 377 fewer than in 1998, yet the number in favour of the scheme or some amendment thereof was only 1265-1186 = 79 fewer than those supporting Scheme C in 1998. This would be consistent with the notion that support for a radical option for the Square tops out at around 1200 people in PL19 and PL20.

Back to main Bedford Square document