WEBVTT

00:00:14.010 --> 00:00:16.258
So I'm going to talk about trust.

00:00:16.258 --> 00:00:18.902
And I'm going to start 
by reminding you

00:00:18.902 --> 00:00:22.749
of the standard views 
that people have about trust.

00:00:22.749 --> 00:00:24.499
I think these are so commonplace

00:00:24.499 --> 00:00:26.857
they've become cliches of our society.

00:00:26.857 --> 00:00:28.876
And I think there are three.

00:00:28.876 --> 00:00:30.613
One's a claim:

00:00:30.613 --> 00:00:35.542
there's been a great decline 
in trust. Very widely believed.

00:00:35.756 --> 00:00:40.547
The second is an aim:
we should have more trust.

00:00:40.547 --> 00:00:45.395
And the third is a task: 
we should rebuild trust.

00:00:46.332 --> 00:00:52.409
I think that the claim, the aim 
and the task are all misconceived.

00:00:52.409 --> 00:00:54.498
So what I'm going 
to try to tell you today

00:00:54.498 --> 00:00:58.896
is a different story about 
a claim, an aim and a task,

00:00:58.896 --> 00:01:03.713
which I think give one quite 
a lot better purchase on that matter.

00:01:03.713 --> 00:01:04.960
First the claim.

00:01:04.960 --> 00:01:07.804
Why do people think trust has declined?

00:01:07.804 --> 00:01:11.545
And if I really think about it 
on the basis of my own evidence,

00:01:11.545 --> 00:01:13.782
I don't know the answer.

00:01:13.782 --> 00:01:19.097
I don't -- I'm inclined to think 
it may have declined in some activities

00:01:19.097 --> 00:01:22.843
or some institutions and 
it might have grown in others.

00:01:22.843 --> 00:01:24.562
I don't have an overview.

00:01:24.929 --> 00:01:28.884
But of course I can look
at the opinion polls.

00:01:29.023 --> 00:01:31.376
And the opinion polls are supposedly

00:01:31.376 --> 00:01:34.912
the sourse of the belief 
that trust has declined.

00:01:35.604 --> 00:01:39.177
When you actually look 
at opinion polls across time,

00:01:39.177 --> 00:01:40.937
there is not much evidence for that.

00:01:41.029 --> 00:01:45.098
That's to say the people 
who were mistrusted 20 years ago

00:01:45.098 --> 00:01:49.467
principally journalists and politicians,
are still mistrusted.

00:01:49.467 --> 00:01:52.962
And the people who were 
highly trusted 20 years ago

00:01:52.962 --> 00:01:56.432
are sill rather highly trusted: 
judges, nurses.

00:01:56.432 --> 00:01:58.757
The rest of us is in between.

00:01:58.757 --> 00:02:01.497
And by the way, 
the average person in the street

00:02:01.497 --> 00:02:03.819
is almost exactly midway.

00:02:03.819 --> 00:02:05.750
But is that good evidence?

00:02:05.750 --> 00:02:10.031
What opinion polls record 
is of course opinions.

NOTE Paragraph

00:02:10.031 --> 00:02:11.375
What else can they record?

00:02:11.375 --> 00:02:13.138
So, they are looking at

00:02:13.138 --> 00:02:16.361
the generic attitudes 
that people report

00:02:16.361 --> 00:02:18.856
when you ask them certain questions.

00:02:18.856 --> 00:02:20.258
Do you trust politicians?

00:02:20.258 --> 00:02:22.657
Do you trust teachers?

00:02:22.657 --> 00:02:24.379
Now, if somebody said to you:

00:02:24.379 --> 00:02:26.096
"Do you trust greengrocers?"

00:02:26.096 --> 00:02:28.459
"Do you trust fishmongers?"

00:02:28.459 --> 00:02:31.038
"Do you trust elementary 
school teachers?"

00:02:31.038 --> 00:02:34.445
You would probably begin
by saying: "To do what?"

00:02:34.445 --> 00:02:37.742
And that would be 
a perfectly sensible response.

00:02:37.742 --> 00:02:39.839
And you might say,

00:02:39.839 --> 00:02:42.876
when you understood 
the answer to that,

00:02:42.876 --> 00:02:46.343
"Well, I trust some of them, 
but not others."

00:02:46.343 --> 00:02:48.046
That's a perfectly rational thing.

00:02:48.046 --> 00:02:50.949
In short, in our real lives,

00:02:50.949 --> 00:02:55.177
we seek to place trust
in a differenciated way.

00:02:55.177 --> 00:02:58.212
We don't make an assumption 
that the level of trust

00:02:58.212 --> 00:03:00.019
that we will have

00:03:00.019 --> 00:03:02.085
in every instance of a certain type

00:03:02.085 --> 00:03:05.658
of official office helder 
or type of person

00:03:05.658 --> 00:03:07.701
is going to be uniform.

00:03:07.855 --> 00:03:11.180
I might, for example, 
say that I certainly trust

00:03:11.180 --> 00:03:13.210
a certain elementary 
school teacher I know,

00:03:13.210 --> 00:03:16.345
to teach the reception class to read,

00:03:16.345 --> 00:03:19.874
but no way to drive the school mini bus.

00:03:19.874 --> 00:03:20.580
(Laughter)

00:03:20.580 --> 00:03:23.320
I might after all know 
that she wasn't a good driver.

00:03:23.320 --> 00:03:26.649
I might trust my most loquacious friend

00:03:26.649 --> 00:03:29.591
to keep a conversation going,

00:03:35.460 --> 00:03:38.127
but perhaps not to keep a secret.

00:03:38.373 --> 00:03:39.744
Simple.

00:03:39.744 --> 00:03:44.178
So if we've got those evidence 
in our ordinary lives,

00:03:44.178 --> 00:03:46.888
in the way that trust is differenciated,

00:03:46.888 --> 00:03:50.062
why do we sort of drop 
all that intelligence

00:03:50.062 --> 00:03:53.403
when we think about trust 
more abstractly?

00:03:53.403 --> 00:03:57.962
I think the polls are very bad guides 
to the level of trust

00:03:57.962 --> 00:04:01.417
that actually exists, because 
they try to obliterate

00:04:01.417 --> 00:04:05.984
the good judgement 
that goes into placing trust.

00:04:06.122 --> 00:04:08.337
Secondly, what about the aim?

00:04:08.337 --> 00:04:11.128
The aim is to have more trust.

00:04:11.128 --> 00:04:13.962
Well, frankly, I think 
that's a stupid aim.

00:04:14.392 --> 00:04:16.337
It's not what I would aim at.

00:04:16.337 --> 00:04:19.728
I would aim to have 
more trust in the trustworthy,

00:04:19.728 --> 00:04:22.211
but not in the untrustworthy.

00:04:22.211 --> 00:04:22.990
(Laughter)

00:04:22.990 --> 00:04:28.461
In fact, I aim positively to try 
not to trust the untrustworthy.

00:04:28.461 --> 00:04:29.726
(Laughter)

00:04:29.726 --> 00:04:33.366
And I think of those people 
who, for example,

00:04:33.366 --> 00:04:37.964
placed their savings with 
the very aptly named Mr Madoff,

00:04:37.964 --> 00:04:39.343
who then made off with them,

00:04:39.343 --> 00:04:40.515
(Laughter)

00:04:40.515 --> 00:04:41.627
And I think of them,

00:04:41.627 --> 00:04:44.736
I think: "Well, yes, too much trust."

00:04:44.736 --> 00:04:48.548
More trust is not 
an intelligent aim in this life.

00:04:48.548 --> 00:04:55.004
Intelligently placed and intelligently 
refused trust is the proper aim.

00:04:55.004 --> 00:04:56.591
Well one once said that,

00:04:56.591 --> 00:04:59.864
one says: "Yeah, ok, 
that means that what matters

00:04:59.864 --> 00:05:04.503
in the first place is not trust, 
but trustworthiness."

00:05:04.611 --> 00:05:09.171
It's judging how trustworthy 
people are in particular respects.

00:05:09.371 --> 00:05:12.599
And i think this judgement 
requires us to look at three things:

00:05:12.676 --> 00:05:15.158
Are they competent?

00:05:15.158 --> 00:05:16.506
Are they honest?

00:05:16.506 --> 00:05:17.951
Are they reliable?

00:05:17.951 --> 00:05:21.693
And if we find that a person is 
competent in the relevant matters,

00:05:21.693 --> 00:05:23.755
and reliable, and honest,

00:05:23.770 --> 00:05:26.057
we'll have a pretty good reason 
to trust them,

00:05:26.057 --> 00:05:28.277
because they'll be trustworthy.

00:05:28.277 --> 00:05:32.035
But if, on the other hand, 
they are unreliable, we might not.

00:05:32.035 --> 00:05:34.298
I have friends who 
are competent and honest,

00:05:34.298 --> 00:05:36.554
but I would not trust them 
to post a letter,

00:05:36.554 --> 00:05:37.929
because they are forgetful.

00:05:37.929 --> 00:05:38.738
(Laughter)

00:05:38.738 --> 00:05:40.986
I have friends who are very confident

00:05:40.986 --> 00:05:43.218
they can do certain things,

00:05:43.218 --> 00:05:47.501
but I realise that they overestimate 
their own competence.

00:05:47.501 --> 00:05:50.415
I'm very glad to say I don't think 
I have many friends

00:05:50.415 --> 00:05:53.835
who are competent and reliable, 
but extremely dishonest.

00:05:53.835 --> 00:05:55.003
(Laughter)

00:05:55.003 --> 00:05:56.658
If so, I haven't yet spotted it.

00:05:56.658 --> 00:05:58.248
(Laughter)

00:05:58.433 --> 00:06:00.411
But that's what we are looking for,

00:06:00.411 --> 00:06:02.798
trustworthiness before trust.

00:06:02.874 --> 00:06:04.776
Trust is the response.

00:06:04.776 --> 00:06:07.045
Trustworthiness is what we have to judge.

00:06:07.045 --> 00:06:08.661
And, of course, it is difficult.

00:06:08.830 --> 00:06:11.698
Across the last few decades,
we tried to construct

00:06:11.698 --> 00:06:15.256
systems of accountability 
for all sorts of institutions,

00:06:15.256 --> 00:06:17.737
and professionals, and officials,
and so on,

00:06:17.737 --> 00:06:22.095
that we'll make it easier for us 
to judge their trustworthiness.

00:06:22.095 --> 00:06:25.349
A lot of these systems 
have the converse effect.

00:06:25.349 --> 00:06:27.389
They don't work 
as they are supposed to.

00:06:27.389 --> 00:06:31.129
I remember I was talking 
with a midwife, who said:

00:06:31.129 --> 00:06:33.961
"Well, you see the problem is 
it takes longer to do

00:06:33.961 --> 00:06:36.527
the paper work than 
to deliver the baby."

00:06:36.527 --> 00:06:37.539
(Laughter)

00:06:37.539 --> 00:06:41.659
And all over our public life 
or institutional life,

00:06:41.659 --> 00:06:44.961
we find that problem 
that the system of accountability,

00:06:44.961 --> 00:06:50.329
that is meant to secure trustworthiness
and evidence of trustworthiness

00:06:50.329 --> 00:06:52.294
is actually doing the opposite.

00:06:52.294 --> 00:06:57.376
It is distracting people who have 
to do difficult tasks, like midwives

00:06:57.376 --> 00:07:00.045
from doing them, by requiring them

00:07:00.045 --> 00:07:02.181
to "tick the boxes", as we say.

00:07:02.181 --> 00:07:04.594
You can all give your own examples there.

00:07:04.886 --> 00:07:07.049
So, so much for the aim.

00:07:07.049 --> 00:07:09.626
The aim, I think, 
is more trustworthiness,

00:07:09.626 --> 00:07:13.577
and that's going to be different 
if we are trying to be trustworthy

00:07:13.577 --> 00:07:16.899
and communicate our trustworthiness 
to other people,

00:07:16.899 --> 00:07:20.338
and if we are trying to judge 
whether other people or office helders,

00:07:20.338 --> 00:07:22.712
or polititians are trustworthy.

00:07:22.712 --> 00:07:25.080
It's not easy, it is judgement,

00:07:25.080 --> 00:07:29.259
and simple reaction, attitudes don't do

00:07:29.259 --> 00:07:31.852
don't do adequately here.

00:07:32.236 --> 00:07:35.494
Now, thirdly, the task.

00:07:35.755 --> 00:07:38.418
Calling the task rebuilding trust,

00:07:38.418 --> 00:07:41.093
I think, also gets things backwards.

00:07:41.093 --> 00:07:46.081
It suggests that you 
and I should rebuild trust.

00:07:46.081 --> 00:07:48.295
Well, we could do that for ourselves.

00:07:48.295 --> 00:07:50.814
We can rebuild a bit of trustworthiness,

00:07:50.814 --> 00:07:54.182
we can do it two people together, 
trying to improve trust.

00:07:54.182 --> 00:07:57.766
But trust, in the end, is distinctive,

00:07:57.766 --> 00:08:00.502
because it's given by other people.

00:08:00.502 --> 00:08:03.345
You can't rebuild what 
other people gave you.

00:08:03.345 --> 00:08:09.487
You have to give them 
the basis for giving you their trust.

00:08:09.487 --> 00:08:12.837
So you have to, I think, be trustworthy

00:08:12.837 --> 00:08:15.044
and that, of course, 
is because you can't fool

00:08:15.044 --> 00:08:18.261
all of the people 
all of the time. Usually.

00:08:18.261 --> 00:08:19.123
(Laughter)

00:08:19.123 --> 00:08:23.128
But you also have 
to provide usable evidence

00:08:23.128 --> 00:08:25.037
that you are trustworthy.

00:08:25.037 --> 00:08:26.521
How to do it?

00:08:26.521 --> 00:08:29.295
Well, everyday, all over 
the place, it's being done

00:08:29.295 --> 00:08:33.960
by ordinary people, by officials, 
by institutions quite effectively.

00:08:33.960 --> 00:08:37.959
Let me give you 
a simple commercial example.

00:08:37.959 --> 00:08:39.959
The shop where I buy my socks

00:08:39.959 --> 00:08:42.251
says I may take them back.

00:08:42.251 --> 00:08:44.134
And they don't ask any questions,

00:08:44.134 --> 00:08:46.144
they take them back 
and give me the money

00:08:46.144 --> 00:08:48.710
or give me the pair of socks 
of the color I wanted.

00:08:48.710 --> 00:08:49.898
That's super.

00:08:49.898 --> 00:08:52.161
I trust them, because
they have made themselves

00:08:52.171 --> 00:08:54.096
vulnerable to me.

00:08:54.096 --> 00:08:56.038
There's a big lesson in that.

00:08:56.038 --> 00:08:59.037
If you make yourself vulnerable
to the other party,

00:08:59.037 --> 00:09:01.589
then that is very good evidence

00:09:01.589 --> 00:09:02.899
that you are trustworthy,

00:09:02.899 --> 00:09:06.225
and you have confidence 
in what you are saying.

00:09:06.225 --> 00:09:08.433
So, in the end, I think 
what we are aiming for

00:09:08.433 --> 00:09:11.878
is not very difficult to discern.

00:09:11.878 --> 00:09:15.496
It is relationships in which 
people are trustworthy

00:09:15.496 --> 00:09:18.628
and can judge when and how

00:09:18.628 --> 00:09:21.204
the other person is trustworthy.

00:09:21.204 --> 00:09:24.490
So the moral of all this is

00:09:24.490 --> 00:09:28.172
we need to think 
much less about trust,

00:09:28.172 --> 00:09:31.211
let alone about attitudes of trust,

00:09:31.211 --> 00:09:35.306
detected or misdetected by opinion polls,

00:09:35.306 --> 00:09:37.496
much more about being trustworthy

00:09:37.496 --> 00:09:40.045
and how you give people adequate,

00:09:40.045 --> 00:09:44.575
useful and simple evidence 
that you are trustworthy.

00:09:44.575 --> 00:09:45.627
Thanks.

00:09:45.627 --> 00:09:47.544
(Applause)