Prof Hibbert: - his Excellency General the Hon David Hurley Governor of NSW to come and open up the 2017 Royal Society of NSW and for academies for (indistinct).

The Hon David Hurley: Ladies and gentlemen, again a very warm welcome to Government House this morning and delighted to have you here for it, I think you'll be very (indistinct) an interesting day. But I acknowledge, of course, Professor Hibbert, Professor Mary O'Kane, distinguished Law Society and Academy Fellows and their representatives and the organisation presenters and members.

I began these series of forums three years ago when I first became Governor and trying to work out how the Governor does his job because it was done manually at the time of the appointment. And much to my surprise I found that with the three Cs, constitutional, ceremonial and community engagements roles of the Governor, constitutional, ceremonial (indistinct) about 10 per cent of my time. 90 per cent of my time was involved with the people of New South Wales.

So I did some strategic planning and came up with a business plan for the House and myself and trying to work out how do you value add in this role to the community. And one of the areas I thought about when looking at my predecessors was the role of Governor Brisbane in the establishment of the Law Society and why was that link in place? Obviously, the roles and functions, the authorities of Governors have changed since Brisbane's days. But the role of the Governor in trying to help the development of [fort] in the early community and the intellectual life of the early community and seeing the great potential that existed in Australia at the time, so could I follow in those footsteps (indistinct).

And I thought one of the things I could do as Patron of the Royal Society is help with the Royal Society to provide an opportunity to have a think tank here at Government House where we could look at some of the bigger issues that are facing us today in a political sense, in a neutral academic environment and so that's the course we have undertaken. And, as I say, look at difficult issues. It's often hard to have discourse and discussion in public life these days without the divisiveness of being drawn to people's attention. If you have two views obviously there must be division and division creates conflict, conflict create news. That seems to be the way of (indistinct). We're not about that, we're about examining issues of importance to our society.

The topic we'll look at today is not new, many of the discussions I had with people last night, but there are aspects of it that have changed. HL Mencken, you may have different views about him as a person but he's very rich in comments about democracy. Democracy is a pathetic belief in the collective wisdom of individual ignorance. 1926 or so that was written. He has a view that, again part of a

discussion last night, our right to individual speech and our right to have an opinion does not necessarily make that opinion right in itself and therefore how do we engage with the community, with people, with institutions, with policy makers when sometimes I think our ideas are (indistinct).

We now have, of course, a transient fake news, alternative facts, post-truth discourse, not new perhaps different titles. Of course, post-truth was the word of the year for 2016 in the Oxford Dictionary. It has now created an industry (indistinct), many books are written on post-truth.

So is it something new or old with a new title? Is it a new phenomenon that we are trying to deal with and when we are trying to deal with? Is it the result of the staggering growth in information data that's available and the social media at the present moment, which has brought it to the surface? IBM, I think I quoted this a number of years ago, IBM has estimated that the amount of data generated in 2002 for the whole year is now created every two days.

Or is something more concerning in play? If we look at the history of our civilisation, primarily western civilisation, rationality has been one of the sort of fundamental stones of which we have stood upon in western society as we have developed our civilisation.

A number of years ago I did a post-graduate course at Deakin University and I had to do a paper on rational decision making and a decision to have a second airport in Sydney. This was in 1993. And I came to the conclusion that we were far away from that decision because if you look at the process we were going through we were not making a rational decision about a second airport. I lay no claim to any decision that's been made recently.

But what are the alternatives to rationality? Of course, objective belief, faith, selective opinions, stand on my ground. So what do they mean for science, for society, for democracies as we know it, therefore our future? Are these really threats or are they impacts that new technology, new ways of doing business have introduced to the society that we have (indistinct)? Is democracy on the decline? Is this a threat to democracy that will increase that decline or are we going through a growth spurt in democracy where it is just (indistinct) democracy has been challenged before?

So why do we wring our hands instead of girding our loins? Believe in it, defend it, we promote it, we take it forward. I go back to Orwell though, a quote from a letter he wrote in 1944, "I fear the horrors of emotional nationalism and a tendency to disbelieve in the existence of objective truth because all the facts have to fit in with the words and the prophecies of some infallible fuehrer. Already any history has in

a sense ceased to exist. That is there is no such thing as a history of our own times which can be universally accepted, and the exact sciences are endangered as soon as military necessity ceases to keep people up to the mark. But if the sort of world that I'm afraid of arrives, the world of two or three great superstates, which are unable to conquer one another, two and two could become five, if the fuehrer wished it." 1944.

I could quote from 1984, which I did last night, about the falsification of history. "I know, of course, that the past is falsified but it would never be possible for me to prove it even when I did the falsification myself. After the thing is done no evidence ever remains. The only evidence is inside my own mind."

So is this another aspect of the issue that we're talking about today? What is truth? What is post-truth? How do we deal with it as a democracy and a society? More importantly, how do we assist decision makers in performing their duties? And that's what we should be (indistinct), to assist, to enable, to take our society forward. So the plus from today I hope is bringing together four academies, which may not on a daily basis come together, that's one of the purposes of this forum, collaborate.

As an aside I did ask the Vice Chancellor's Committee of all our New South Wales Universities to do a task for me a few years ago and which when I made my pitch the first question back was "Do you want us to work together?" Yes, that would be nice. Okay. But I shouldn’t go any further down that track.

Today I think is a day for some very intriguing presentations. I think at the end of it hopefully we can come out of this (indistinct). My desire would be (indistinct portion).

Prof Hibbert: Thank you. I'd just like to say a couple of things about the way that today will happen. The last two forums (indistinct) panel discussion and the people decided that that didn’t work all that well (indistinct) last year, so we had to put questions and comments at the presentation (indistinct). But it will be Paul Griffiths' job to try and (indistinct) the time. I'd also like to acknowledge our two overseas speakers who are running around the world, James Wilsdon and Sir Peter Gluckman. They were both going to be in Sydney and agreed to speak here today but then James Wilsdon got a senior academic position at his university and had to go home. And Sir Peter, of course, got a new New Zealand Government, which has rather serious calls on his time. But both of them very kindly have accommodated our various problems.

So James Wilsdon did his very best to record his talk, failed late last night but, in fact, we have the text of his talk (indistinct) will be presented by Paul Griffiths for

which we thank him for doing that at short notice. Sir Peter Gluckman was on a plane this morning coming from New Zealand and arrived (indistinct portion) and then had to fly back home. (Indistinct portion).

We're formally supported by the Office of the New South Wales Chief Scientists and Engineer and in recognition of that support and indeed all the contributions made by the Chief Scientists Convention with the Royal Society of New South Wales, I should now like to introduce our other most favourite person other than His Excellency, Professor Mary O'Kane to make some opening remarks. Thank you.

Prof O'Kane: Thank you, Brynn. And above all, thank you, your Excellency, for hosting this and encouraging us so much. It's wonderful to be able to sponsor this and to know that there is such a distinguished audience here taking part in the discussions today. In speaking I too would like to acknowledge the traditional owners of the land, the (indistinct) people of the (indistinct) Nation and pay my respects to those past, present and future.

I'm only going to make a short number of comments and from a very particular angle of the role of the Government, as Peter Gluckman would say, Chief Science Advisor. It's a very timely topic, and his Excellency has pointed out, it's not a new (indistinct). Rationality indeed is a very old term and the concept is very old and one that we value greatly.

I see great examples of what I think we're talking about today. And I was amused when we had the science awards here a few weeks ago, his Excellency was clearly thinking about what he'd say. And at that point I was thinking about what I'd say and I'm sure all the speakers today have been challenged by the topic to know just what is it that we are dealing with here. And in my role as Chief Science and Engineer, and as I'm leaving it very soon, I've used this opportunity just to sort of think back over the sort of things we've encountered and what it sort of means in a wider context.

The Chief Sciences and Engineer is the bigger part of the role, so the Governor talked about his role, my role has been there's a certain element of promoting science and innovation, very important. But a large part of it is dealing with the wicked problems, the wicked policy problems involving science and engineering that the Government sends to the Chief Sciences and Engineer for some enlightenment or some comment or whatever.

And if you look up what is a wicked problem and you say the Wiki definition, a wicked problem is a problem that's difficult or impossible to solve because of

incomplete, contradictory and changing requirements that are often difficult to recognise. People often criticise Wikipedia, it's a brilliant definition.

It goes on to say "The use of the term wicked here has come to denote resistance to resolution rather than evil. Moreover, because of complex interdependencies the effort to solve one aspect of a wicked problem may reveal or create other problems." I can only give you several examples of how true that is.

Some of the examples I've dealt with in the nine years I've been in the role include one we're probably best known for, the review of coal seam gas activities in New South Wales get a run in the Fin about once a week or once at least every two weeks. More recently the review into the decline of koala populations and the koala State strategy, the State koala strategy. I never quite know what a State koala is. The Energy Security Taskforce that has taken so much of this year and has been so very fascinating and, of course, also being involved in Alan Finkel's review.

The work on rail coal dust emissions in the Hunter. The very controversial work at the moment on tunnel air quality, which we've been doing for some years but it's one that keeps turning up and brings us onto the front of the Sydney Morning Herald, as you'll have seen a few Mondays ago. The PFOS contamination whether it be at Williamtown, the most difficult problem in the State, the Shoalhaven, and in many sites across the State, particularly as it comes - the Williamtown one it comes off Commonwealth land, so we've got all the Constitutional problems that go with that.

Sea level rise, rat population on Lord Howe and medicinal cannabis, another great giving topic in this area. And it's over those areas we see some incredible behaviours. And a lot of them don't look at all rational, a lot of them look very, very strange when we start in on the inquiries. They also are areas where people get enormously distressed, enormously worked up, and so we're dealing with a community where the behaviour is not always particularly rational.

We also see very interesting coalitions, famously in the Lock The Gate Coalition you saw the farmers and particularly the NFF, National Farmers' Federation working with the Greens and Lock The Gate in a very tight and effective coalition.

So what's going on? I'll just sort of unpack the coal seam gas one a bit. The coal seam gas issue is relatively straightforward. Coal seam gas technology is very well developed in the States in it's great hunt for coal bed methane, which is the same thing, different words. And how to drill for it, how to frack where necessary is something that is not hard and not difficult to do. And it's actually the notion of gas from coal seams, of course, you know, it's generally a by-product you want to

get rid of. And indeed, in the war here in Sydney a lot of cars were powered by gas from the Balmain Colliery. In fact, I actually live right over that colliery and right beside the gas well. No problems with that.

But what was very clear in the inquiry was that while the technology was straightforward and the technology do to deal with that produced water was reasonably straightforward. The regulatory arrangements were not. And the notion because it's a very distributed industry of getting appropriate regulation in place and making sure it was enforced turned into something of a challenge, so that was one thing. There were land access issues where some of the people in the early 2000s who'd had exploration licences caused a lot of trouble going onto land and that's where the term lock the gate, of course, comes from.

There was a great fear of fracking. Why? The belief that chemicals would lead to water pollution and there were health concerns, really big health concerns round that. A lot of that, of course, was prompted from the movie Gasland but was sort of amplified here. There was a belief that fracking could cause - well, it does cause induced seismicity but that this could cause, you know, major earthquakes. There was worries about subsidence. There was worries, the greater whole issue of what did this do to ground water to surface water. And a great belief that in many cases that it would completely wreck Australia's aquifers and that they were the life blood of Australia and the whole country sort of imploded (indistinct).

And, of course, a belief that the aquifer could cause the drawing down of water and there'd be a lot less water available for farming and other activities. There was the concern about gas in the drinking water and as anybody who's lived on a farm with lots of (indistinct) Australia will know, there's been methane in the water forever. But pictures of Jeremy Buckingham lighting the Condamine certainly got people interested when he almost fell out of the boat, something he enjoys talking about.

Very great worries about produced water. Worries about radioactive produced water. Worries about dealing with the salts that came from produced water. Worries about impacts on air near the coal seam gas wells. Massive amount of discussion and we spent a lot of time trying to understand what these issues were and how serious they were, as you know, came out and said that the technology was perfectly applicable but the (indistinct) had to change and there needed to be good communication and very good consultation.

But let me then say what was really going on with all these issues and there was a basis with most of them to some extent. Yes, you can get subsidence where you've had a lot of gas wells but it's very, very small in many cases. And we spent

a lot of time with Geoscience Australia working out how to measure it remotely from satellite using the very - Australia was incredibly good in the geomatic space and even with all our expertise we can just measure it in certain places.

So there were things but it had been badly amplified, so what was going on? I think there were three things. First of all there was what the Governor referred to as sort of the post-factor issues. Maybe he used a slightly different term, (indistinct). But where people had a starting point for their reasoning that was very odd for whatever reason. And it really was brought home to me at a consultation at Campbelltown in the Coal Seam Gas Inquiry because we did loads of consultation where two Carmelite nuns came, it would be two, of course, and one of them stayed relatively silent as Carmelites often do. But the other one was, you know, somebody worthy of [Aquinas] in the way she reasoned and went through a series of arguments to explain why coal seam gas was a total disaster.

I could barely believe it because she was saying all sorts of odd things (indistinct), you know, conclusions from it were deeply logic. And I said to her at the end quietly afterwards, you know, (indistinct portion) I could tell her why they were wrong. And we talked about it and realised that because of her personal life all she had to go on was (indistinct portion) respect for anyone who works on the Sydney Morning Herald (indistinct portion) Herald. And because this was her source of (indistinct) she had decided there really was (indistinct portion) in many cases people were not trusting the scientists or not trusting official people in Government (indistinct portion) gurus and they were believing them thinking they were more trustworthy.

So what happened? These people were often very rational (indistinct portion). And this often was people who (indistinct portion) very sensible but didn’t have time to investigate (indistinct portion). They point to the fact of chemical spills issue, you know, the PFOS issue at (indistinct portion). They pointed to Chernobyl (indistinct portion). And they say we can't trust science where there are downsides to it. (Indistinct portion). You know, it's that issue and I think there's a lot of (indistinct portion).

(Indistinct portion) the negatives in a relative sense and then reason from there (indistinct portion) where people are out there campaigning and very worried but the issue is something close to their hearts that they often feel they can't say because they'll look (indistinct). The biggest one in this issue is land value. Some people will talk about it openly, others come at it in a very crab like sideways way where they're very worried about coal seam gas on their property or, and I think more genuinely in the case of the PFOS, although the coal seam gas can do it too, it's going to drop the value of the land.

It took me a long time to see it in certain cases but in the Liverpool Plains the worry about coal seam gas really often turned about the fact that people were then having problems having bank loans. They were already highly mortgaged and the thought the coal seam gas was coming was causing banks to freeze somewhat. So even if the value hadn't gone down it was expressing itself in difficulty with loans. Certainly, at Williamtown it has come through in a drop-in value in the contamination there.

So land value. Identity is another issue. I live at Williamtown, I live an alternative life where I feed myself and my family, I live off the fish, we home slaughter meat, I eat vegetables and you can't do it because of contamination. It's completely changed the lifestyle I live. It's another very big, and I think very legitimate argument.

There's the don't trust Government. So they won't necessarily say it to me but they will - all the public service, that's the bit they won't really talk about to me, but it's very clear they don't. And then you get the things that sometimes are where people are quite upset that somebody else has benefited. So later I think Simon Chapman's going to talk about the windmill issue. I'd just make a point that one of the things we saw with some of the windmill issues was that people were distressed, that their neighbour got the windmill and all the money that went with it and they had the downside of trucks going over their land on the way to it and got very little money from it. And every day they got to see the blasted windmill that was giving the neighbour a lot of money.

There are things like that and often people don't wish to appear selfish and won't say what the real reason is but it is distinctly rational to oppose what's come in. So I think what I'm saying here is that in the post-truth world there is often great rationality but you need to go back to what is the truth or the lack of it. So what do we do in this case? As many of you know, the way we've tried to approach these things is to try and understand the issue, the real issue, to try and build some level of trust and to try and provide some level of information. We listen, we look at the literature, the grey literature, the scientific literature. We hold massive community consultations. We work closely with stakeholder groups. We work hard to be seen, to manage conflict of interests and be seen to manage them.

We recruit a range of experts in parallel to write for us, so that there are many opinions and run independently. We publish reports, we investigate compliance. Chris Armstrong, my deputy, developed a wonderful system of workshops where he would bring together people from all the stakeholder groups to solve a problem. So one issue was how the water regulations should be changed and he got everyone into the room and they'd be from Lock The Gate, they would be from the

Government agencies, they would be from the coal seam gas drillers, and he would get them to all work on the problem. And when you'd go in in the morning they'd all be looking very frosty and people were sitting in groups. By morning tea people were being polite. By lunchtime they were working with butcher's paper together and by the end of the day they'd be going to the pub. And it was wonderful to see it.

We also worked very heavily with the press. We very strongly promoted open data and open Government. We're very careful with our language. And I noted what the Governor said about language last night. Really careful because we note the pejorative use of language, like chemical equals bad, renewable energy equals good and, you know, very careful not to get into that.

So in conclusion where does this leave us? I think it leaves us with an issue where we need to go back on many fronts to do the sort of things I've been talking about that we did in the office. But also to go back to education, particularly emphasising how people should analyse and pose and understand problems and how to specify a problem and that needs to be a strong generic skill we teach kids.

But the other thing I think we need is to educate people in the history of science to understand the contributions of science while also understanding that there are some downsides. And to me I think even if we can just do that it's important. It's also important, as Peter Gluckman will I think say to you, that people in roles like mine are not seen as advocates, they are just advisors. It's very much a case of just the facts, ma'am. And it's not to say other scientists shouldn’t be advocates but that in Government in building trust it must be seen as just the facts.

And I think it's so important that we do today, so important that we actually do talk about these issues because as I've seen they can cause massive trouble and massive distress. And I saw a great quote in the Financial Times the other day when Richard Flanagan, the Booker Prize winner who lives in Tasmania, was talking about his new book called "First Person". And it's really about - it's a fictionalised account of John Frederick the great conman that you'll remember, you know, particularly in Victoria. And Flanagan says, "The terror of becoming the first person trapped in someone else's lies of having unknowingly traded my freedom for oppression," and then he goes on to say "It's a bad thing."

And I think that is what is often happening in our world today of people are somewhat unknowingly have traded freedom from oppression because they've been trapped in someone else's lies. And I think it's our job to try and work out how that doesn’t trap you. Thank you.

Prof Hibbert: Thank you very much, Mary, for starting us off. I was going to say the business of the day now starts in earnest but I think we've already started in earnest with your excellent remarks. And so I call upon Professor Paul Griffiths, Fellow of the Australian Academy of the Humanities, Fellow of the Royal Society of New South Wales, to take charge of the proceedings and introduce our first speaker.

Prof Griffiths: And so it gives me great pleasure to introduce the Vice President of the Royal Society Donald Hector AM. Donald's based in Sydney but works around the world with clients in Australia and the UK. He's somebody who works in the intersection of science and industry with extensive experience in large scale industrial processing, the evaluation and commercialisation of new technologies. He's a former Managing Director of Dow Corning in Australia, also Asia and Pacific Speciality Chemicals. He's been a non-executive Chairman and Director of many ASX listed and private companies and also non-profit organisations. And Donald's going to frame the issue for us by introducing us to the problem we're confronting today, that of rationality post-truth and the threat that it poses to (indistinct).

Prof Griffiths: Your Excellency, colleagues. Well, what a mess. Why can't people be sensible? Wherever we turn there are astrologers, homeopaths, conspiracy theorists, miracle workers and anti-vacsers. The politicians prefer to follow their gut instinct rather than evidence based rationale. The Internet has made everyone an expert. Two thirds of Americans get at least some of their news from Facebook and over half get some from Twitter. How much substance can be embedded in 140 characters? Is it true that only twits tweet?

Are the post-modernists and the relativists in the ascendency? What has happened to reason? The great enlightenment philosopher David Hume said, "Errors in religion are dangerous but errors in philosophy are only ridiculous." This was one of his rare mistakes. Rejecting established sources of reason and accepting that belief should have equal sway with fact puts an open free society in great danger.

The advances made in human civilisation in the last 600 years have been greater than the previous 60,000. In 1840 there was no country in the world where the life expectancy at birth was greater than 40 years. Today, just 180 years later, there is no country in the world where the life expectancy is less than 40 years and there are several countries where it's double this.

The rediscovery of Greek philosophy during the Renaissance, the emergence of the scientific mathematics, flourishing art, music and literature together brought about the agricultural revolution, the scientific revolution, the industrial revolution at an extraordinary period of human creativity. Of these the scientific revolution was

the most important because it changed the fundamental paradigm of Middle Ages, Christianity and the ancient world. Belief gradually gave way to evidence of reason.

The gains were greatest and emerged earliest in what are now called the developed countries, most particularly those of Western Europe and North America, but the phenomenon has now spread worldwide. Today most prosperous countries share a common feature. Although far from perfect they have developed or adopted institutions in areas of law, politics, health, education and social institutions such as universities and a free press.

That places great value on evidence and fact. These institutions are the foundations of today's civil society. In such an environment when inquiry is rigorous and subject to review by one's peers and key to this is our notion or modern notion of knowledge. As the Oxford English Dictionary puts it "The apprehension of fact or truth with the mind. Clear and certain perception of fact or truth. The State or condition of knowing fact or truth."

Why is this emphasis on truth so important? Because it led to the settling of disputes with evidence and reason rather than force and this then became the foundations of institutions that people could trust. Today's topic truth, rationality and post-truth is important because of the threat to these institutions posed by the emergence of post-truth. What is post-truth? Simply that objective facts are less influential in shaping political debate or public opinion than appeals for emotion and personal belief. One might be tempted to say that Twitter Trumps fact.

I'll discuss truth and then I'll take a look at rationality and then I'll briefly outline why I think post-truth is so dangerous. But first I'll make three statements upon which my subsequent remarks are based. First, there is a physical world independent of human thought. If I cease to exist the world keeps going. Second, from birth every human acquires a body of knowledge that represents the physical world they experience through their senses. This is their subjective knowledge, it sits within them. And third, there is an independent body of knowledge that's been developed through human thought and communication. This is objective knowledge. It's the full range of shared ideas such as stories, writings, art, music mathematics and so on.

As far as I know the first philosopher to bring this together quite so succinctly was Karl Popper. It wasn’t original. Popper drew upon philosophical thinking that had emerged over the previous two millennia but he did put it very clearly. He referred to these as three worlds and claimed that they are this three distinct ontological

states. Now, some philosophers will dispute this but it's a pretty good way of thinking about things in the context of today's discussion.

Now, in considering truth philosophers generally look at two issues. What is meant by the words is true? That is to say truth predicate. And the second is the criteria for truth. For example, if I say the book is blue how do I determine whether or not the book really is blue? An example might show why this distinction is so important. Pontius Pilate was famously reported to have asked the question, "What is truth?" He got it wrong. He should have asked "Is he guilty?"

And the point is that it's important not to mix up what the question of truth is versus whether or not we can establish whether something is true or false. I'll expand more on the meaning of truth in my written paper that will be published in the journal but today I'll just refer to the two theoretical approaches that have been most influential in the last hundred years or so. These are the correspondence theory and the coherence theory.

Now, both of these are called substantive approaches. They hold that a notion, the truth, exists and it has two parts. The first part is a property of a statement that purports to represent a truth and this is called a truth bearer. The second is a theoretical cognisor, someone who can critically evaluate the truth bearer. Correspondence approaches propose that truth is contained entirely within the truth bearer. That is to say the statement about whether or not something is true. And that it represents the way the world is, so it's independent from the cognisor and should be able to determined objectively.

Coherence approaches on the other hand argue that truth is contained not only in the truth bearer but also includes relationships between the truth bearer's statements and the cognisor, so it contains an element of subjectivity and it's not independent from the cognisor. If, as I do, you have the view that a physical world exists, what Popper called World 1, and that's independent of the human mind, then it follows that there must be truth bearers that can be independently and objectively evaluated. That is to say observations about the physical world must be viewed from a correspondence perspective.

Hence science is predominantly about correspondence, making propositions and evaluating them independent of the observer. Now, there are all sorts of philosophical objections to this. There's a strong argument that much scientific inquiry is socially determined even down to the questions that scientists decide to investigate. But it's difficult to avoid the conclusion that there should be able to be truth bearers formulated that can be objectively evaluated even if we could never really achieve truth observer independence.

On the other hand Popper's World 2 and many World 3 phenomenon, that is to say the subjective and objective knowledge cannot be dissociated from the cognisor because they're entirely products of human thought. Thus they can only be evaluated using the coherence approach.

Now, surprisingly the philosophical literature of the last century or so puts the correspondence theory and the coherence theory in opposition to each other. But even if you accept that dubious claim, it's only the case when they're referred to as or considered as to be as definitional theories of truth. That is actually referring to the meaning of truth. But when they're just used in the context of criteria for truth to determine whether something is true or false, then the two approaches be complimentary. And the only theoretical limitation is that they can't provide sufficient justification to determine truth with absolute certainty.

Now, there's a fair few words in that, so maybe I can give you an example that might clarify it. I can make a statement the book is blue and assert that that statement contains the truth. The coherence theorist might then ask, well, what do I actually mean by blue? What happens when I perceive blue light? Is it the same as someone else's perception of blues? Difficult questions.

The correspondence theorist on the other hand says that the statement doesn’t require someone to think about it in those terms, it's either it's true or it's false. Now, I can take a spectrophotometer, I can measure the wavelength of the light and if it's 475 nanometres, around about that area, then I can say that's blue because that is what blue light is.

But that's not really a complete answer. The most complete answer lies in the combination of both the coherence and correspondence approaches. If the light is at 475 nanometres it's blue, so the book is blue. But the perception of blueness may be different from person to person. Now, I happen to be colour blind. I know that my - or I'm fairly confident that my perception of blueness is different to about 93 per cent of the men in this room and about 99.5 of the women in this room. So we can't be certain how another person perceives blueness.

But science can provide us with the means to find an objective answer to the question and I think that's the important point. It must come back to scientific objective inquiry wherever we can. And I think this distinction is at the heart of the point that CP Snow was making at his famous and controversial lecture The Two Cultures. The scientific method is largely based on the correspondence approach. But recognising that some questions are socially determined whereas the social science and the humanities refer more to the coherence approach because of the subjectivity in most of the issues that they consider.

The problem is that scientists and technologists are reluctant to recognise the social determinants that influence their research and the outcomes while those in the humanities and social sciences can be dismissive of expert opinion even when it's based on overwhelming scientific evidence. If we really want to see knowledge advance we should recognise the importance of both approaches to truth and use them together.

Now let me turn to the subject of rationality. All conscious animals need to make sense of the uncertainty they encounter in the world and must either control it or adapt to it. To do this they form mental representations of the world based on the information they receive through their senses. They then react to this and behave accordingly. That is they form a theory of reality, a world theory by which they relate their own existence to the real world phenomenon they encounter.

This form of cognition is intuitive. All animals have intuitive thinking in their thought processes. In humans intuitive thinking is experiential; it relies heavily on visual insight and the recognition of patterns that emerge from complex systems. It's oriented towards immediate action, our reflexes are governed by it, and it leads to the formation of images, which are persistent and very slow to change. It's very hard to change your intuitive images.

Intuition is experienced both passively and subconsciously and is heavily affected by belief and emotion. Judgments arising from intuition are compelling and they appear to be infallible, they appear certain to us. To the extent that people are often seen as irrational if they don't share our intuitive judgments about things. We just find it very difficult to relate to them. It's often thought of as being imaginative, creative and even mysterious.

But humans have also developed a second form of thought, which is analytical in nature, and it's based on reason. This form of cognition is logical and it derives from conscious understanding and appraisal of real world phenomena in the context of your own thoughts. Analytical thinking is slower to process but it can change rapidly, eureka moments. It exists in the abstract and is denoted through language and other symbols such as numbers. Unlike intuition analytical cognition is deliberate and conscious. The individual controls its own thoughts and has the capacity for self-awareness and to be self-reflective. It's based on evidence and logic even though the logic might be flawed.

But importantly, it's retraceable, so we can go back through our analytical reasoning and look for mistakes and correct them. Thus the complete theory of reality for a human is a cognitive system consisting of a world theory that develops

from intuitive thought and a complimentary self-theory that comes from analysis and reason.

Now, these bimodal systems of cognition are not new. The ancient Greeks had them, for example, philosopher and people such as [Pascale] (indistinct), Simon (indistinct) and a whole bunch of other people have constructed them in similar ways in the last 50 years or so or 100 years or so.

They've generally been taken to be dichotomous rather than a complimentary continuum that recognises the importance of both form and cognition. But if we do consider the two as a continuum they give us a much greater insight into the commonsense nature of human thought. As far as we know humans are the only species that have developed such a sophisticated analytical reasoning capacity and it's made us very successful. It's the combination of these two aspects of human thought upon which our view of rationality is constructed.

Our belief systems are largely the product of intuitive thinking and it takes a great deal of effort for us to undertake the rigorous analytical thinking for us to be truly rational.

Now, ultimately the purpose of all this is to determine whether a judgment or a choice or a decision is likely to be successful and there are two essential elements to this. First, is the judgment coherent with the prevailing paradigm? And second, does it correspond to establish (indistinct) facts? Both are necessary but neither is sufficient. For example, a rationally determined judgment might not be accurate because it's based on a wrong paradigm. And a judgment made through erroneous thinking or is based on a wrong paradigm might be accurate purely by chance.

In other words for a judgment to be ultimately successful it needs to correspond with observed facts and phenomena and it must be coherent with our best understanding of the way the world works.

Now, this all sounds pretty straightforward but cognitive psychologists have found we're prone to major errors both in our intuitive and analytical thinking. And I'll discuss two of the more influential areas of research into this.

An early pioneer in the area was Kenneth Hammond who in the 1950s developed a theory that had previously been developed by Egon Brunswik on perception. And he observed that people respond to various cues that they perceive and interpret in situations. Each individual receives different cues and interprets them slightly differently and this gave rise to what Brunswik referred to as the Lens Model. So just as an optical lens, which presents a different image to observers

depending on where they're placed relative to the lens, in much the same way people form different perceptions of situations because of cues they receive and their interpretation is different.

So to sort of summarise that it's expected that people will reach different conclusions about the nature of the same problem from apparently identical observations. And I think that's a really important point, we all receive different cues. The information looks to be the same but we all make different interpretations because of our different perspectives.

The second stream of research that's become very [influenced] in the last couple of decades and has landed a couple of people Nobel Prizes, two American psychologists Amos Tversky and Daniel Kahneman found that both laymen and experienced practitioners are subject to bias in thinking. They investigated why people make apparently simple mistakes (indistinct). And then they confirmed this in finance, the judicial system, medical diagnosis, choice of treatment, policy formulation (indistinct). And this research led to a very pessimistic view about human judgment, that it's irrational and it's untrustworthy.

But many of the researchers who reached these conclusions overlooked the original Tversky and Kahneman research but said that bias is primarily a (indistinct) intuitive judgment on analytical reasoning. And further research found that often by understanding the source of bias it can be removed. So if we give people training and statistics the probability they come up with better answers. And if we give them (indistinct) an inference it enhances their analytical thinking. Together with rigorous application some biases can be reduced or entirely eliminated.

So very successful people such as the people in this room seem to have the ability to meld the insight and creativity that derives from intuitive thought with a power of analysis to recognise the differences in perception and the bias introduced due to our intuitive thinking. This process of creativity combined with rigorous criticism enables them to develop a deep, rich subjective and objective knowledge base and thereby they form a more comprehensive understanding of the world.

So in this brief review I've argued that there's a remarkable consistency and convergence in the philosophy and psychology around both the nature of truth and the criteria for distinguishing between truth and falsity. Both are important in understanding the way in which humans make complex decisions and try and form rational judgments.

Let me now turn to post-truth and why I think it poses such a threat to free open societies. If you look back over history whenever there's been a major change in

communication technology social disruption and change follows. Sometimes it's for the better but often it's for the worse. The printing press was used to great effect during the Reformation with the distribution of drawings and pamphlets. The first English newspapers were started in London in the 1660s at the time of great social upheaval and at a time that also gave root to many of our modern (indistinct). (Indistinct portion) and daily newspapers became possible. Together with photography, which was also invented about this time, newspapers were a major influence in the American Civil War.

Not long after the invention of motion pictures they were seized upon as a (indistinct) tool and were used to persuade public sentiment during World War 1. Russia and Germany both embraced motion pictures and established Government sponsored film industries. In 1933 Hitler created the Reich Ministry for People's Enlightenment and Propaganda run by Josef Goebbels. It was used to great effect in the Holocaust. Wherever you find a totalitarian regime you'll find a State sponsored ministry of information.

Now, the difference between totalitarian propaganda machines and the free press of open societies is that the free press aims for truth in reporting no matter how imperfectly that might be achieved. This holds the establishment and its institutions to account and thereby helps to build and maintain our trust in them. Totalitarian propaganda units deliberately create distrust by circulating disinformation and misinformation to conceal unpleasant aspects of the truth in order to support the prevailing regime.

Now, the development of the worldwide web in the 1990s with 24 hour availability of news has marginalised these established news media. The emergence of social medial with the extraordinary penetration of Facebook and Twitter has brought about a fundamental change in the way news is delivered to consumers and the political discourse unfolds. News is no longer distributed via universally acceptable media rather algorithms used by Facebook, Twitter and Google deliver news based on your search preferences.

These companies don't uncover news themselves but they parasitically harvest news that's gathered by companies that invest in the human and financial resources needed to report it and then they grab information from other sources as well. This so-called news is not about the dissemination of objective true information, it's about marketing a commodity called content regardless of its truth to an audience segmented down to the individual driven solely by data analytics, marketing strategies and search engines. By their very nature these appeal to and reinforce personal bias and prejudice.

Shrewd commentators from shock jocks to politicians can now exploit this to directly target the individual, play to sentiment, and to shape public opinion. In such an environment truth becomes one of the first causalities as the sheer volume of disinformation and misinformation swamps rationally determined knowledge. This has the same effect as totalitarian propaganda machines. It erodes people's trust in established institutions.

For the evidence of this look no further than the misleading innuendo and deliberate lies that were propagated in the Brexit campaign, the 2016 US Presidential Elections, and virtually every election in Australia in the last decade, and the endless discussion around climate change.

So what can we do about it? The challenge is predominantly one of leadership. Leaders should critically evaluate propositions in the light of fact and reason while at the same time recognising their own fallibility. We should be clear on what we mean by truth. We should insist that the criteria we use to distinguish truth from falsity are clear. We should recognise the shortcomings of human cognition. We should insist on the same rigor from others. We should be vocal in our criticisms when we see truth being compromised. We must not let public policy making enter the domain of unsubstantiated, untrue dogma and belief. We must protect the institutions of our society by holding those who run them to account and supporting them in adversity.

The more we strengthen these institutions the more people will be inclined to place their trust in them rather than be ill-informed and deliberately (indistinct) chatter that they find on the Internet.

Two centuries ago Keats wrote "Truth is beauty." Two weeks ago the leader in the Economist said "Truth is hard work." Both were right. Thank you.

(Indistinct portion)

Prof Chapman: (Indistinct portion). So this a very, very fresh topic to me because (indistinct portion) and more about that later. Is the screen on already? Yes. The world faces an existential threat from climate change and the transition to clean renewable energy is front and centre of global hopes for avoiding some of the worst scenarios.

Today Australia has no peak national body or Commission for climate change yet thanks to the efforts of four cross-bench politicians, I call it the four horsemen of the anti-wind farm apocalypse, whose votes were courted by an appeasing Government, we have a Commissioner for wind farms. The National Health and Medical Research Council has no dedicated program of research focused on

climate change but it does have a dedicated research program on wind farm disease, which as we shall see, is demonstrably a non-disease.

Today I will consider how this happened and what it says about the erosion of truth for the post-factual era. But first I want to give you some historical context because it's very important to understand this is a point that's been made before, that what we today call fake news has long been part of popular culture in the form of factoids. Factoids are items of unreliable information that are reported and repeated so often that they become accepted as fact.

Now, social media, as the last speaker emphasised, has massively facilitated the contagion of factoids. Bogus statements passed around face to face social networks in the pre-digital era moved at glacial pace compared with the speed at which claims circulate today. New technology has a very, very long history of attracting prolonged impassioned and often crackpot attacks from those both fearful of and hostile toward Mephistophelian, not a word to get your head around when you've got a cold, (indistinct) that offends the existing order of things.

Linda Simon's fascinating history of electricity, dark light and anxiety from the telegraph to the X-ray notes that "Although the discovery of electricity generated excitement and electrical companies worked hard to build the market for electrical power." For more than 30 years afterwards Edison's invention the business section of Lower Manhattan barely 10 per cent of American homes were wired. And even after the First World War that had only risen to about 20 per cent.

Now, one reason for this was community anxiety about the safety of electricity was widespread with many news reports being published about the calamities that electricity caused those foolish enough to embrace it. But some also worried very much about going blind from reading by electric light. And in 1888 science noted a new disease called Photoelectric Ophthalmia described as "Due to the continual action of the electric light on the eyes the patient is awakened in the night by severe pain around the eye accompanied by excessive excretion of tears."

Now, on September 24 in the same year the British Medical Journal carried a report on the newly popular telephone, saying that it could cause telephone tinnitus and claiming that victims "Suffered from nervous excitability with buzzing noises in the ear, giddiness and neuralgic pain." And the article contextualised the perils of these new contraptions by saying that "As civilisation advances new diseases are not only discovered but are actually produced by the novel agencies, which are brought to bear on man's body and mind. Almost every addition which science makes to the convenience of the majority seems to bring with it some new form of suffering to the few. Railway travelling, it's a (indistinct) liquid in the shape

of slight but possibly not an unimportant jolting of the nervous centre. The electric light has already created a special fear of ophthalmia and now we have the telephone indicted as a cause of ear troubles, which react on the spirits indirectly (indistinct)."

George Willard (indistinct), a prominent US neurologist promoted what became the common diagnosis of neurasthenia at around about 1869. And his central thesis was that modern living in the case of life among the well to do was causing a proliferation in a range of aggressive symptoms. And among the causes of all this nervousness he included several new-fangled inventions, wireless, telepathy, science, steam power, newspapers (indistinct). In other words, modern civilisation.

Now, I'm old enough to (indistinct) through evidence free public anxieties about television sets, electric blankets, microwave ovens, power lines and computers. My family was the first to have a television set in Ophir Street, Bathurst. All the children in the street would gather around except those from one family who were banned from coming there because of the danger of the rays coming through the screen.

In recent years we've seen apocalyptic predictions made about the mobile phones (indistinct) brain cancer what smoking did to lung cancer. But unfortunately for these forecasters the incidents of brain cancer has flat lined for over 30 years while mobile phone use has been almost (indistinct) for about 15 years.

In 2006 (indistinct) two authors writing in that esteemed journal (indistinct) predicted that by the end of 2017, 50 per cent of the world's entire population would be suffering from electrosensitivity and hoping to beat a retreat to the world's ever (indistinct) electricity free haven. Now, there are only about 30 days to go, folks, so watch the traffic jams to the Blue Mountains.

The most recent panics about the modern health worries include Wifi, smart electricity meters, solar panels on roofs (indistinct portion).

So my new book with Fiona Creighton, which is published yesterday (indistinct). Today I'm going to summarise why it's clear, adverse reactions to the (indistinct). I'll then spend most of my time talking about the opposition to wind farms in Australia and the formal and factoid science that has driven it.

So in our new book we list 247 different diseases and symptoms in humans and animals, which have been attributed by wind farm opponents to wind farms, and particularly to subaudible infrasound. I hit upon doing this one wet Sunday afternoon and I went to Dr Google and started putting in wind farms and then just

random diseases and by the end of the afternoon I had about 60 and today I've got 247. These include lung cancer, skin cancer, haemorrhoids, gaining weight or losing weight. But my favourite of all is disoriented (indistinct). Now, most of the 247 are classic symptoms of anxiety. They're things that can happen to you when you are very worried.

So from at least the time of Francis Bacon in the 15th century scholars have observed that people can worry themselves sick. Bacon wrote that "Infections if you fear them you can call them upon you." The nocebo effect, the evil twin sibling of the healing placebo effect, has been documented in the vast research literature in both clinical and real-world settings. When some people are exposed to frightening information about agents or exposures, expectancy effects are just as powerful as placebo effects can operate to make people sick with worry or anxiety.

However, there have been by last count 25 scientific reviews, which have been published since 2003, which have all concluded that there is very poor evidence for any claim that wind turbines are the direct cause of any disease. For any social scientist such as myself (indistinct) uncontested evidence that point unavoidably to a conclusion that wind turbine syndrome is what I've dubbed a communicated disease, not a communicable disease but a communicated disease. You catch it by hearing about it and then worrying about it.

What we do know from the science is (indistinct) observational studies et cetera there are a small minority of wind farms, which have a small minority of residents who claim to be affected. The direct causation hypothesis would predict that all wind farms should affect some people and that's far from the case. The great majority of complaints occur in English speaking nations despite the proliferation of wind farms in Europe, China and many other non-English speaking nations. So this is a disease which mainly speaks English.

Wind farms with a history of being targeted by opposition groups are more affected by wind turbine syndrome. Just six wind farms in Australia, and there are around about 70, had 74 per cent of all known complaints and they have all been wind farms which have been targeted by the anti groups.

Those with negative views about wind farms are more likely to report symptoms than those with positive views about them. A point that's already been made by Mary, those being paid to host turbines very rarely complain (indistinct) suggesting that the drug money maybe a very powerful preventative.

Claims about (indistinct) susceptible individuals, for example, like those who get motion sickness while others don't struggle to explain why there are apparently no

susceptible people, for example, in the whole of Western Australia to the whole of Tasmania with no records of anyone ever complaining about wind turbines in those two States.

Claims about over 40 families having to effectively abandon their home have never been validated with those who are making such claims saying that many of the wind farm refugees don't want publicity. Well, I mean most people who complainant about things do want publicity, so for some reason we're supposed to believe that the people who complain about wind farms (indistinct portion).

While some complain of acute effects within minutes of exposure the first known complaints about wind farms date from 2002 although many wind farms were operational for many years before that. So why then were there no reported acute effects occurring prior to 2002?

Experimental subjects who were randomised to be exposed were not exposed to negative news footage about wind farms and then exposed to (indistinct portion) show that prior exposure to anxiety producing messages increases the reporting of symptoms even to sham wind infrasound.

So we devote a chapter to exploring the interesting views in our book of several of Australia's most prominent opponents of wind farms. And I'm going to give you a few vignettes of these people now. Sarah Laurie is an unregistered doctor. She told a South Australian court in 2011 that "Wind turbines can make people's lips vibrate from a distance of 10 kilometres away." And I think that's from about here to Chatswood, right. And she's also said that "These vibrations are sufficient to knock them off their feet and bring some men to their knees when out working in the paddock." So the program Mythbusters I think, you know, would be fed this information because they'd do a wonderful sort of challenge (indistinct portion).

Laurie also claims that "Some people are so exquisitely sensitised to certain frequencies that their perception of very, very low frequency is right off the shape of the bullet curve such that they can, for example, from Australia perceive an earthquake in Chile." Now, you can see how far Chile is from the east coast of Australia.

Mr George Papadopoulos who is a pharmacist in the Yass area, there maybe such a person, he has written that he can perceive wind farms from up to 100 kilometres. Now, I think that's as the crow flies from about here to Lithgow, okay. So these are the sorts of people who put in submissions to various inquiries, who write to Government officials, who attend meetings and demand answers about their claims from scientists and so forth.

Mr Papadopoulos at the time worked as an assessor for something called the Geovital Academy. This is an entity which sells blankets, shields, paints and pillows to protect gullible people from the evils of electromagnetic radiation invading their homes. Its website once had an endorsement from I quote "Nobel Prize winner Ivan Engler. Now, no one named Ivan Engler has ever won a Noble Prize for any category. He may well have won a noble prize, whatever that might be.

Mr Noel Dean as an objector from the Waubra area in Victoria, he once told an anti-wind farm meeting in 2013 that "Wind turbines started charging his mobile phone without it being plugged in." Now, this extraordinary claim I think would be of great interest to manufacturers of mobile phones who apparently have not been advised about this remarkable charging ability. And something all of us would be very keen to know about I think.

The next person is someone who's a regular guest on the Alan Jones program and gardener, she's perhaps Australia's most prolific wind farm complainant, and she believes she is adversely affected by wind turbines "Even when they're switched off." So where we go with that claim is anyone's guess.

And finally, Bruce Rapley who in 1995 published the visit to New Zealand of a prominent Australian anti-immunisation advocate told the 2015 Senate Wind Farm Committee when invited to give oral evidence that "In the future I believe the adverse affects of wind turbines will eclipse the asbestos problem in the annals of history. In my opinion the greed and the scientific half truths of the wind industry will be seen by history as one of the worst corporate and Government abuses of democracy in the 21st century."

Now, this is the sort of clap trap that, you know, exists in this debate. I looked up how many people have been affected by asbestos, it's something in the vicinity of 125,000 people of occupational exposures. And, of course, I think it's about a third to two thirds of occupational cancer deaths are caused by exposure to asbestos. So this character believes that wind turbines are in the annals of history going to be worse.

So I think by far the most egregious of all these inquiries, and we've had three Senate inquiries in five years and two State Government inquiries, but by far the most egregious of all these was the 2015 Senate Inquiry. This was headed by Senator John Madigan who was a blacksmith before entering Parliament. The Madigan Committee's report is frankly a travesty of science. It failed to even mention what is universally acknowledged in the scientific field here to be the largest, most robust and important longitudinal study of wind farms and health,

which was run by Health Canada. And that study provided no support for the direct causation hypothesis. It had been published before the report came out, it was in various submissions to the Madigan Inquiry but they didn’t mention it.

The $2.5 million office of the wind farm Commissioner released its first report in 2017. And as anyone following this issue closely would have predicted it was not stampeded by complainants. Wind farm opponents have grasped the straw that the evidence that wind turbines are dangerous is very poor and they argue that we therefore need to invest in research that they just know will eventually prove their point.

There's also I think very poor evidence that you UFOs, the Loch Ness Monster and leprechauns exist. But no serious scientific body thinks that investing research in such claims is a sensible thing to do other than unfortunately the politically pressured NH and MRC, which in 2015 allocated $2.5 million, as I mentioned earlier. And a senior NH and MRC official at the time wrote that "The decision to allocate funding to wind turbine and health research reflected the macropolitical environment."

Now, let me finish by describing the tactics which have been used against my efforts, not just me but a lot of these are against me, to ask awkward questions about the claims being made by wind farm interests. These have included serial complaints to senior officials in my university, that I was belittling wind farm victims by questioning some of these claims. Their claims are apparently beyond question. Taunts that I refuse to ever meet victims and see for myself what they were experiencing. And I can honestly say hand on heart that I've never received such an invitation to go and meet with of those victims but publicly I was taunted and have been taunted repeatedly for refusing to do it.

Complaints to my Institutional Ethics Committee that I was conducting research without ethics approval. Well, that went through full investigation, I was completely cleared, I wasn’t. Constant false claims that I'm in the pay of the wind industry, I'm not. Regular attacks on my academic credentials, you know, I'm the worst possible sort of scientist, a sociologist, can you imagine. I've actually just concluded a defamation case, not in this area but with someone who made the same claim about me saying that I was masquerading as a medical practitioner and (indistinct) holiday in Asia.

Regular attacks, I mentioned that. Attacks under Parliamentary privilege by two politicians. Two defamation suits. Regular slander on an anonymous website. The history of social panics over new technology show that they do have a natural history. There are doubtless few people left today who still fear television sets and

microwave ovens. They hey day of fearing cell phone towers came and went in the 1990s. I know because I wrote papers about it at the time and monitored the media coverage of it.

Wind farm anxiety is now thankfully rapidly receding as we saw from the first report of the Wind Farm Commissioner. The desultory complaint volume submitted to him show that I think the phenomenon is almost past. But the delays that this panic caused in driving Australian renewable energy harvesting were major. Our book's final chapter explores the lessons in how we might avoid the next wave of modern health worries.

I'm going to end with a slide that has nothing to do with what I've just said but I thought I just couldn’t resist showing it to you. It came into my inbox this morning and I put it up on Twitter. This is the former Chief Executive of British American Tobacco and, as you know, what that man's statement of duties is is to try and maximise (indistinct) products right around the world. On the other side of the screen that you probably can't read very well is a (indistinct portion) mobile phone reception because he believed that electromagnetic rays and radiation might pose a major health (indistinct) were exposed to it. Just think about that. Thank you very much.

Look, I think it's important to be sceptical of all the time. I was made Australia's sceptic of the year by the Australian sceptics, so I have a pedigree in scepticism. But I think there is no particular reason to be particularly sceptical about new technology more than being sceptical about anything else. But I do think that it's interesting to historically look back and see some of the commonalities in community anxieties about new technology. I supervised a fascinating PhD about 20 years ago, which was looking at the anti-immunisation lobby. And the roots of that go way, way back to the time just after (indistinct), it's not a new phenomenon at all. But essentially there was, you know, a sense that this was going to happen (indistinct portion) quite unholy and unnatural.

And when the sort of industrial artifice came, you know, in the form of the pharmaceutical industry came in and was involved in it, it just added a whole new dimension to that whole concern.

Prof Chapman: Look, there are, as I said at the end of my remarks, there probably are still a few people who absolutely refuse to have mobile phones. I had a person come and visit me in my office with a worried post-graduate student. She had a meter and she was an opponent of Wifi and she ran her meter over my office and looked at me with great concern and told me that, you know, did I realise what I was being exposed to. And I thought to myself, I sat and listened to her for about an hour

because I was just fascinated in the narrative that came out. But I think these people are few and far between and natural histories of the ebb and flow of particular anxieties I think have been with us for a long time and I'm very pleased to say that I think that the wind turbine one is on the wain very much.

Prof Chapman: Look, in case I gave the impression that people suffering from wind turbine syndrome are somehow making it up or are malingerers or just deliberately playing through a script, I don't want to give that impression at all. The people who are suffering from wind turbine syndrome are suffering from the symptoms of anxiety essentially and those symptoms can be measured, they are not just things that they report. They can be measured on CAT scans, all sorts of things like that. So psychogenic illness is a (indistinct) real experience which distresses people.

And I think one of the most distressing things about some of these phenomenon are the people who are oxygenating that anxiety in communities. (Indistinct) this disease, it's not at (indistinct) proportions by any means at all. If you do, as I've done, go through the submissions and there are sometimes many hundreds of (indistinct), the same names (indistinct). And indeed in the book we look at the phenomenon, well, how many landowners who are getting income from hosting turbines have ever complained and there are only two and I won't (indistinct) both of those if you look carefully at them I don't think so.

MORNING TEA BREAK

Prof Griffiths: Welcome back to the second session. Unfortunately, our speaker in this session, James Wilsdon isn't able to be with us, so I'm going to read his talk and present his slides. I'm just going to speak in the first person, so I here is James Wilsdon, imagine somebody slightly less bearded and with slightly less prominent glasses. Prof Wilsdon is a political scientist at the University of Sheffield and Professor of Research Policy. He's also Director of Impact and Engagement and Associate Director in the Grantham Centre for Sustainable Futures. He's a program leader in the ESRC centre for global higher education and Vice Chair of the International Network for Government Science Advice, which is partly what's talking about in his talk today. He is a regular contributor to the media and one of the editors of The Guardian's political science blog on science and research policy. So his work is exactly - his research is precisely focused on the issues we're discussing today and while we're sorry not to have him here, I'm very glad that he's allowed us to present his material.

I'm speaking as James. In this brief talk, I want to explore what happened in 2016 and why, what does Brexit say about the relationship between evidence expertise and policy, is this the beginning of the end of UK evidence-informed decision

making and what are the prospects for evidence and expertise in post-Brexit Britain?

Over the past 20 years, the UK has built up a strong reputation for the quality of its scientific advisory system, as exemplified by its network of scientific advisers in almost every department of government and by its willingness to experiment and innovate with new approaches to evidence-based policy making. Its early adoption nudge approaches to behaviour and What Works evidence centres being two recent examples.

But this seemingly progressive arc towards the ever-greater uptake of evidence and expertise in decision making took a major knock in June 2016 with the result of the referendum on UK membership of the European Union swinging narrowly 52 to 48 in favour of Brexit. This was despite a mountain of evidence and the near unanimous support of experts of all kinds for remaining in the EU. Here we have examples of long lists of business leaders, economists and scientists all arguing for the UK to remain in the EU.

The referendum process itself was marred by exaggeration and the use of dubious facts and figures on both sides, but particularly by the Leave campaign and by accusations of outside interference in the democratic process by a range of murky and unaccountable actors, including the Russian government. More evidence on the scale of this interference is coming to light on a daily basis, with clear parallels to aspects of the 2016 US parallel election. But were the activities of Russian Twitter trolls enough to swing the outcome? This seems less likely and we also know a lot about the underlying economic and social insecurities, dislocations and inequalities that gave rise to the 52 per cent vote for Brexit.

Concern about mass migration, post financial crisis austerity, more inchoate desires to strengthen UK sovereignty, to take back control, all played their part. As this polling shows, what the vote highlighted more than anything was two very different value sets held by almost equal proportions of the UK public. So what's going to predict voters going for either leave or remain, their background views about the value of multiculturalism, is it a force for good or ill, whether social liberalism has been a force and even their views on feminism are strong predictors of their voting pattern. The older you were, the more likely you were to vote for Brexit. The more educated you were, the more likely you were to vote for remain.

Now another striking feature of the EU referendum campaign was the introduction into British political discourse of an unusually critical stance on the value and legitimacy of evidence and expertise, most notably in the now infamous remark by Government Minister Michael Gove that people have had enough of experts. To

be fair to Michael Gove, the full version of his quote was a bit more nuanced: I think the people of this country have had enough of experts, from organisations with acronyms saying that they know what is best and getting it consistently wrong. Nonetheless, his remarks are seen by many, particularly in academia, as a sign that something has shifted in the British body politic, that this is more than just an ongoing and gradual decline in deference to authority; it's more visceral, more angry.

Other episodes in recent months have highlighted such concerns. For example, this reaction back in January by British tabloids to a ruling by the Supreme Court that Parliament needed to vote before triggering the article 50 clause that initiates the process of leaving the EU. Even in the more hysterical parts of the British press, it's been alarming to see senior judges branded enemies of the people simply for doing their job.

So is this all a sign of the new post-truth politics that we inhabit? That's been the topic of numerous books in recent months by academics, journalists and political commentators. In some ways, things have changed. The combination of vested interests, whether Moscow or Murdoch, the echo chamber effects of social media, the powerful yet unaccountable algorithm all pose significant challenges for the operation of liberal, evidence-informed democracy.

But while post truth was the word of the year in 2016, it's hardly a new problem. Politics has always had a relationship of convenience with empirical reality and science was never pure, as the historian of science, Stephen Shapin reminds us in this recent collection. What Brexit and Trump have jolted is not the status of truth, but the assumption that liberal, rational, cosmopolitan democracies, informed by relevant evidence, will lend a majority to options that appear self-evidently preferable to those who have benefited from that same liberal, rational, cosmopolitan order, EU membership being an obvious example.

But the alternative truths experienced by many in our society, especially in socioeconomic terms, are very different. So while assumptions of a rising tide of evidence-based democracy in the UK have taken a knock, I think this is less a crisis of truth or of expertise and more a crisis of democracy. In seeking to renew the legitimacy of expertise and scientific advice, our starting point should not be to dismiss populist movements or reassert the self-evident superiority of rational decision making. We need instead to start by repairing our democratic institutions and the cultures that support them. Part of this requires greater humility on the part of scientists and experts, acknowledging that we as a community have too often uncritically aligned ourselves with the winners at the expense of the losers, as this

prescient piece by Colin McIlwaine in Nature argued six months before Trump was elected.

Turn back to the Michael Gove quote. For many people, the idea that expert views align with their interests or reflect their own experience is highly debatable. As this piece described, in Newcastle just before the referendum, a Kings College London professor invoked the views of leading economists before inviting the audience to imagine the likely plunge in UK gross domestic product. Back yelled the woman: that's your bloody GDP, not ours. In her brutally simple criticism, she has a point and populist politicians or social media warriors can too easily tap into these anxieties caused by globalisation and rising inequalities and channel them towards resentment.

But we shouldn't despair. In the UK, as in Australia and elsewhere, evidence and expertise are being sought with growing urgency across a proliferating array of policy and public questions. At the same time and often on the same issues, the legitimacy of evidence and expertise has rarely been so fiercely contested, the Brexit referendum being an acute case in point. Paradox coexists with the possibility of new evidence-formed decision making. We need to better understand what lies behind the former and forge alliances to advance the latter. This is why the International Network for Government Science Advice was set up, which I'm sure Peter Gluckman will say more about in his talk this afternoon.

Operating under the auspices of the International Council of Science, ICSU, the INGSA's membership now includes almost 3000 practitioners, academics, knowledge brokers and policy makers. Its focus is on assisting the development of effective advisory systems and the individual skills and institutional capacities that these require, irrespective of particular structural arrangements, through workshops, conferences and a growing catalogue of case studies and other guidance.

In delivering Brexit, decoupling structures for scientific and technical advice can at first glance seem deceptively simple. In many areas, UK institutions map onto EU counterparts, the UK Food Standards Agency coexists with the European Food and Safety Authority. The European Medicines Agency coexists with the UK medicines and healthcare products regulatory agency. Why not shift responsibility from Brussels to London and let us Brits get on with the job? However, as I argued in this Nature piece a few months ago, the difficulty is that UK and EU networks of expertise, guidance and oversight are complementary and have developed in tandem over many years. Generations of British scientists and experts have shaped EU frameworks and vice versa. Around every issue that is codified in law

or regulation, there exists a softer sphere of influence, information exchange and standard setting.

So in animal health, the European Food Safety Authority plays an important role in coordinating data and evidence about emerging livestock diseases. The UK benefits from being part of a network of EU reference laboratories which coordinate surveillance, risk assessment and epidemiology on a range of transboundary diseases, such as avian flu. The Food Standard Authority has drawn heavily on the European Agency's meta-analyses and sophisticated protocols around risk and uncertainty.

In the life sciences, the UK's three per cent share of the global pharmaceutical market is dwarfed by the EU's 25 per cent. This brings significant benefits from regulatory harmonisation through the European Medicines Agency. If EMA licensing was no longer to apply, the association of the British pharmaceutical industry warns of a delay for up to a year in British patients looking to access innovative treatment.

Finally, turning to environmental protection, a recent inquiry by the UK Environmental Audit Committee estimates that up to a third of EU legislation will be difficult to transpose into UK law and those protections for wildlife, for habitats, for biodiversity that can be transferred, will then be detached from the underpinning sources of expert advice, no longer updated, with no UK body to enforce them. Over time, the UK can build up new advisory and regulatory capacity, but this won't be quick or easy.

So as a community, committed to strengthening evidence-informed decision making, we need simultaneously to work on the structural, social and political dimensions of the problem, to rebuild and develop new capacity at the evidence policy interface, but also to address the underlying causes of disaffection with experts. Drawing on the latest work on the evidence of evidence use, of which these two reviews are good examples, this is what we in the INGSA network like to call the science and art of scientific advice.

Providing scientific advice in a reflective way that requires learning from mistakes and is humble in the way it makes its case often requires a shift from scientific advice to knowledge brokering. Brokering requires persistent interaction with decision makers and their context. Brokering necessitates diversity of perspectives, epistemic, institutional and cultural diversity, diversity in disciplines, in methods, in mechanisms, in sectors and institutions, in experiences, ideologies, background, culture and so forth. Brokering means keeping it complex; there is no single privileged view of a complex problem and finally, brokering means providing

multiple alternatives. Given uncertainties and diversity of knowledge and values, there are usually multiple plausible pathways into the future and choosing amongst them is inherently political. There is a strong focus on experimentation and learning in this approach to scientific advice.

So can Brexit be for the UK, or indeed Trump for the UK, not a moment of undoing, of unravelling all that has been achieved, but instead a point of disruption from which we pause, learn and regroup? In a thoughtful new paper, the Science and Technology Studies scholars Sheila Jasanoff and Hilton Simmet make this cautiously positive case, asking whether, quote: the post truth moment can be reframed as a moment of revelation that neither facts nor values can stand alone in a government founded on the principles of truthfulness and inclusive public debate. They suggest that, quote: without renewed attention to the norms that shape the practises of public science and public reason, it would not be possible to guide fortune's wheel expertly along the arc of justice.

On a bad day, of which there are too many right now in British public and political life, the views I've just presented make come across and naïve, as wishful thinking. But much as I lament the result of the EU referendum and wish it could be halted or reversed, I also refuse to believe it is the death of democracy or the beginning of the end of evidence-informed decision making. That story still has many chapters to be written. Thank you.

So resuming my role as Chair, I'd like to suggest that we - I believe we have Peter Gluckman here now. Is Peter in the room or are we still waiting for him to arrive? Oh I'm sorry, I thought he was straight after James or did we reorganise the program? On my version of the program he is still - oh we've swapped, I'm so sorry, I've got the old version of the program. In that case, I'm going to suggest we try and get ahead of time and try to have all question time after Sir Peter Gluckman's talk, since his talk is paired with this one and I think he'll be able to address questions that relate to James Wilsdon material; they're research collaborators.

In that case, reorganising rapidly my notes, I now have the pleasure of introducing Emma Johnston who has recently moved to the position of Dean of Science at the University of New South Wales. She is a leading authority in marine ecology, formerly Pro V-C Research at UNSW as well. She's head of the Applied Marine and Estuary Ecology Laboratory at UNSW, she's led major research projects for industry, government, ARC and the Australian Antarctic Science Program. She was the inaugural Director of the Sydney Harbour Research Program at the Sydney Institute of Marine Science. She's also President-elect of Science and Technology Australia and an important public advocate for science and the

increasing participation of women in scientific research. So it's with great pleasure that I invite her to speak to us and her title is also on the wrong page of my handout, but it will no doubt come up in a minute, so sorry. Why are scientists are so quiet? The cultural and philosophical constraints on the public activities of scientists.

Prof Johnston: It would not be the first time that I had arrived early to a position and in this case in a very esteemed line-up. It's my great pleasure to be here today. I join you today as a practising scientist and a practising science communicator rather than someone who is an expert in the theory of either. So I'd like you to take my comments today as notes from the field, so to speak, or letters from the front, indeed it sometimes feels.

As a young marine scientist, I was fascinated by a very odd type of creature, some of you may be somewhat familiar with barnacles. Anyone know barnacles? Yes? Yeah? General nodding. So if you haven't looked at barnacles very closely, they're upside down prawns surrounded by their own concrete cage that they make themselves, grasping at waves for their entire life. So it was an odd thing to be fascinated by, but I counted literally thousands and thousands of them during my career. Much later I discovered that I was not even the biggest fan of barnacles, in fact Charles Darwin had been probably the world's biggest fan of barnacles ever. I read his meticulous, painstaking study of the world's barnacles - I read of it, I haven't read the full documents yet. It was an effort that consumed eight years of his life and it ended in a bout of ill health.

As an ecologist and a sometimes evolutionary biologist, Darwin's theory of natural selection has influenced everything that I've investigated and how I've interpreted that investigation, so it's part of my lens on the world. But it was Darwin's reason for embarking on his global barnacle study while leaving, get this, leaving his sensational draft On the Origin of Species by means of Natural Selection in draft form, locked in a draw at home, unseen and unread for eight years, that really touched a nerve.

Darwin's obsessive journey from 1846 to 1854 into what Rebecca Stott calls barnacle darkness in her wonderful book, was partly driven by curiosity, so of the more than 1000 species he brought back to London on the Beagle, there was only one in the end that he had not been able to catalogue and describe and that tiny soft dun-coloured creature that he'd found in a conch shell on a Chilean beach turned out to be a rare burrowing barnacle. But it wasn't just this troublesome, scientific loose end that drove Darwin to spend so long finessing his books on barnacles. Darwin actually had an instinct for postponement. He realised that he needed to prove himself as a scientist and as a systematiser, if he was to be

listened to when he did finally publish his most important work On the Origin of Species, or arguably, from my perspective, his most important work.

So he gave his wife detailed instructions on how to handle the publication of the Origin, should he die before his barnacle study was complete and it did nearly kill him. But the first book, Living Cirripedia: A monograph on the sub-class Cirripedia, with figures of all the species, meticulously detailed, won him the Royal Society Medal in 1853. He still had the Balanidae to go, not that you would really care about that, but I did and together with his geological treatise on coral reefs, the barnacles books established Darwin as a scientist who had won his spurs. Stott argues that without his barnacle spurs and without his barnacle contacts, On the Origin of Species by means of Natural Selection would have been very differently received.

When Darwin finally published his theory of natural selection in 1859, quite a few years later, he had a global web of scientific contacts, forged through his barnacle work and a ready-made community of colleagues that were there to recognise the importance of his new theory. He was taken seriously, not as we know by everybody, but by a sufficient number of his peers. Not what was almost 160 years ago, but the story is still relevant today and particularly so in this apparently confounding post truth era. I say confounding because we, as academics, have all played our part in building or reinforcing our global culture of knowledge making. The many different hurdles, gateways we've put in place to weed out unreliable, ill- conceived, biased, incomplete information, designed to ensure that by the time we present our knowledge to the world, it's as close to complete as possible, as reliable an understanding of the world as we can manage.

But as senior academics, most of us are probably confident in the authority with which we publish and in the credibility of our work. But this structure and this culture has also had perverse consequences and they go back to Darwin's story and in some ways my own story. We have built a knowledge hierarchy and a similarly strict professional hierarchy which has not only protected the veracity of what we produce, but has actively discouraged scientists from putting forward bold ideas, or taking part in public debates, particularly young scientists who are, as Darwin noted, yet to win their spurs. The result, I have observed, has been to slow and quieten an entire profession.

Many successive generations of scientists have assumed that the discovery process is mostly about generating research outputs and that their job is to generate new knowledge, not to advocate, not to argue, but to let the facts speak for themselves. As a result, there are very few scientifically trained public intellectuals, very few scientifically trained politicians, because our structures do

not support them or encourage them and scientists rarely see themselves as public intellectuals or advocates. Indeed, many scientists understand very well that to actively seek the public spotlight risks drawing the contempt of their peers. In fact, despite the many passionate arguments and discussions behind the scenes that we do have and some notable exceptions, some visible internationally recognised science voices, the public face of the scientific community is largely hesitant and tight-lipped.

In my own case, I was acutely aware of the scientific hierarchy as a young academic. I felt just as compelled to speak publicly about science then as I do today, but I made a concerted effort to remain quiet, to recognise and behave in accordance with my then junior standing. If I spoke out at all, I spoke strictly within the direct realm of my active research. I did not use my expertise to comment on other matters of the day, even if they were marine matters. In Darwin's time, the quiet that this hierarchy and culture engendered was perhaps not such as an obvious a problem as it is today, it was still a problem, but perhaps not so obvious. The ability to contribute ideas at that time was already limited to those with access to finances, a printing press, a stage, a pulpit, a soapbox and audiences, too, were relatively small and we've heard today about how slowly messages moved.

Even in my own early career, and I won't give away too much of my age, but during the early rise of the internet, there was no broadcast media available to anyone with an opinion, access to a keyboard, the internet. We still had many reliable mass media gateways through which pre-vetted information flowed and many publications had specialist science writers who we could trust to do our communication for us and who also investigated us, investigated the investigators. Now, as scientists, we find ourselves scrambling to find a foothold in an environment in which everyone has a voice, in which the truth can be virtually impossible to distinguish from fake news and everything else in between. As the Yale science communication theorist, Dan Kahan, recently wrote, the problem is not the much-maligned lack of scientific literacy in many of our societies, although scientific literacy is highly desirable, it's not essential for the public to recognise what it is that science knows. The real difficulty for audiences, Kahan argues, is identifying who knows what about what and distinguishing the currency of genuine scientific understanding from the multiplicity of counterfeit alternatives. Everybody appears to be peddling facts these days.

But what does it mean for scientists if the cream doesn't necessarily rise to the top in an information free-for-all, as we had optimistically postulated in the early days of the internet? Well, personally I think we need to recognise that we are at a turning point. We may, in the future, look back at the dynamic changes we're witnessing as the catalyst for remaking of much of what we long understood as the

rules of science. First, we need to turn our attention to and seek to understand the profound impact of the new information technologies on who we communicate science, but that's just the most obvious issue. I'd like to explain why I believe this must also challenge us to rethink what we teach in science education and ultimately how we do science, that is, how we create knowledge, which is our ultimate goal.

Before I go on, I want to pause and to say I'm not a pessimist of any type; I wouldn't have got this far if I was, especially counting all those barnacles. Change always throws up new opportunities, but we do need to be able to recognise and grasp them.

So getting back to today's topic, what is the future of rationality in a post truth world? Well, fake news, propaganda, bare-faced lies are, of course, not new. More than three centuries ago, Jonathan Swift famously noted that falsehood flies and truth comes limping after it. The British novelist, best known for Gulliver's Travels, was also an astute political commentator and published various pamphlets expressing his concerns about what we might today recognise as post truth facts. But in Swift's time, the distances and the speeds at which falsehoods could travel were very limited and so too was the size of the audience they could reach.

I have a more recent example. I was in East Timor in 1999 in the lead up to the independence referendum, supporting the students who were campaigning for a free East Timor at the time. I was assisting the students when reports of a militia attack came in and student fatality numbers varied from zero to 25, so we were rightly panicked about this. My East Timorese friend, who was used to having no internet, no reliable media, no news, nothing that he could trust, explained the knowledge vetting process and the knowledge arrival process. He said to me not until you have heard this same report from three independent sources can you confirm its truth. So he had a system, a well-known system that all the East Timorese were using of defining truth. This process took days before we knew how many students had been killed. Fortunately, it was two, not 25.

Today, falsehoods do more than fly. They seem to arrive fully formed in our consciousness via our screens. The internet has dramatically accelerated and amplified the sensational, the unreliable and the blatantly untrue. We all know that, but there is something else we need to consider about the design of or the shape of the virtual world. Before instant digital communication, in many countries we did have those gateways. We chose news and views via publishers that we trusted and we had that information vetted first. In researching this speech, I came across the multiple websites and Facebook pages for the Flat Earth Society. Have you seen them? You should have a look after. They claim to be places for free thinkers

and the intellectual exchange of ideas and their latest crowd funding campaign is raising funds to launch a satellite to prove that us round-Earthers have been conning them all along.

The websites look professional enough and the satellite plan has all the hallmarks of a scientific investigation. So maybe you think that's a ridiculous example, but one that goes to Kahan's concerns about the challenge of recognising credible scientific information. This is especially so when the tools of science, in this case a satellite, confuse the issue or as he puts it, pollute the scientific communication environment. So two things troubled me: first, the information looks so genuine, but I have to honestly admit I couldn't work out if the sites were a joke or not. But secondly and most importantly, in the virtual world, the glossy claims of the Flat Earthers or anyone else without knowledge or authority, are only one click away from CSIRO or NASA, my own university or any of the academies. We know this flat virtual space is fuelling some troubled communication practises like false balance. When one side of an argument is just as accessible, vocal or visible as the counter view, then we're at risk of assuming an equivalence, that the two sides are actually a balanced debate.

For scientists, the obvious example recently is the way in which this faux duality has bogged down the climate change debate in Australia and beyond. We see, for example, Professor Brian Cox, physicist, seated alongside the former One Nation Senator and vocal climate change denier, Malcolm Roberts, on ABC TV in the name of balance and within minutes, a lifetime of study and research becomes equivalent to an ill-informed opinion.

We also know that any opinion, bias or prejudice can find validation somewhere on the internet and that automated content selection that we've heard a little bit about today, those algorithms, reinforce particular views. It's very difficult to counter selective exposure, selective perception and selective retention. Others have talked in detail and will talk later in the day with considerable insight and knowledge on these topics. In the domain of science research, we know that genuine science news does move very, very quickly online, does spread quickly, but it's the rumours that have greater staying power for much longer on platforms like Facebook.

For climate change mitigation, which is a topic very close to my heart, this means we find ourselves stuck in a repetitive and redundant debate, when we should be channelling our energies towards solutions. This is not, of course, a circumstance caused by the new information technologies, they're the tools, but there are many with vested interests who exploit them and there seem to be just as many with preconceived ideas of how the world works and conspiracy theorists who just want

to use them. As scientists, trained to be quiet, we find ourselves on the margins rarely being heard above the din.

I had a telling exchange at the local beach recently where I'm a surf lifesaver. I'd recently hosted an episode of Catalyst on the ABC which asked a very, very serious question about whether we should intervene and seek to promote adaptation in the laboratory so that we can see the Great Barrier Reef with corals or their symbionts, more resistant to the negative impacts of rising water temperatures. My surf lifesaving colleague mentioned he'd seen the show and thought it was fantastic. He loves the Reef and he's really worried for it. He seemed to be patting me on the back and then he immediately went on, without pausing or even adding a but, and in a rather conspiratorial voice said something along the lines of this climate change thing, you know, I don't think it's happening, do you? Obviously, I was absolutely astonished that even when I could get past the exposure problem, the selective exposure problem, I could not get past the selective perception, the selective retention of information with this character, so I'm not so optimistic about that component.

At the same time, another factor has come into play. As a public advocate for evidence-based action to offset, mitigate and ultimately reverse climate change, I'm regularly on the receiving end of various trolls' extraordinary views. Trolls use the kind of insults that we would not consider hurling in person, but with the anonymity of the online space, their inhibitions seem to melt away. So, to our longstanding cultural constraints that discourage advocacy and agitation, I would add the undeniable pressure from trolls. So what, as a scientist, do we do? First and foremost, we have to get in the information communication game. We cannot deride social media, we must join social media. Again, that might seem obvious, but how we do it is a bit more complicated.

If we scrutinise the way our knowledge system evolved over the centuries, it wasn't a bad model for the circumstances of the past. Discouraging researchers from speaking out until the knowledge they were generating had been vetted and verified and they themselves had built a generous and considerable cache of context, was a powerful way to build our credibility. Our quiet culture did help strengthen the knowledge system. Now, however, everything has been turned on its head and the silence and hesitancy of scientists is putting our knowledge system at risk. The question becomes how do we raise our voices while retaining the rigour and the reliability of knowledge creation.

I don't want to depress anyone, but I'm sure many of you are familiar with the emerging interest in citation analysis. A decade or so ago, a library and information science researcher from Indiana University put many academic noses out of joint when he revealed that 90 per cent of the journal papers published are never cited by anyone and that half are never read except by their authors, referees and journal editors. Publishing in Physics World, Lokman Meho called this a sobering fact and approximately at the moment one article a minute is added to PUBMED. So are it's 26 million-or-so papers to date a knowledge triumph or a tragedy? While we've become very good at adding to the global knowledge vault, we're not very good at getting that high-quality information out and in the face of today's sometimes savage and frequently ill-informed attacked on science and scientists and our findings, I think that speaking out well beyond our conventional outlets can only strengthen our position.

Yet, we are so accustomed to building our careers on the back of peer-to-peer communications that we may not regard talking to the public as part of our remit or talking with the public, I should say. We need to build this into our promotion and rewards process and our training of young scientists because we really do need to be able to explain complex concepts for diverse audiences and we do need to engage much more with other academic fields so we begin to understand things like the power of message framing, that even the font you choose influences how people view the information you are presenting. There's no shortage of empirical research that points to the best ways to convince an audience that your information is valuable and genuine and this goes back to the way scientists can help make it easier for the public to distinguish between credible information and the Flat Earth Society. We may need to make a concerted effort to brand ourselves and credible, engaging, interesting sources. To do so, we must be able to explain what we do, why we do it and why it matters to anyone. That is the so what of our work.

But branding and comms won't win this battle alone. What then would it actually take to turn the quiet culture around? I feel incredibly fortunate to have chosen to major in both the philosophy and sociology of science, alongside ecology when I was at university and what I learnt then informs everything I do today. However most scientists, most of my colleagues, our scientific elite and most science students, our scientific future, have not studied any aspect of our scientific culture or how systems of knowledge making have been built; no philosophy, no sociology, no anthropology. Scientists are mostly unaware of all of the hard work that has been put in by successive generations of philosophers and sociologists to situate scientific knowledge within our cultural and social mesh. Most scientists, in

fact, would deny that science has a political element or that observations can be biased.

When we begin to look deeply at how knowledge has been constructed, we can no longer think of science as pure and I think we saw a book mentioned just recently, never pure. When we understand that, the cultural pressure not to engage in public debate begins to appear deliberate and duplicitous. It also invites us to consider other ways of knowledge making which personally, I believe, can only make us better scientists. This would give us room to rebalance the biases in Western scientific culture that have, for example, largely excluded women and non-Western forms of knowledge. I believe that social, philosophical, historical study of science and knowledge production should be an integral part and an integrated part of the science curriculum and being a new Dean of Science, I may be able to do something about that, at least at my own university. This will help us evolve our practises, but it goes to my ultimate point. If we begin to think about how we make knowledge, not just how we communicate knowledge in this post truth era, this throws up a fundamental challenge, that is, to examine the way we do science, indeed the way we do all research and it's a rare and a marvellous opportunity to have all the academies together to discuss new ways forward.

The process of research has long tended to prioritise and reward isolated development. It's fundamental science that wins Nobel prizes and we understand the importance of this research because history has taught us that from fundamental knowledge, much else, much of it unanticipated and unimagined, flows. That's certainly true, but as scientists and as Mary O'Kane mentioned at the beginning of the day, we are also solving complex contemporary problems and to do this effectively, we know we need to work across academic disciplines and we need to collaborate with a whole range of professions and industries and decision makers and communities who have an intimate understanding of and a stake in solving the many multifaceted problems we're seeking to address. This provides us with an opportunity to think about creating knowledge differently. Research practises are evolving. Collaboration and interdisciplinary research is about the co- creation of knowledge. If you co-define and co-create research, you involve your partners, those with a stake in the problem you are solving in the entire process, from whoa to go, in the process of discovery.

You may find that your work is taken up even before it is published and so your research may have an impact before that first precious paper comes out and this real-world impact plays an important part in the public debate. It is visible, tangible evidence of the value of an evidence-based approach. Moreover, the very relationships necessary for collaboration create valuable new pathways along which credible information and ideas automatically flow. When we involve our

partners in the scientific process, they learn the strengths of our method and the rigour of our approach, they develop respect for this form of knowledge creation and can explain the process to their friends and they improve the process of that knowledge creation. So that's one part of the answer. But what about taking it another step back and asking ourselves to think more deeply about how we identify the gaps in the knowledge and the problems we would like to solve. Likewise, scientists tend to look at them in isolation when I believe we all and all of our disciplines in fact, could and should be working much more closely together.

I see this all the time in my own field. We ask a contained question, we attend to what is up close, we produce the new knowledge and then we wait for it to be taken up. We may be identifying important problems, but without a plan for finding a workable, economically or socially acceptable solution. Take marine pollution, my area of expertise and a very quick case study for you. The work's not yet published but we've done the first sampling in identification of microplastics in Sydney Harbour, sediments and fish. Most of the debris, the plastic debris that we've found is from clothing fibres; it's not, as we might have thought, from rubbish like plastic bags or bottles or from packaging, it's synthetic fabrics, especially acrylic, nylon and polyester. They shed lots of fibres when they're washed and these are very significant but a largely unrecognised source of marine plastic pollution. We don't yet fully understand the impact of these fibres on the ecosystem.

So what might we do as scientists? I could publicise the work and that's important, but do I try to get Australia to ban the use of polyester clothing? Goodbye weekend fleeces, goodbye onesies; could be quite good actually. It wouldn’t be a very popular policy. We know synthetic fabrics have revolutionised life for people worldwide by providing relatively cheap clothing and bedding, often with very favourable insulation qualities and we also know that alternatives like cotton have their own environmental impacts. It takes 3800 litres of water to produce one kilo of cotton, for example and growing non-organic cotton currently uses a quarter of all the pesticides used worldwide. So what if we could take those results and immediately set up collaborations with communities in industries, instead of sitting back and waiting for them to find their way through the academic publishing system. We could collaborate with material scientists who understand synthetic fibres and exactly how and why they break down. We could collaborate with engineers who understand washing machines and chemical engineers and nano materials experts who understand water filters. Communities might want to join us as citizen scientists and could try out different clothing materials and brands in their own washing machines as part of a collaborative, multidisciplinary investigation.

Over the past decade or so, successive federal governments have recognised the value of co-creating knowledge, but it's mostly in terms of collaboration between academia and industry as a means of driving innovation and, in turn, of securing Australia's future economic prosperity. Personally, I think the issue of collaboration is about more than facilitating industrial translation. It's about re-imagining everything we do in science within a social, cultural and economic context. As part of the big picture, it's about doing science very differently, facilitating engaged science, funding more diverse partnerships, doing research together. For us pre- interneters, this might seem like a huge challenge in terms of our academic culture, our skills, our practises, but over the many years I've been teaching, I've seen waves of change moving through the system.

Today's students and early career researchers are digital natives and they are more open in the way they do science already. This means they are expressing online, in real time, their enthusiasm for something that they are discovering, in much the same way as they might report on a social event. They are tweeting from the lab and suddenly their friends and family and followers are commenting and contributing, they're engaged with the very practise of science. Next-gen scientists are crowd funding their research, they are running citizen science projects, their professional organisations are engaging with communities. Next-gen researchers do, of course, understand the importance of verification, but they don't feel the weight of the cultural muzzle in the same way as Darwin or even myself; generational change is already underway.

Interestingly, the World Economic Forum identified the spread of misinformation online as a major risk in its global risks report as early as 2013 and it has since responded with a series of conferences and workshops about science communication, canvassing how we might counter fake news. But recently the WEF Young Scientists, a select group of the world's most promising scientists under 40, have been drafting the Forum's universal code of ethics for researchers. The very first responsibility on the list for researchers and organisations they represent is to engage with the public. This, in my mind, represents a very significant cultural change. The code goes on to exhort scientists to pursue the truth, maximise benefit and minimise harm, engage with decision makers, support diversity, be mentors and be accountable. Its message is that we must talk and engage, agitate and argue.

I'm confident the three matters of which I've spoken represent a positive way forward for science in the post truth era, lifting our voice, critically analysing our history, culture and practice, evolving our knowledge production to engage communities in the entire practice. In the past, we've mostly converged on the best evidence for, say, the value of adding fluoride to water. But we're now operating in

a polluted science communication environment, with lots of toxic messages muddying the waters. Researchers tell us that people acquire their scientific knowledge by consulting others who they identify with, who share their values in whom they therefore trust and understand. That, in my view, is a good reason for us to take stock, to take steps to address the limitations of our own culture and begin to dismantle our silos and to build diverse partnerships, all of which can make us part of those trusted conversations. Or at the very least, my lifelong interest in barnacles suggests a place we definitely don't want to find ourselves, stuck to the same old science rock, increasingly irrelevant and drowning in a sea of noisy change. Thank you very much.

Prof Griffiths: Well after that highly original, fascinating talk, I suspect we'll have a storm of questions.

Prof Johnston: About barnacles.

Prof Griffiths: Of course.

Question: Well thank you very much, Emma, it was very inspiring.

Prof Johnston: Thank you.

Question: One of the things that's worried me for many years is the failure of the scientists and non-scientists taking account explicitly of social responsibility and I was greatly influenced in my whole approach to change by two important pieces of work, one by Chomsky called The Responsibility of Intellectuals where, in effect, he said, it's not enough to make napalm, you've got to take account of what use is made of it; and the great book by Australia's most progressive social science and historian thinker, the late Hugh Stretton, called The Political Sciences, in which he argued that to believe that analysis and ideology could be separable was to corrupt your students and to have a false philosophy and he went through all the social sciences showing examples of this. But that had a tremendous effect on my teaching and my work. I never started a course of lectures after that without telling people what my philosophical, religious and political values were and what approach I took to economic analysis. I'd say I don't expect you to agree with me, but I wanted you to know where I come from.

Prof Johnston: I think that's a great example of how just giving people the basic understanding of how social science works and how philosophy of science is developed, can help them then teach, interpret but also do the science, that they do more rigorous research that they bring. It's very, very odd to me that we can create a knowledge creation system which is designed to remove bias, but then pretend that we are not biased ourselves. I mean the very structures that we use in science are very,

very good at removing much of the bias, but unless we recognise that we can hold bias and we do hold biases, then we can tend to let those processes fall by the wayside and also not recognise the biases that we haven't been designed to remove in, for example, the definition of your problem.

Question: Wonderful talk Emma, I really, really appreciate it, I give one like that myself, so we'll swap.

Prof Johnston: Yeah, we'll swap. Mine's new.

Question: That value system that you talked about, about scientists and people in the academy feeling that it's somehow unseemly to be engaging with the public, it is not how people who are most revered within the academy see their own role. I had an (indistinct 2:50:33) a few years ago with people like Wayne Hall and Sally Redman and so forth. We looked at six areas of public health and we got people around the country who were active researchers in each of those six areas to peer vote the six most influential people in those areas in Australia. With the exception of one individual, every one of the people we then interviewed said that they believed that they have a duty and a responsibility to carry the implications of their research into the public and political domain, to try and make sure that people who are in a position to act on it and who should be aware of what they have been finding, did that. So I think that the people who today denigrate people like you and I as being self-promoters or show ponies or whatever, are often people who are not particularly influential themselves. I think that institutions have an enormous responsibility to try and reward academics who do that. We've all been in those interview situations, you know, promotion interviews where research teaching, service to the profession, service to the community and the full criteria, the service criteria, in terms of the community one, it's often regarded as fluff and it shouldn't be.

Prof Johnston: So Simon made a very good point that people who have the most influence and effect over public debate but also over policy in government are usually the ones who are using their scientific voice or their research and intellectual voice in public and it's generally recognised by many researchers that they have a responsibility to communicate their results. But the point that you made at the end was really - I've already just lost my-----

Question: The point was that academic promotion ought to be taken far more seriously.

Prof Johnston: That's right. So academic promotion, whilst it's already - always been the case that people need to make an argument as to what they're contributing to the profession and what they're contributing to their communities, that aspect has often been inadvertently down-weighted where research and education activities directly have

been given a higher priority. What I was going to say is one of the most effective ways I think to rebalance that is actually use a scientific technique and actually weight all of those equally and actually score them so that, for example, at the University of New South Wales, you have a score out of three for research, a score out of three for education and a score out of three for your community engagement, global impact and leadership. You have to get a total score of six, so you can have research-focused people and you can have education-focused people, but you can also have people who are completely focused on engaging and using their evidence to inform policy and they can do extremely well.

Question: Thank you Emma and I'm pleased you referred to Charles Darwin. What I observed, that in politics one of the common attitudes to appropriate these statements of important writers like consider what's being done with Henry Lawson and Adam Smith, the same with Darwin. What Darwin was about really was the importance of adaptation, not the importance of competition. But economists, a lot of economists say, with respect to one in particular here, have adapted, have appropriated the competition approach into economics and life as if that has been justified by Darwin's studies and is somehow inherent. Do you think it is important that biologists actually speak out more, point out that role?

Prof Johnston: That's a really good question and people will use scientific information selectively, won't they and for their own purposes. In fact, they will often do give really great examples of Darwinian behaviour in terms of trying to promote themselves. I think the thing about any scientific philosophy and those that become quite familiar to people is that they can be cherry picked and information in general can be cherry picked and I think that's been one of the hardest things for scientists to effectively engage with people who are pretending to be scientific. So using scientific information, but picking the eyes out of it, they can in fact make very convincing arguments to people who aren't scientifically trained. That's one of the biggest issues I have in the climate debate, is that often I might be making a very coherent argument based on hundreds and hundreds of papers, but I can be effectively dismissed by somebody who has just cherry picked three or four graphs and interpreted some aspects of it in a highly biased manner. It's really hard for communities to pick between my argument and that other person's argument if we're using the same language and we're using the same graphics and individuals.

I haven't quite worked out how to deal with it, which is why I've suggested that we actually need to engage people in the act of doing science and research, in the act of collecting data, because the most effective means of owning and understanding the rigour and the strength of the scientific method is by being actively engaged in that process.

Question: Prof Emma, thank you for that. I was very interested by the idea that we might want to teach our young scientists more about serious philosophy, anthropology, sociology and I certainly benefited from it in my undergraduate education. Where do you see, do you see there's much scope for that in Australian universities?

Prof Johnston: I think there's an enormous amount of scope for it. There are two ways we can do it. We can make subject compulsory and that's the one that's going to have the most resistance because there's a feeling in universities that we're already trying to cram too much content into people. So that's one component and I think we would have a cohort of undergraduates in every degree who would like to do one or two or maybe an entire major in philosophy and sociology of science and we should provide that. I think part of the problem is we're not even making that available to those who choose it at the moment. It's only a few universities who can provide those subjects.

The alternative method, which I think is for the masses, will be to integrate history and sociology into the way we teach our content matter and that is going to be far more effective, but it requires that all of our university experts in their various areas are also enabled and understand the history and philosophy of their own subject matter. So then the resistance is going to be from my academics, I don't have to learn that. So we do have problems; we have one of access and we have one of fitting more content into complex degrees. I'm not a big fan of the American model of degree where you go in and you do a general arts degree, I know some people are - sorry, a general degree, including arts et cetera for science. From my perspective, high school is where we do our general education and at the university we do higher education and that doesn't mean that you're restricted to only science subjects, because you can study the history of science, you need to study around your specialisation, but not necessarily do multiple components of that education within the degree.

The reason I do that is because degrees are long enough. We have a three-year undergraduate degree, an honours year and a PhD. If you do all of those in a row, you're at university to become an entry level researcher after eight or nine years. If we extend that much more, what we do is we actually make it less accessible to huge sections of the population but structurally we make it particularly less accessible to women because we're basically making them study and not earn any money up until the point at which they're right in the middle of the time in which they would want to have children. Then we create a situation where we don't give them a permanent job for the entire period in which they would be wanting to have kids and that's structurally where, in the university system, we are losing all our women, is between the PhD and that continuing position. I think to extend the degree any longer makes that situation worse.

Prof Griffiths: I'm afraid we're literally running up against time for the next talk, so we'll just get behind, so I'm actually going to have to cut off discussion.

Prof Johnston: Thank you very much and if you want to follow me on Twitter, I'm @DrEmmaLJohnston, because I know you're all on it, right?

Prof Griffiths: I'm sorry, the trouble with these - some of our previous forum, the talks are too interesting and we don't have enough time for all the questions. After Emma's talk, one of the ideas here is that the structures through which we do and make science have evolved over the 300 years of the history of modern science, so we should remember that academic publishing is in fact a model that had to be decided on in the late 17th century. Some people thought collective anonymous publication was the way to go and as you can see, that didn't catch on, I guess we can sort of see why.

One recent development I think we can all see has been a really progressive development in the way that science has been structured, is the creation of the body of structured access to that enormous vault of knowledge that Emma Johnston referred to, which is known as the Cochrane Collaboration. Our next speaker is somebody who has been very heavily involved in that process for a long time, so I'm very pleased to be able to invite Lisa Bero, my colleague from the Charles Perkins Centre, to come up. Lisa directs our evidence, policy and influence collaborative research program, her work concerns biases in the design, conduct and dissemination of science and has the aim of reducing and improving the use of research for policy and practice.

Prof Bero: Thank you Paul and I want to thank Brynn for the invitation to speak here today before all of you, it's been such a stimulating day so far, so I hope I can continue that. When I sent Brynn my draft title, he said, it's okay, the subtitle's okay, whatever it means. So I think I'll start there, with putting the cart before the horse, we'll talk about what that means a bit. So that's an expression, actually, for doing things in the wrong order and as we've heard today and all throughout the morning, there's been a great deal of investigation and concern about the role of evidence and policy making and there are also many interventions going on to improve the uptake of evidence in policy. Many of these interventions actually deal with reducing the impact of conflicts of interest on policy making and making evidence more important in the process. I'm really interested in conflicts of interest because I think they're something we can actually do something about them.

But I want to refocus now a bit and I think we've been putting the cart before the horse, because we really should be paying more attention to what influences the evidence itself and I think this is a great follow on to Emma's talk because I'm

going to dig in to some of these specific biases we've been talking about. We know that conflicts of interest can bias the design, methods, conduct interpretation and publication of evidence and I'll give you some examples of that from my area, which is biomedical public health research. These bias findings have led decision makers to underestimate harms and overestimate effectiveness of interventions. The research community has responded to this by increasing transparency about our research enterprise, but there is much more we can do. We can strive to reduce the influence of these conflicts of interest on research so we can have trustworthy evidence and we need to do that through structural reforms, which I'll talk about near the end of the talk.

First of all, what is a conflict of interest? We all need to have a common understanding of this term. This is perhaps the most commonly used definition of conflict of interest. It's a circumstance that creates a risk that professional judgment or actions regarding a primary interest will be unduly influenced by a secondary interest. For example, a primary interest would be doing unbiased research. A secondary interest might be personal financial gain related to the outcome of that research. As we've heard this morning and I completely agree, there are many, many interests that scientists bring to their work. Science is not value free, we all have our own beliefs and experiences that we bring to our work that shapes it, but these are different from conflicts of interest. Conflicts of interest are a risk, they don't necessarily mean that judgments will be biased and conflicts of interest are always real, they're not potential, so someone can have a conflict of interest, I have a conflict of interest, but they don't always result in biased evidence, so what might not be real is the bias. That's what we have to sort out when these conflicts result in bias.

Now in politics, conflicts of interest are of course very well understood and I don't think I could say those three words without mentioning Donald Trump. There's been much, much written about his lack of disclosure of his income taxes and how changes in tax reform might personally benefit him, nepotism in his office, how government policies in the US will affect his stock holdings, his leasing of government property, his foreign holdings, on and on. But just to reference Simon's talk this morning, I thought this was an interesting conflict of interest, so Mr Trump owns two golf courses in Scotland and he's recently been lobbying politicians in the UK to oppose windfarms. This is not because he believes they're bad for health or for the life or birds, there are no echidnas in Scotland that I know of and not because this might contradict US or UK energy goals, but because they would lower the value of his golf course property. I think everyone gets that, that's a conflict of interest and we talk about it a lot in politics.

Now in science and health, we are still having trouble sometimes recognising conflicts of interest and there's frankly resistance to this. These are disclosures from actual journal articles. For those of you who read journal articles, I'm sure you've noticed we're doing a lot more about disclosure. That one on your upper left is hard to understand because it talks about actual competing interests and then potential interests that are on file, so what's really the difference? The one on the right is interesting because it said that the researchers were given an opportunity to disclose, but you don't really know if they did or not. Then the one at the bottom's interesting because in the acknowledgement section, it says that the study was funded by a variety of government funders, but then there is this little term duality of interest that says it was also funded by a drug company which happened to make the product that was being tested in that paper.

These are not really easy to understand sometimes, these conflict of interest statements and what's even worse, is that we can have very, very - and don't even attempt to read that one on the left, but that is just an example of a conflict of interest statement you typically find in a medical journal article nowadays. You're just swamped with too much information, nobody ever reads these. Then we also get statements that are cluttered up with - statements that you're just wondering how it's relevant to the article. The one on the right and I apologise it's fuzzy, but I wanted you all to know that I am not making this up, this is from a real article and basically this was an article that was examining the association of HIV transmission and male circumcision, so it was an important public health question. The authors not only disclosed their Jewish faith, but they also disclosed whether they and their children had circumcision and how the circumcision was done and what really startled me was that one of them is married to a non-practising Catholic, which doesn't seem to have much to do with anything. This is the kind of thing that is cluttering up the literature and it's also making researchers very confused about and not taking seriously, actually, conflicts of interest. The bottom line is disclosure is difficult to enforce and it actually does not prevent bias in research, as I will show.

Overall researchers, at least in my area, are still in a great deal of denial about conflicts of interest. These are quotes from interviews, studies we've done with researchers in various countries, in various fields. They'll say, I'm not influenced, my colleagues are influenced, but I'm not. That's a very common one. I have ties with all the companies, so I am not influenced by any. I'm just helping out my patients. Then the last one, I recognise that I'm in conflict but believe that I can handle it, if I couldn't handle it, the conflict, I wouldn't have gotten involved. I think this is a really telling statement, because this investigator does not recognise that

preventing bias is not about personal responsibility; we really need institutional changes to reduce bias.

I now want to talk about what I mean by bias. We've heard this mentioned over and over this morning and there are very specific definitions. We do quantitative studies in bias. This is a meta-analysis and it shows each horizontal line there is one study that has examined the association of pharmaceutical industry funding with the outcomes of a study on a drug. You can see there are a lot of lines there, so a lot of people have studied this in a whole variety of different areas and it's not just drugs. Basically, what this slide shows, when you put them all together in the meta-analysis, is that the industry sponsored studies are about 30 per cent more likely to find that a drug is effective compared to studies with other sponsors and this is even your controlling for the methods of the study. So that's a really important point. What we would like to see would be that larger effect estimates are associated with the intervention itself, not with the sponsor of the study or not with some methodological characteristic of the study. That's what's wrong with these studies, is that the effect is associated with the sponsorship.

What's going on? That's an observed association, how can this possibly be happening? There are many ways to bias a study and all of you are researchers and you're familiar with this. The first way is in how you ask your question. Many corporate scientists I've worked with, they'll say, well you know, sometimes we just don't ask a question if we don't want the answer, you can frame it a certain way, you know, it's a very honest answer. As we've heard, that our own personal interests will drive our questions, so that's one way to bias a study. The next is the methods we actually use in the study, so the protocol by which it's conducted. The next is what goes on behind the scenes, so even if a study has a very rigorous protocol, is it actually conducted according to the protocol? Then the third way is science doesn't really exist, at least right now, until it gets published in some way and so the last way to influence the study is by whether it's published in full or not. I'll just give some examples of these different types of biases.

We're doing some work now where we're looking at research funded by Coca-Cola and we're following up the studies that they’ve funded. We're looking at the topics of those studies and interestingly, of the 200-some studies we've looked at, about half of them focus on physical activity and actually a much smaller proportion of them focus on the effects of sugar on health. This is a way that a funder can actually distract from questions about harm of their product and focus on something else, so let's focus on physical activity, not the effects of sugar. This is incredibly similar to research I did about 15 years ago looking at tobacco industry sponsored research on second-hand smoke. An organisation called the Center for Indoor Air was fully funded by the tobacco industry and they published primarily

studies on the effects of carpet off-gassing, whether having green leafy plants in your office was good for you, basically they studied anything in indoor air, other than second hand smoke, that might have an influence on your health. Then they used this over and over and over again to distract from the harms of second-hand smoke. So how the research question is framed is important.

Now the methods of studies can introduce bias. For example, we know in drug studies, that randomisation is - poor quality randomisation will overestimate the effect of a drug. Again, I'm just using drugs as an example, so that's a bias, poor randomisation overestimates the effect. If the randomisation is not done correctly, if it's done like an iPod shuffle, for example, which I thought was randomised and I actually wrote Apple about it and they said, no, no, it's a shuffle, that's why you're detecting a pattern, then that's not truly random and that will introduce increased defect estimates in that study. I work a lot with animal researchers and we also have evidence that blinding or a lack of blinding will overestimate efficacy estimates. Animal researchers say, well you can't really blind the animals but of course that's not the point, we want to blind the outcome assessors. The people who are making the measures need to be blinded and that protects against performance bias or measurement bias, so these protect against different biases.

Now what's interesting is industry sponsored studies and those with conflicted authors, tend to use very similar methods. So bias in those studies is more likely due to the questions or to the next source of bias I'll talk about, which is selective reporting bias. Now selective reporting bias, I think of it as coming in a few flavours. First is there may be selective analysis reporting. If there's a study and the outcome is depression, depression may be measured on different scales and those might be reported as medians or means or trend over time, so depending on how those outcomes are analysed, only some of those analyses may be reported and those may be the ones that look more effective. Another layer is selective outcome reporting, so take a study with depression for an outcome and this is a real example, the depression score was measured at three months, five months, nine months, 15 months and maybe all of those points aren't reported in the publication, maybe only one is and that's the one that showed an effect on the depression scale. That's selective reporting on outcomes. Then we have publication bias, which is when an entire study is not published.

I've conducted a series of studies looking at selective reporting biases in drug and tobacco research and I'll just give you an example from drug studies. This actually was prompted by a med student working with me at the time and she had the great question, are all drug studies that are submitted to the USFDA for drug approval, are they published because as a clinician, I would like to have the same information available to me that the FDA did to inform my prescribing. Basically I

would love to do a similar study at the Therapeutic Goods Administration here, but they don't make the information public like the FDA does, so this is why we use the FDA. The simple answer to this question was no. Over a period of years we found applications for 33 novel new drugs and these new drugs were approved on the basis of 164 drug trials. Of those 164 trials, about 80 per cent of them were published in the medical literature, so that's not too bad; most of them were published. But if you look at it by drug, it's kind of interesting. So all the trials were published for only half the drugs, which is just amazing to me you don't have that information. No trials were published for two of the drugs and that included a drug that was only approved for children.

We also found examples of selective outcome and analysis reporting. The FDA has the clinical study reports from the companies and these study reports list all the outcomes and 41 of the primary outcomes were missing from the papers that eventually got published and none of these were favourable to the drug, so they all gave it a thumbs down. Interestingly, the papers also had 15 extra outcomes that the FDA didn't see and all of these gave the drug a thumbs up, they were all favourable. We also saw some changes in statistical significance with the published papers showing more statistical significance. The bottom line is that all of these changes and selective reporting meant that the scientific studies published about the drug made the drug look more effective than it actually was based on the data in the clinical study reports. We can't do this in ever field. In every field we don't know what's not getting published or what's not getting out there.

Things get really interesting when you start to look at how studies are being conducted behind the scenes and litigation has actually given us glimpses into what's happening. I think it's been the only good thing for me about living in the US so many years, is that our legal system has given us a number of settlement agreements where the courts release previously confidential documents from cases that have investigated harms from tobacco, drugs and chemicals. All of these documents are made publicly available on the internet, so any researcher can go look at them. This has given us some particular insight into how industry sponsorship or conflicts of interest have affected the publication of science. Again, I'm just using the drug industry as an example and we did a series of papers around a drug called gabapentin and found that research and scientific publication were actually part of the industry's marketing strategy. That's really important because it's telling us we think research is always done for new knowledge and I think we would really like that, but in some cases it's done for other purposes or used for other purposes. It's not just marketing, it could be used for advocacy, whatever.

What we found was this strategy came down to managing the publication of every individual study for this drug. This is a document from a lawyer and this study, if positive, that means statistically significant, will be published in medical meetings. Then this is talking about other studies and this particular study will not be published. The lawyers wrote a document saying exactly which studies should be published and which shouldn't and we've looked at all these studies and basically it's aligned with their statistical significance of the efficacy outcomes. Every one of those clumps of numbers at the bottom there is a study and going up the vertical axis, you get a greater P value, so statistical significance studies are at the bottom of the graph and the ones that are not statistically significant are at the top. What you see in the upper corner in the red dots, the unpublished studies are clustered up there and they're not statistically significant. This was the plan that was described in the documents and that's exactly how it played out.

Lastly, we can have bias in interpretation of results, otherwise known as spin and again, we're very, very familiar with spin in the media, we're familiar with spin in politics and for a scientist, what this means is it's not about the numbers, right, the numbers don’t change but it's how it's interpreted. As we heard this morning, I think spin is not necessarily a bad thing; people will interpret their data in different ways and I think it's just important to realise how that might affect and assessment of whatever is in the study, an assessment of a drug. We've recently conducted a systematic review of studies of spin and we found 35 studies that assessed the prevalence of spin by different study designs. What amazed me is it's actually pretty high; so observational studies, almost 90 per cent of the ones in the samples had evidence of spin. It was less prevalent in randomised control trials, a little over 30 per cent and in systematic reviews, it was even less common. Spin was defined in many different ways in these studies, but the most common manifestations were making the results look larger than they were, claiming that there was statistical significance when there was not and inappropriate claims of causality. Whenever I present this or talk to the public about it, the question always is, well doesn't peer review catch that? Actually, no; I advise people not even to read the conclusions of studies.

Why does all this matter? It matters to me because I'm actually quite an optimist and I do think we can use evidence to influence policy. I see that as the foundation for systematic reviews that are being increasingly used to advise policy, many governments now require systematic reviews for environmental risk assessments, for example, NH&MRC is writing new guidelines about use of systematic reviews in public health guidelines, including those wind farm guidelines. They're being increasingly used but if the underlying evidence is biased, then the systematic

review will carry forward that bias unless we can assess it and account for the bias somehow. It can be biased in any of these areas.

We have another issue which I think really gets at to the heart of what we've been talking about today and that's that in the Institute of Medicine in 2009 did a report on conflicts of interest in medicine and this report put it very well. It said conflicts of interest not only hurt the integrity of scientific research, but they also hurt the public's trust. I think that's what we're really grappling with and that has been the downside of much of this coverage of conflicts of interest, is that it hurts trust and people don't really know how to assess the evidence. I think it's very, very important to note that this is not a problem of bad apples or a moral failing of individuals somehow, that much of this bias may not be a conscious effort of individuals and also that it extends across various types of funders and financial conflicts of interest, it's not just the pharmaceutical industry which I used as one of my examples here. To fix this, we really require some structural solutions and we need to give greater weight, in my opinion, to research that is free of conflicts of interest.

I'll just end by talking about some of these structural solutions that are certainly happening in biomedical research and that we hope will expand into some other areas. The first is disclosure and as I noted, disclosure is far from perfect, but it's better than 20 years ago when we didn't know what funding sources were at all and when there were no conflict of interest statements. I think we need structured reporting of this information, just like we have structured reporting of methods, because as you see, it's very hard to interpret these conflict of interest statements.

We have a lot of investigation going on with different funding models, particularly for pharmaceutical research. For example, Italy has a program where pharmaceutical companies pay a fee to the government that is proportional to the amount of money they spend on advertising in that country and then that money is all pooled together to fund studies that the pharmaceutical companies themselves would not fund. These tend to be post-marketing studies of harm or systematic reviews that actually compare head-to-head comparisons of drugs. We have very little data on that; most drugs are compared to placebo not head to head, so how do you know which one to use? This is a very novel program; Spain experimented with that for a little while and has no funding for it now. Also people are talking much more about consortia of funders working together just among companies and with government to reduce some of these biases about the questions.

Next we have protocol registration and this is used to combat reporting biases, so the protocol can be checked to find out if the study has been published and if it was done according to the protocol. Many top quality medical journals now, like

Lancet or BMJ, JAMA that you might have heard of, they actually have editors on staff who will check submitted studies against the protocol to make sure this is a high resource activity. Certainly not every journal can afford to do that, but it's a great way to reduce these reporting biases. There are many, many moves toward open access to data; the Cochrane Collaboration has been extremely active in this area during my tenure when I was co‐chair of the board and basically these help us combat reporting and analysis bias and spin, because it makes the raw data of the studies available to researchers when the study is published. Again, it's a resource‐intensive activity, if you have the resources that data can be analysed in your own way and that can be made transparent. There is a lot of momentum around open access data these days.

We also have reporting guidelines and so this has been a whole variety of journals and I know this is really getting an uptick now in some of the environmental journals I've been working with, to have very structured reporting of the research and this helps us get a completeness of reporting and to minimise spin and ironically, they do not include reporting guidelines for conflicts of interest, which I wish they did. Then we have evidence synthesis, so rigorous evidence synthesis that's done by independent organisations, such as Cochrane, always includes an assessment of the risk of bias of the studies included in the synthesis. This is good because it lets people know about the body of evidence on a particular topic, where the biases might lie and whether those biases can be controlled for in some way. The bias is assessed not by the people doing the research, but by a group that's synthesising it.

I'm actually quite optimistic about teaching people critical appraisal skills. I do a lot of work with journalists and consumers about how to critically evaluate research. I'm even involved in a project, my colleague Andy Oxman in Norway, where we're working with school children in Uganda using a cartoon book that shows them how to evaluate information that they get from studies about their health. It's been very, very successful, so I really think we can give people the skills they need to evaluate research. I don't think we can shy away from, in some instances, excluding people with conflicts of interest from certain decision‐ making areas. This is done sometimes on guideline panels, for example, where people are going to be involved in the recommendation made by the guideline,

they can't have any conflict of interest that would be related to that recommendation. The most easy example again is drug, if you have a conflict of interest related to a drug company and those drugs would be recommended in the guideline, you're excluded from making the recommendation.

I think we do need to put the horse back in front of the cart and prioritise our solutions to minimise the influence of conflicts of interest on the evidence itself. Thank you.

Prof Griffiths: So we have five minutes for questions.

Question: Thank you, that was a really interesting talk. It's begging the question of if we ask to be more integrated, more engaged in research, how that may actually increase the potential conflicts of interest and how do we manage that?

Prof Bero: Yes, so I work at the Charles Perkins Centre and that's the whole reason I came to Australia, because it's such a multidisciplinary place to work. I think you're question, it raises two points. If you mean collaborations across disciplines, I think that actually really helps to minimise conflicts of interest, as a matter of fact. If you mean collaborations with industry and I know industry in Australia is quite broadly defined, then I think that's when we need to have some of these structural solutions in place. For example, at the Charles Perkins Centre, we do review any commercial engagements with industry according to a risk benefit framework. There are certain things, because we're a public health oriented group, that we would just exclude collaborating with that industry because they have a bad public health track record. If we collaborate with other industries, we would make sure that there are mechanisms in place that would prevent some of these biases I'm talking about, like any control over publication or any control, like total control, over the research question. I do think we need to have those structural mechanisms in place and I'm certainly not opposed to collaboration.

Question: My question is related to the fact that there's probably around Australia drawers full of negative results and in a sense we're biasing hours of scientific output because we don't want to publish our negative result and nobody wants to ‐ journalists don't want to publish it. How real is this an effect on knowledge and its progression?

Prof Bero: I only really know about drugs and tobacco as far as the extent of publication bias and there it's so great that I would imagine in other fields it's great. One thing we're starting at the University of Sydney right now is a repository for animal studies. It's very hard to know how many animals experiments aren't published, but we're going to take every protocol that's approved by the Animal Ethics Committee and we're going to put it in a registry and it's going to be open to other universities as well and we'll be able to get a handle on publication bias in animal studies. I'm very glad you said that mostly it's the ‐ there's very good evidence showing that publication bias occurs because investigators don't submit the papers. There are actually ‐ negative papers and we've done research showing this, are accepted by journals at a pretty much equivalent rate as the positive papers, if they're interesting and they meet other methodological criteria, but investigators are much less likely to submit them or they give up after a few tries. I'm always asked how many of those I have lying around and I think I still have three studies I probably never published, except on the internet now; I did get them out there.

Prof Griffiths: A final question.

Question: It's a comment actually on the conflict of interest. The entire Sri Lankan team of research on (indistinct 3:34:46) is charged with making a decision on a carcinogen and I happened to be involved in the helicobacter one, but the really good strategy was they would have three agents and they had a team of eight from each who had to have peer reviewed data and when that position is made, all 24 of us had to vote on it, so in a way, you got rid of your own inherent bias because of course you know helicobacter and it's property and that was a very good situation of coping with a slightly different bias, our own natural bias in our own research. It was a very effective strategy.

Prof Bero: Yes, yes, I'm familiar with that model and I think that's, in the environmental area, that's one reason a lot of the agencies like EPA has now been mandated by congress in the US to do systematic reviews for their risk assessment. Those reviews will be done by a group that's independent from the committee. Then again, you can do a consensus process on the committee or you can do a voting process, which makes it very clear who's going in what direction.

Prof Griffiths: Well it's now time for lunch, so let's thank Lisa. So we have an hour for lunch and we're back here at 1:40.

LUNCH BREAK

Prof Griffiths: So welcome back. It's an enormous pleasure for me to be able to introduce our next speaker, Sir Peter Gluckman. Peter Gluckman is somebody's work I've admired for a long time. You can read about his long and distinguished career as a medical researcher in the conference materials, but for me also Peter is somebody who's made important contributions to thinking about evolution, I'm thinking of a book that I use a lot; Plasticity, Robustness, Development and Evolution, which he co-authored with the late Sir Patrick Bateson of Cambridge University who passed away last year, it's a great loss to science. So contributions to basic science, Peter's been a very effective communicator of his work and work in related fields. Some of the books that he's written with Mark Hanson in English by medical researcher Fat, Fate and Disease, which I think is an excellent introduction, and Mismatch which is another book really focusing on the crisis of obesity that faces us and its biological causes. But for the last, I think, nine years now Sir Peter has been the Chief Science Advisor to the New Zealand Government, so he's had an enormous wealth of experience in dealing with the nexus of evidence and politics. He's going to talk to us today about the role of evidence and the expertise in policymaking, the politics and the practice of scientific advice. Peter.

Sir Peter Gluckman: Thank you very much Paul, it's a delight to be here and thank you to the governor for being here to listen to these words. I'm going to focus on three questions and in many ways they follow on from Emma's talk somewhat. What is the role of the institution of science in the world where trust is declining? How do we ensure respect for scientifically derived knowledge in this environment, and particularly for policymaking? How do we ensure that policymakers are more likely to take into account the role of scientifically derived evidence in their decision-making? Now I'm not going to dwell on the post-truth, post-trust, post-elite, post-whatever world we're in now because I'm sure it was talked about this morning, I'm sorry I wasn't here for all the talks. But let's just remember the manipulation of facts and evidence is not new, it's been going on since religion was invented, since various forms of power structures developed ten thousand years ago in villages and in cities. What we have rather is a massive amplification of the effect because of the powers of digitalisation, which have got many effects which are listed on these slides, which I won't go into now. But it's also had this dramatic effect of changing

the positioning of the different publics in relationship to the policy community and it's increasingly affecting the way the policy community operates. Oops, I've gone too far. Virtually every government at any level - every issue they face has a scientific component to it and here I must emphasise I'm using science in the most broad definition you can imagine to include the knowledge based humanities as well. But we also need to remember that science will never alone make policy, which is why I've eradicated the words 'evidence based policy making' from my lexicon, because evidence can inform, it cannot be the only construct in which policy was made. And where science is of most use is actually where the science is most contested. Governments are usually making decisions in situations where the science is not complete, it can never be complete and it's often most contested. And we now face this challenge that the science of the most interested governments is actually in areas which are most contested in terms of public values, and I'll come back to that momentarily.

And so the issue is how do we ensure that the science is reliable, robust and how will it be used? Will it be used well or will it be misused or ignored altogether? And I think we need to remember that science is not the only form of evidence. For most people, science is not their primary sources of evidence. For them, evidence is tradition and folk knowledge, evidence is the knowledge that's within their peer community, it's religion for some people and it's anecdote, experience and observation. And certainly, for most politicians it's anecdote and observation are the primary things that influence them. So where does science sit in that hierarchy and how do we work to ensure privilege for science in that hierarchy? And of course, as we've talked about by the previous speakers, science is defined largely by its processes. Science is not a collection of facts; science is a collection of processes which are defined to eliminate bias to the extent they can. But that's not to say that science is value free as the last speaker spoke about, of course there is values all over what we choose to study and how we study it. But in the context of my talk the most important value judgement within science is over the sufficiency of evidence on which to reach a conclusion, which might have impact on society, and we'll come back to that because I think many of the debates that we have are really over the quality of evidence and its sufficiency on which to draw a conclusion. As Heather Douglas wrote about in her brilliant book, blocked on the name, Heather Douglas and Ken, I just had a moment (4:38:32) collapse, never mind. It's this inferential gap between what we know and what we conclude, which is of so much importance in policy space. And within all this we're really talking about the changing nature of science. Science has changed dramatically in the last 50 years and it's going to change much more in the next decade or two, as we see the shift from linear to complex science, from deterministic to probabilistic science. And from what (04:39:06) and Jerry Ravetz would call post-normal

science, that is science where we're dealing with systems, where it's complex, there's unknowns and no matter how much science we do there'll still be unknowns left at the end of the day and uncertainties. The stakes are high, decisions are urgent and it intersects dramatically with community values, and those community values are in dispute; climate change, environmental matters, public health matters. Virtually every contentious issue that government thinks about actually falls into this definition. It's complex, we don't have all the answers, it intersects with public values, which are in dispute and of course that's where a lot of the conflicts emerge and where the difficulties of how policy and science intersect are so great.

And now we're seeing a new phase of development, which again was talked about by the previous speaker. How do we address these conflicts? The emergence of extended peer review with the community rather than just professionals that review science. The true development of co-design and coproduction all part of the solution. But that's not my talk for today, that's another talk. I've already said all of this so I don't dwell on it but I said that, we'll just move on. The other side of this is that unfortunately because we are engaging in science which engages with disputed public values, science can easily become the proxy for debates which are not about science. We've seen that in climate change where the real debate is an economic debate and it's about intergenerational and north south economic issues, not about the science of climate change. But we've seen it in GMOs, we've seen it in fluoridation of water, we've seen it in the United States about stem cells and about reproductive technologies. There are many issues in which its easier for people to debate complex science and cherry-pick the odd observation, rather than deal with the true issues that cause the debate. And in my experience the best way to deal with climate change sceptics has been to challenge them and actually say, "You know this is not a scientific debate. You know this is really a debate about values and you're not being honest and having the debate that you should be having." And we have now a lot of evidence, particularly from the GMO and from the climate change literature, of course, that just pushing more science of people with different world views will not resolve the matters or make matters worse. And so now we come to this issue of trust in science as an institution, which the last two speakers also alluded to. Clearly this has become more complex in an environment where science is now dealing with these complex issues with societal values in dispute. But I think there are bigger issues as well and some of them being alluded to. The sides of the endeavour, three million papers last year, seven million authors, 30000 allegedly peer reviewed journals, most of which are never read and yet less than 100000 people with an H factor, not that I like H factors greater than 10. Think about that system. We've had this massive utilitarian transition in science, which we've all welcomed because it's

allowed governments to put more money into science, but science has now been possessioned in a much more utilitarian way and then that's led to this raft of incentives, particularly on universities, which have led to this bibliometric disease, which I would love to treat and I'm not sure how to treat.

Now we're seeing the overt politicisation of science, which is somewhat new. We're seeing these proxy debates and this issue of the relationship of science to the publics and had I had a chance to ask a question of the last speaker, I would have said the biggest challenge we now have is actually what guidelines and ethics should surround public communication by scientists, because on the one hand as citizens they've got the right of free speech, but on the other hand they're standing up and saying they are speaking for science and there are some real issues there that we have not grasped with and I encourage you to look at the Japanese Council of Science work done after the Fukishima debate to see how they are struggling to deal with this. And then we have - again it's been mentioned - intellectual silos and the real challenge of trans-disciplinarity. How do we marry the humanities and social sciences with the natural sciences? We say we do it, very few people do it. So onto the main part of my talk. Science and policy are very different cultures, they have distinct methods and epistemologies. The arrangements between them are influenced very much by societal culture and so forth and what had now come very clear is there's a need for boundary structures to act as translators between these two communities and I spend most of my life these days around the world helping countries through - in my chairmanship of the international network and government for science advice - thinking about these issues. As I've said, policy is really determined solely by evidence. Policy is really made around a whole lot of considerations, public opinion, political ideology, electoral contracts et cetera, et cetera. But what science can do and uniquely do if it's well presented, is deal with the issues of the evidence of need, the possible solutions and the impacts - the multiple impacts - of any possible solution chosen. There's challenges at the interface, as are summarised here; too much science, the changing nature of science, the post-normal nature of science, the different perceptions of risk that scientists have, which is often actuarial as opposed to the perceptions of risk the public have and the perceptions of risk that politicians have, which is largely about the [ballot] box. And all of this plays out, there's different perceptions of expertise, increasingly policymaker or policy analysts think that Wikipedia or Google searching is enough on which to come to a scientific conclusion. We have hubris on behalf of the scientists, we have hubris on behalf of the policymakers and there's all sorts of issues at the interface and I could go on and on and on. Now for scientists they think, "Hm, how does policy work?" And you see this beautiful picture that looks like that. I assure you it's a total myth, it looks far more like this, which is the simplified version, because how policy

emerges is unclear, it comes from formal and informal actors, elected and unelected actors that somehow coalesce somewhere in this process to influence - in this case - executive of government. Very, very complicated and if you think about how lobbyists work, how we work when we're lobbyists, how the private sector works, how public opinion works, the media works, you immediately see how confused and complicated policymaking really is.

And so the issue is how does evidential input work? Well evidential input has to work all over the place in this system and I think this is a really important point, which is often forgotten, that it needs a concerted effort to maintain evidence in front of the policymakers. So what are the primary functions of science advice in this setting? First of all I think it's to help the policy community actually understand complex system; be it a social problem, be it an environmental problem, be it a transport, an urban issue. Often they have seen bits of it and I think it's system thinking by scientists that helps so much in clarifying what can be done. Secondly, it's about helping the policymakers see the range of options that could be applied and understanding the implications of each of those options, because policymakers always have options, they always have the option of doing nothing to start with, which is usually their default position and from then onwards, you've got a range of options which will always have spill over effects. And thirdly there's a role in having and evaluating policies that have been implemented and then there's some special areas in emergencies. Most emergencies be them natural disasters or a terrorist event, have a scientific or a technological component. I could tell great stories about the need to make sure the policymakers understand what the evidence is saying in such situations. Then you have the issue of technology and forecasting a big story in its own self and I don't have time today (04:48:47) diplomatic issues but if you think about all the global challenges that we face and encapsulate in the SDGs, Sustainable Development Goals, many of them have a scientific basis and science diplomacy is going to be critical at the global level and thinking that through and I'll come back to that on my last slide.

Prof Griffiths: So in its simplest, you can see policymaking is about making choices between different options which affect different stakeholders in different ways, with different consequences, many of which are not certain. And I think that the major role and the core presumption of scientific advice is that it's more likely that government can choose between the options in a way that will be of better outcomes. Now this is also not often appreciated, policymakers have limited bandwidth. The policy cycle is shorter and getting shorter because of the impact of digitalisation. The science they need is usually incomplete and much ambiguous and the last thing I will ever let any of my science advisors say to a government is the words 'more research is needed.' Governments have to make decisions, if they can't make a

decision then it's don't have a policy acceptable solution to them at that moment, they move on and drop the issue. The other thing is you cannot expect politicians to be scientific referees and that's where I think there's going to be a real issue about scientists acting in the public space, that you can see contested science being argued in a way that can be very confusing. I think it's going to be really difficult and we need to think this one through carefully. What are scientists good at? We're very good at problem definition. Climate science has done a great job. We're less so at finding the solutions that the science tells us about because usually it involves different disciplines to the disciplines that define the problem in the first place. Climate change was all about physical scientists, climate change solutions is about economics, about social science, about different technologies et cetera, et cetera, it's got a whole different basis to it. Often scientists approach the policymaker with a fixed solution in mind and considerable hubris and that's just a no-go. And so the issue in my mind is how can we understand this and use this to improve science talking to policy and policy talking to science? Well there's many potential elements in a science advisory ecosystem. I've just listed them there from the role of individual scientists and universities, research institutes, through to the national academies, the government advisory boards, to science advisors such as myself, the role of parliamentary libraries and so forth. There's an immense number of possible players in this ecosystem and you don't need just one, you need several. What have we done here? It's jammed, can you go to the next slide please?

Sir Peter Gluckman: So what I've done here is break this up into what I think are the four categories of roles in this interface. There's the knowledge generators, the scientists who generate knowledge, there's the knowledge synthesisers and we heard from one just in the last talk, from the Cochrane Centre and so forth. People that aggregate the knowledge and try and make sense of what it means. And then there's the knowledge brokers who have to translate that science to the policymaker and translate the policymaker's needs to the scientists. And then there's the policy evaluators and you can see that you need more than one structure in your interface if it's to be effective. Then you can think about other ways too. You can think of another dimension. You can think about policy for science, how the science system operates, which is what scientists often think about as the only role of evidence in policy is to give them more money. Then there's evidence for policy, as I've talked about, policy implementation, policy evaluation; not the same thing. A rise in [scanning] in crisis management. And again you can see that there's a raft of structures and institutions needed to achieve an interface. I've used this word 'brokerage' and I want to talk about it a little bit more. Roger Pielke wrote a book, The Honest Broker, in which he defined that there were different ways we can communicate. We can be advocates where we go in there wanting a

particular solution or a particular outcome or we can be brokers where we actually transmit the knowledge in an appropriate, reasonably values-free way, because it can never be absolutely values-free, to the policy community, and that's about what is known, what is the consensus; if we go beyond the consensus, why are we going beyond the consensus? What is not known? And often the most important thing you can say to a government is, "We do not know." Other caveats on the data, the inferential gap between what we know and what we don't know, what are the risks involved? What are the options and trade offs? What are the consequences outside the science area that each option might bring? I never make a recommendation, I'm always talking about what are the implications of each option. It's for the policymakers to make the values judgements weighing up all those other considerations that come into play.

And then you have this other divide. People like myself are inside the system. I can talk to the prime minister or ministers any day. I talk to the cabinet office literally everyday and that means that I can see all those different interactions that are going on and their complex [message] cycle. That is the advantage of science advisors and scientists within the system, they can see what can be done. On the other hand, they're not as fully independent as an academy or academics on the outside. But they are more designed to do the deliberative reports, the formal reports that can impact on governments and what I keep saying is we need a balance between internal and external mechanisms. Another way to look at it is to think about informal and formal mechanisms. Informal mechanisms is what I do when I brainstorm with the prime minister or a minister, when I say 'prime minister' that's the most stupid idea I've ever heard. Or, "Prime Minister, would you like a report on this?" Or whatever or, "There's a problem with your thinking" and you can challenge and it relies on a very trusted relationship between science advisors and the executive of government. On the other hand you've got the writing of the formal reports and that depends on how it's done. Is it a requested report or is it a profit report that's not needed? And one of the problems of reports is often it's there to show off the intellectual brilliance of the report writers rather than to actually answer the questions that policymakers need and I think that's something that the academies are working around the world to try and learn about.

And that brings me to the role of academies, seeing we're speaking here with the Royal Society of New South Wales. Academies have a critical role. They are a place at which multiple disciplines can come together and write a critical report, a report on any subject. But sadly most academy reports, even from the Royal Society of London, end up immediately on the round filing bin on the floor and are not read and that is because most are not - shall we say - negotiated before they're started with the government of the day to see if the government actually

wants to get it. Because if you put a question forward that the government doesn't want to hear the answers to, it's probably not going to succeed. Often even when they are given a question by the government, they don't realise what the government needs by way of response. There's a whole lot of [issues] here and I think academies have a challenge in this post-trust world of how they'd reinvent themselves but that's another story. So in summing up, the skillset needed at this interface whether it's outside from academies from other think tanks, Cochrane centres or whatever, and that needed [inside] are compatible but slightly different and I've emphasised in black what I think are really key for those who are inside the system. I think anybody who's engaged in this interface needs to understand the complexities of the system. They need to get beyond single disciplines and realise that virtually everything that a government deals with in science has a social component to it as well as a natural science component to it. They need to - I think - employ brokerage rather than advocacy. If you go in there saying, "We know the answer to this, you must do that." You'll find a thousand policy analysts writing immediately reports why that's not the case and the scientists don't understand what they're doing. It needs diplomacy, it needs policy entrepreneurship, it needs good communication to the four audiences; the politician, the policymaker, the public and media and the science community. The four distinct audiences all of which need to be communicated with. We need to understand this new [environment] we're going to go into, humility is the most important skill you can have in talking to a policymaker. You must never try and take their role away from them, they are the ones who are there to judge the role of all those other mentions. They are the ones that need to opine on values and consequences, not us. And you need to maintain integrity and trust with all four audiences and there's obviously a hierarchy of trust. I can't do my job if I don't have the trust of the prime minister, the ministers, the policymakers but it's also critical I have the trust of the public. The science community worries that I don't advocate enough for them, but that's not what I'm there to do, I'm there to help a government make better policy choices, not just to fight for another billion dollars for science.

On the other hand, the most important thing academies can do, they have to maintain the trust of the academic community, otherwise they lose their standing as an academy. So you see there's different hierarchies of trust involved. There are real challenges here. You need an ecosystem and few countries have a comprehensive ecosystem. Britain does reasonably well, I think New Zealand does very well, I'm not going to comment on Australia; when you deal with climate change I might. We have these challenges, what is a fact? Is robust science available? Who defines the knowledge is robust and reliable? We have this huge emerging issue of social licence. As the innovation and science machine gets

faster and faster with the nanotech, biotech, digital tech, geo-tech, wherever it will be, there'll be more and more issues of social licence emerging and frankly I don't think the natural scientist community and the innovators have been thinking enough about how to develop and maintain social licence. I'm heavily involved with the OECD on the issues of what will be the impact of AI and all that's associated with it on the concept of human wellbeing. What does it mean at a level of individual, the level of society, at the level of the nation state and its power? Because there's no doubt that there's a changing dynamic in the jurisdictional space. And what I've argued for in this talk is that any effective advisory system needs to have an informal, an internal component, but it cannot work unless there's an effective external deliberative component coming from the broader science community and particularly from academies. I just want to give you one advert from my organisation, the International Network For Government Science Advice, which is an autonomous part of the International Council Of Science or the International Science Council, it's about to be renamed with the merge of the World Social Science Council. We deal with all these issues at many levels in capacity building. We have over 3000 members now in about 80 countries working with many, many governments and individuals and out there on the website at the moment is a manifesto about how science and science (05:02:06) needs to be developed to deal with the sustainable development goals. It's open for comment from anybody 'til the end of January and we would welcome comment because it's going to be presented to the STI forum of the UN in June next year and the website is there and membership of the organisation is free, so you might want to join up. Thank you very much.

Unidentified: So we have ten minutes for questions.

Unidentified: So Peter, thank you for a marvellous talk. I wonder if the essential problem is a paradigm of science. It's a quest for total understanding and it leads to things like the precautionary principle, 'You don't do it if you don't totally understand it.' If I cross across to engineering, it's all about risk. If you're designing an aircraft, if you overcompensate for safety, you don't get off the ground. Now I just wonder if in our scientific advice we teamed up with the engineering elements of risk we might do very much better in providing the sort of advice we need and I'd simply offer the comment that in some of the environmental advice we've provided in Australia that has worked exceptionally well, the community understands risk. The human condition is living with it and I just wonder if in fact we can understand our barriers.

Sir Peter Gluckman: Politicians are better at dealing with uncertainties than scientists are. My experience of politicians is they do understand that virtually every decision they make will have unintentive consequences but the system is open and has far more complexity to it than any analysis will give. I think the issue of the precaution of

principle depends on what it means, there are at least 27 different definitions, I think, in the literature and you've got these extreme definitions that lead to total inaction, you can't have any innovation without some risk and so I think governments actually understand risk quite well, I think the problem is - as I've said before - the definitions all at risk means are different to the different entities. My office is working on the publication of a national risk register for New Zealand with the cabinet office, which will be a really innovative idea for this part of the world, Britain's had one for several years, and it's been interesting to watch that even in the preparations of the heat maps around likelihood and level of impact, the debates of perception that go on between scientists who use a purely actuarial value, the fact that impact has a whole lot of qualitative components to it and it's been really interesting to watch these different perceptions of what these things mean. Now I think that we've just got to be - I think your first point in your opening remark was really important. I think scientists do [two one] things, they think they know all the answers to everything and they usually don't know the answer to what the policymaker really needs to know. They're awfully afraid to say, "I don't know" or, "We have these uncertainties" thinking that the policymaker will see that as an excuse not to listen to the science. In my experience, the policymaker is comforted by saying, "We don't know everything but the evidence we have would favour these options over these options for the following reasons." Now when it comes to the issue of precaution, I think there we're dealing with the problem of perception of risk and there we've [got to do] far better at how we communicate what we mean to the public. I mean we lost in New Zealand the GM debate 20 years ago because it was immediately taken over by Frankenstein foods and everything else that went on. It's very hard to get that back. On the other hand, sometimes you've got to just work with governments to say, "Some things have to be done." The big one in New Zealand has been fluoridation of water where 20% of people think it's a conspiracy to do God knows what to them and vehemently oppose it and they get themselves elected onto every local authority controlling water and getting government to actually override that and say, "No fluoride is a major public health benefit" is when you have to deal with this issue and just ride over and that's a difficult discussion in harm reduction science. We're dealing with at the moment - we have to put folic acid into bread, that's another contentious issue.

Unidentified: A wonderful presentation, I very much enjoyed it. A few years ago I did a study with colleagues here where we interviewed about 25 to 30 very senior policy activists, you know, ex premiers, health ministers, senior bureaucrats, head of NGOs, whatever, and the core question was, "Who do you consult with? Where do you get your expertise from?" That sort of question. Much of what you said, of course, they put back particularly about trust. But many of them emphasised something that you would ask anybody in this room, who do you choose to hang

around with? And it was intensely personal factors, I like those people, they're good company. You would choose to be with them, things like that, intangibles that you'd never read in a textbook about how to go about this stuff but so many of them said it.

Sir Peter Gluckman: I agree with that, absolutely, which is why trust has got more to it and I emphasised diplomatic skills as well. We did two things in New Zealand you don't do here or one thing you don't do here. We have departmental science advisors, so as well as myself reporting to the prime minster, I have (05:08:17) departments of government, a science advisor, as they do in Britain and they're all part-time academics doing the role for either 5/10ths or 8/10ths of their time. Their primary role is to get alongside the analysts within the departments and influence what they do. We work as a group as well and that's done a lot, the government loves it because it's breaking down silos between ministries and it's probably being very effective, particularly in the social sector in that regard. So much so that treasury now relies on the science advisors to give them independent advice on budget proposals in that sector. And the second thing I've done is twice now over six years; three years ago, we have done surveys, very structured interviews of government departments on their attitudes to science evidence and it was the first one which was so disappointing that led to the prime minister agreeing we could have departmental science advisors. The second one showed some progress but what was most interesting to me was we did all this work, extensive interviews over a year with every government department, I went in to talk to the prime minister about it and he said, "Don't tell me, I bet I can rank the departments that use evidence well and use evidence badly." And he got it right. He knew, he knew, that was John Key. But the other thing that I want to mention in that - and you might from that work out which departments might not have been doing - using evidence very well was some departments actually said specifically, "We don't want to talk to scientists because scientists only lobby us and present the arguments that will give them more money. And for that reason we explicitly exclude them from our inputs. You can see why some might have grabbed that view, now it's a cynical view and it's a view that hopefully we've eradicated by inserting science advisors into the department, but there is that echo and it's real and I think that's why intermediaries are another part of the solution.

Unidentified: One more quick question.

Unidentified: Great talk, thank you. I've been clicking away at the slides so I hope you don't mind if I plagiarise wildly.

Sir Peter Gluckman: I'll email them to you if you want.

Unidentified: That'd be great.

Sir Peter Gluckman: Anybody who wants them can have them by email.

Unidentified: Lovely. So my question really is of all of the groups that you listed in that ecosystem, the only groups of scientists per se, if you could call that where they're acting as a collective, were the academies and they were doing reviews.

Sir Peter Gluckman: Not really, there's [what] works units, there's a range, there's think tanks, there's a range of other organisations that do it collectively. There's government advisory committees, there's a broad range of structures that do that and clearly whenever I set up a panel, I mean we just did a major report on water last year, enormous report, we had 30 different water scientists working on it plus another 10 doing peer review on the report. So I don't think that's quite fair.

Unidentified: So I guess my perception of think tanks is they don't tend to include many scientists, active scientists, whereas the organisations that do include them are the academies, for example.

Sir Peter Gluckman: Yeah.

Unidentified: So generally - and I'm talking about natural scientists, I guess, at the moment, rather than social scientists - governments would be interacting with academies or they'd be interacting with individual scientists.

Sir Peter Gluckman: Correct.

Unidentified: Or teams of scientists that they've put together. Is there a value in creating another type of component, which is the science think tank or the science policy think tank?

Sir Peter Gluckman: Well there are some around, for example, the (05:12:19) Institute in Holland, Netherlands, which is admittedly funded by the governments, so it's a government funded institute, is specifically there to be an independent think tank on new technologies and it does a marvellous job and if anybody's interested in the issues and challenges of new technologies, reports from the (05:12:38) Institute are just superb. Nordics have a number of these institutes in different areas, Finland has a couple, Denmark has a couple, so it's not it's not been strong in the Anglophone dimension but in the European dimension there's a lot of it. But they don't have science advisors so it's interesting that you get these cultural differences, there's not one non Anglophone country that has an individual science advisor.

Unidentified: Thank you Peter.

Prof Griffiths: So one of the major topics in all the talks so far has been the changing landscape of media technologies and our next speaker, Professor Andrew Jakubowicz is going to address that fairly straightforwardly. He's the Professor of Sociology in the School Of Communication at UTS, he's written books on multiculturalism, on media racism and diversity, and on cyber racism and community resilience. The title of his talk I think gives it all there; Algorithms of hate, how the internet facilitates the spread of racism and how public policy might help stem the impact. Thank you, Andrew.

Prof Jakubowicz: Technology, it's working, yes. Who recognises that green person? Anyone? How many recognitions? One, two, three, four, five, six, so the trendies and those under 40, okay, right. Our friend here is a gentleman by the name of [Pepe]. [Pepe] was once an extremely nice human being created by a very good American comic artist on the web and he was designed to convey a sense of sort of reckless optimism about the world. [Pepe] was appropriated by a group of gentlemen who operate out of a particularly dark part of the web called 4Chan and [Pepe] began to transmute into the symbol of the alt-right. During the election campaign in the United States last year [Pepe] became a hunter who went out trying to destroy Hillary and the democrats and there is, in fact, a rather remarkable picture of the magnificent 7 [Pepes] all of which seemed to have heads associated with Donald Trump, Jared Kushner and other members of the Trump team. So [Pepe] became really, really important and earlier this year the creator of [Pepe] tried to rescue him, right, the designer wanted him back and over the last eight months or so there has been a fairly internecine and hidden struggle amongst online cartoonists to rescue [Pepe] and the meaning of the internet.

So I want to start off with [Pepe], he's one of my favourite characters and empathy trumping hate is an attempt to try and capture one of the dynamics underlying this whole question of truth, post-truth, alternative realities and so on. Also on the screen there is a header from a recent article that came out in August this year, which looked at the way in which algorithms, which underpin and operate the internet, actually generate hate and really that's the focus of what I want to talk about. Because in my analysis I'm suggesting that the science that went into building the internet, the science that keeps it alive actually has a series of direct social consequences and if we don't address those social consequences then other social consequences will emerge and continue much to the detriment, I think, of everyone in societies such as ours. Okay, so the abstract you've got, this is a quote from a couple of weeks ago from a gentleman called Tim Berners-Lee, I'm assuming now we're in safe territory, Tim Berners-Lee, we're safe with him. Okay Berners-Lee was the guy who basically invented HMTL and created the world wide web. He runs a thing called the W3 Foundation, which is an

organisation which is committed to trying to make the web freer and he had this to say - and this is not the first time he's said it, he's just saying this from the beginning of this year at least, "My vision for an open platform that allows anyone to share information, access opportunities and collaborate across geographical boundaries has been challenged by increasingly powerful digital gatekeepers whose algorithms can be weaponised by master manipulators." Now the discussion of Brexit of course directly addresses that, Trump, the (05:17:59) that exist in Russia and so on. All of these are part of this story but in an earlier version of this Berners-Lee said, "And the thing that worries me most is that whatever it is we've created we've licenced racism to run free across the planet and the consequences of that for civilisation and democracy are very, very sordid if they're not addressed." In fact I don't know if people have seen in the last week or so the Lowy Institute has produced a new report looking specifically at the impact of social media on democracy and they take a more general view but within that the sorts of arguments I'm making today can be can be seen. Okay.

It all became fairly clear, I think, to people who watch this earlier in this year in March when advertisers across the world began to boycott alphabet, which is Google, YouTube and all those companies, and Facebook, because they'd discovered - or in fact newspapers looking for truth had discovered - The London Times and the Washington Post, they had discovered that there was something going on with the algorithms that drove the internet and what was going on was that all over the world when people who were excited by the idea of violence, misogyny and racism and so on clicked on their favourite sites, they would get served advertisements from Coca Cola, from a whole variety of big advertisers and the advertisers discovered that in order for these particular sites to work - particularly young men with money who are a very, very attractive component of the internet audience, the internet customer base, were being sought out by the automatic algorithms that ran many of these sites and when those boys went off to look at racist and sexist and violent material, the advertisers would follow them and so their advertisements would get served up on these sorts of sites. Not only that, every time the advertisement got served up on the site and there was any sort of click through to the advertisement, the site would get paid. So what we had was this rather interesting model where if you happen to be, for instance, a member of a neo Nazi group, you and your mates can sit round and just hit your site time after time after time and every time an ad comes up, you click on the ad, every time you click on the ad, the payment goes through to your site. So we had this situation earlier in the year where multinational corporations were funding many of these extremist and dangerous sites. The YouTube problem was the first but it grew and it expanded through all of them and I did a little experiment yesterday, before yesterday, in preparation for this talk, I'll come back to that, there

we go. This is a site called (05:21:17), people know (05:21:19)? It's run by a guy called John [Bannon] who was the advisor to President Trump, he's gone back there now. When I went to it - this is revealing a bit about where I buy my clothes but leave that aside for the moment - when I went there as soon as I clicked on it up popped an Oxford Christmas sale ad, right? Now had I clicked through on that, I carefully didn't, a small payment would go across to (05:21:50). There's also for people of a certain age the ad below that is for a particular osteoarthritis pill, I might have been more tempted perhaps on that one. Anyway, so what we were seeing in this stuff is a dynamic in which the fundamental structure of the internet drives certain sorts of economic and political directions. Over the last 12 months it's become clearer as we understand how the web actually operates as a structure of discrimination all the way through from the political economy through to the practices. An organisation, for instance, called the algorithmic justice league, one of my favourite sort of new online groups, points to what they call the bias of the code at [gaze] and they've gone in and they've actually looked at the arithmetic that drives this stuff and tried to work out exactly how the biases emerge and operate.

Science early in the year had an article which reported that [it's] an area of machine learning of semantics, there's an automatic shaping of the learning to reflect human biases in terms of race, gender and disability. So the very way in which things are read automatically, the way in which AI works is infected - is a good word - in sort of infused with these sorts of social differences. Facebook itself has discovered under some pressure because it's only the last year or so that it's been doing this, that when it looks at who it actually employs, it has a very peculiar structure and it's not surprising that therefore that Facebook makes certain sorts of decisions. It's diversity report from the United States where it employs nearly 13000 people, shows that 40% of their employees are Asian-Americans, 45% are white Americans. Racial categories in the United States are very curious, so just bear with me on this. Only 3% are black, even though the proportion of the black population is very much higher than that and only 5% are Hispanic. Facebook has begun to recognise that the fact that it employs particular sorts of people actually affects what happens, similar sort of discovery that was made by the Australian Defence Forces over a decade ago where it began to realise if it only employs certain sorts of people it'll get certain sorts of decisions and it's begun to diversify and actively recruit, so it's got a real mix. Facebook over the last year as well has been caught up in what's called 'multicultural affinity targeted advertising.' What this means is that there are certain categories of advertising that are targeted towards certain multicultural groups or targeted to avoid certain multicultural groups. So let's say if you're offering expensive, easy to get credit, this will target people who are American blacks. Why is that? Not because anyone's consciously

done it but because all the characteristics drive those algorithms in those sorts of directions. Darwin's been around a bit today so I thought I'd drop him in as well.

In 1875 when Darwin was trying to make sense and actually apply many of his findings to the human society around him, he noted that the truth in science - two great words we've talked a bit about today - should ensure that when men are exposed (05:25:34) to the truth of science, this should lead us to extend our sympathies to all men, that is he argued that the more you understood about the nature of the human race, the more you were going to be convinced that your empathies should extend to everybody irrespective of their racial background. Race as an idea has been extremely controversial and difficult to sort out. At the end of the second World War when UNESCO was actually trying to work out a position on race, the taken for granted attitudes of the day, the scientific attitudes of the day concluded that races were real categories, the human species was indeed divided into really different races but those races were inter-breedable, it's quite an interesting concept of how we thought about races at the time. This created over the next decade or more, this created a huge debate amongst biologists, anthropologists, social scientists and so on, who weren't prepared to sit with what - even at that stage - many people recognised was a very, I guess, 19th century and biologistic perspective. By 1967 when the United Nations was trying to in fact reach an agreement about what should be done about racism globally, a new perspective came into existence and this was the perspective that said while race may be a useful term to label differences between people, it didn't actually allow you analyse anything, it didn't allow you to predict anything and it definitely didn't allow you to say anything about hierarchies of different peoples. They made a very strong position, which remains to this day, that racism has no scientific basis whatsoever. Some of you may know that as we sit here today in Geneva, the Australian Government is being put through its paces by the United Nations Committee on the eradication of all forms of racial discrimination and there is a conflict of two realities being played out. There is a government reality, which says we are doing extremely well and we're an extremely equal and calm and joyous society, and there is the NGO presentation which is suggesting something different and they're both based of course on versions of what the truth is and what science is.

At the end of 1967 when they finally reached their conclusions on this, the UN said this, "Racism (05:28:15) the development of those who suffer from it, perverts those who apply it, divides nations within themselves, aggravates international conflict and threatens world peace." So by that stage it was fairly clear that the global community had understood that racism was going to be a continuing challenge to peace and human development over the coming period and so in fact

it has turned out to be. This is the Australian version of [Pepe] because local and global developed fairly recently over the last twelve months by an Australian, I guess you'd call it, alt-right, neo Nazi, white power player on the web. I don't know if you can read that down the back, this is called a groyper, right, it's a particular animal, it's called a groyper, and my friend who pulled this one up for me, [Cassie McBride] who is a social media activist who works on issues to do with racism, describes this as an exploitable illustration of Pepe the Frog resting his chin on his interlinked hands, variations of which are commonly used as profile avatars amongst the alt-right and new right on Twitter and the alternative social networking service of Wrong Think. Any people here hang out on Wrong Think? So it's worth going there just to feel your reality totally disrupted for two or three minutes. So groyper, in a sense, represents the idea of a [nativist] Australian white power push, which is expanding through the Australian internet environment in a very dramatic way. The work I'm referring to now is based on a project that began about five or six years ago, which is known as cracker, the cyber racism and community resilience project. The book I'm afraid governor is not yet available in my hands or you would have got one of them too. It's just been released and what we were interested in was discovering how the internet actually forwards racism. How does it make it possible for racism to continue? what are the main patterns of racism? What happens to people who encounter racism online? Who generates racism online? And what can civil society do about it? Now in the time I've got left, about another 10 minutes or so, I'm not sure I'm going to be able to make my way through all of this but again in the final paper I'll do it in a lot more detail.

So how does cyber racism grow? Well we argue that there are basically four key elements, each coming out of a different scientific and knowledge base that we need to grasp and then the problem is to understand how they interact with each other. The first of these is - and I've made some reference to that - is that the algorithms that underpin the economics of social media work in particular ways which are constrained by the profitability of the companies that generate them and that they are agnostic about their impact; that is these algorithms only have one criterion when they're put into effect and that is the optimisation of economic return. And in fact when it has occurred in the beginning of this year when Facebook and Google and so on decided they actually had to do something about the fact there was a strike by their major advertisers, they discovered that they couldn't actually rewrite the algorithms, there was no way they could do that. What they had to do was employ significant numbers of people and go into a sort of craft environment where individuals had to actually go out and look at stuff and try and organise for it to be reported and then stuff had to be taken down by hand and so on. It's a very laborious process and it demonstrated for people, like myself, how much the political economy of the underpinning structures of the internet produced

these sorts of outcomes. Secondly there are all the questions around state ideologies that actually constrain intervention by governments because governments can do things, we've seen how they've operated in different environments depending on the policy advice they get from the scientists and other people and the other lobby groups. The United States has very few constraints because of the first amendment and the various other elements in the Constitution. Australia sits somewhere suspended between the United States and Europe. After Brexit it's not clear at all where Britain will sit on all of this stuff because there are European human rights legislations which are likely to drift away as the new British Government takes control of these things but it depends where you are. The most successful intervention strategy of any government is the one being carried out by the People's Republic of China. It has almost total control over everything people do on the internet in China and basically racism doesn't happen on the internet in China unless somebody says that it should do, though there has been some trouble in relationship to Japan, actually trying to control that.

The third element which really takes us into the psychology of internet users is the social and psychological relationship between humans and computers and the sorts of environment that that generates. And finally of course there's the global and regional rising ethno-religious conflicts under the impact of globalisation where the internet has become a critical part of the infrastructure that allows those conflicts to flourish. And out of that you get cyber racism and in the Australian context cyber racism sits, if you like, between online cyber bullying on the one hand, and online radicalisation on the other. While government - there's very strong government focus on the bullying side of things, particularly for children and on the radicalisation things where most of it's focused on Islamic radicalisation but there's now an increasing interest and focus on white power radicalisation in Australia. Racism is a critical part of both of those but cyber racism itself until approximately a month and a half ago, was not addressed by any Australian Government policies partly because of the long-term desire by the government to get rid of Section 18C of the Racial Discrimination Act, which meant that when they were setting up, for instance, the (05:35:24) cyber bullying environment, they were very careful to draw a line between that and anything that might interfere with freedom of speech for adults in relationship to questions of race. So it's all been very, very messy but it does produce an interesting particular pattern in the Australian context.

This is not an easy thing to read, what I'm going to try and do here is explain what this table means and, again, in the printed version it'll be much easier to [pursue]. We did a survey of some 2300 online internet users and we asked them about their experiences with racism, their engagement with racism and so on. We also

asked them about their attitudes and values on a whole series of questions to do with race and we ended up with a matrix which basically shows three relationships to racism as a target, as a perpetrator or as a bystander and three broad perspectives on racism; people who are strongly racist, people who are moderately racist and people who are anti-racist. When we press the button and made it all go clunk, clunk, what we discovered was this in simple terms; about 10% of internet users have identified themselves as targets of racism. We know that about 30% of internet users have actually experienced racism and we know from a recent study of young people by the E Safety Commission that came out about a month ago that around 60% of young people have had a recent encounter with racist material online. So about 10% or 11% actually perceive themselves in the general population see themselves as targets. About 3% to 4% of people actually identify themselves as perpetrators, that is are happy to identify themselves as somebody who has produced material that they accept other people might think is racist. That may be an under estimate but it's helpful to get some sense of the general [type]. Bystanders make up about 85% of internet users, so bystanders are very important. But what we also discover is that of that 86% somewhere around two thirds of all bystanders are moderately racist in their views; that is they will see racism, they'll let it go through, they might even find it funny or cheer it on or whatever. So the internet is a fairly buzzy place and if you're a target - and we've done quite a lot of research with targets - it can be an extremely uncomfortable place.

The impact on targets looks something like this. Using a sort of visualisation, analytical visualisation tool, we went through the transcripts of a number of group discussions with people who identified themselves as targets of racism from about half a dozen - maybe more - of different ethnic and racial groups. What they reported was that they felt attacked, they were very angry, they felt that their race was being used to abuse them, that they could find very little way of being protected and that their general response was to retire into spaces which they felt were safe or safer. So effectively one of the most important impacts of racism on the internet is to drive people who are the targets of racism off the general internet into small pockets and colonise the internet as a place where racist values are normalised and that's more or less what we've seen happening over the last 12 months. In order to try and deal with this Google and a large number of these other companies over the last two years have looked at using AI to try and identify and block racist posts, material, all sorts of things on the internet. Starting at about June last year in the United States in 4Chan was a project which was called Operation Google. Operation Google was specifically developed and designed by the people associated with it to knock out the capacity of Google and the other companies to use AI to shut them down and the detail of that is actually really

fascinating, I'd be happy to answer questions on that. The main impact of it was to effectively destroy the capacity - at that time - of Google and Facebook to knock out the use of racist ideas and racist [notions]. I don't know if any of you have tried to use the Twitter or the Facebook or the Google technologies which allow you to try and identify racist material, but in general it produces a hit rate of around 10%, it's still very, very poor and they're finding it very difficult to do.

So what are the problems then for public policy in the last few minutes? In Australia the platforms are uncontrolled and unregulated. It's possible to protect racism from national sovereignty. One of the sites that I've had particular trouble with is beautifully protected, it sits in five different locations across the planet, none of them are in Australia and each of them are able to operate fairly much independently of each other. The actors are anonymous of course. Racism in Australia is extremely easy to do compared to say with Europe, it doesn't mean there's not a lot of racism in Europe but it's much easier to be racist in Australia to a fairly general level than it is in Europe. The platforms are in control of what happens, they can choose whether or not they wish to respond to this stuff, whether you happen to be the Australian, for instance, working on their comments page or Sky News or Facebook or whatever. And one of the things we discovered was that the majority of internet users who - if you like - are soft racists, often don't recognise what they've encountered as racism, so they don't know and part of the issue is if they don't recognise, then they don't react to it and there is no framework or structure or process in the Australian context, unlike for instance in the United States, where anyone is bringing to their attention that this material should be considered racist. So currently targets are expected to act on their own or just they're abandoned. Bystanders are allowed to get on with it more or less, though we know that high status bystanders who are not from the group that's targeted have a great deal of impact on other bystanders and we're now starting to see in New South Wales, through multicultural New South Wales and in Victoria the beginning of social media strategies which are picking up on this research and looking for ways to do it. And racist proponents who ultimately make up maybe 1 in 200 of active internet users, these are the proponents who are actively racist, they are proponents who are unknowingly racist but the ones who are really out there wanting to do as much damage as they can, they produce most of the harassment and their task - or the task they set themselves - is how to create followers and they do that extremely successfully through a number of different approaches.

While there's been an attempt to harness the power of algorithms, to find hate speech, it's still extremely difficult to do and in the Australian context there's very, very little investment at any level by government in actually trying to do this. And

our argument, I guess, is that if governments have a responsibility in the broad for the wellbeing and the social harmony of complex multicultural societies, this is one of the areas where there should be increasing research and increasing focus on how to do these sorts of things more particularly. I think as the chief scientist from across the ditch was suggesting, there is a certain lack of courage in the political class in Australia over these sorts of questions, not simply in relationship to this one. So we're looking now then at how we might go ahead in building resilience, a social program that might counter this. There are moves ahead to create a national body that might push for civil society pressure in these sorts of areas. There are laws again, the ones created in New Zealand seem to be working fairly well and could work fairly well in the Australian context in relationship to the harmful digital communications act that operates there. There are possibilities that we might be able to empower the human rights commission to actually take some autonomous action, though after the belting they've had over the last 12 months they may not be too comfortable at accepting that challenge. And there's also many possibilities in civil society to call out advertisers who allow their brand to be attached to hate sites. There's an organisation in the States called Sleeping Giants and similar sorts of things are starting up in the Australian (05:45:34). There is a need to support the targets of this sort of stuff, to educate and empower bystanders and perhaps to spend some time developing some alternative algorithms that might start to cut into this sort of thing. I live in hope that Australia will sign onto the European Convention on Cyber Crime in relationship to cyber racism. I live in hope though I'll wonder if I'll ever see it of Australia withdrawing its reservation on Article 4 of the United Nations Agreement on the Elimination of Racial Discrimination. Every time Australia turns up in Geneva it's asked to withdraw this reservation, which they put in in 1966, every year every time it's asked it says, "No" and we need to recognise that civil and robust conversation is an important part of our democracy and it can be destroyed if we don't try and protect it.

So this is the book in case you're wondering. It's up on (05:46:38) it's just been released, Cyber Racism and Community Resilience and now I think I have not too many minutes. Oh dear.

Unidentified: A quick and perhaps unnecessary question. When you carried out your survey, did you present a definition for racism? And I ask this because DNA wise the differences in our DNA based on racism - as most people understand it - are much smaller than most of the other person to person differences.

Prof Jakubowicz: The operationalization of the concept of race and racism or the concepts of race and racism we're very careful about. We used a survey that's been used for the last ten years offline and essentially what it is is we use a number of different

definitions of racism, we operationalize those through a series of attitudinal questions, which have been thoroughly tested and reworked over time. We then ask people about those questions and then we can sort of collate their responses to those questions to give us an idea of how racist they are. So we don't treat race as an objective category, we treat racism as a behavioural and attitudinal manifestation.

Unidentified: Have you got a means to where you could get a measure or an index of cyber racism by country? I say that because the OECD is doing a digital government (05:48:34), it's their big project and the biggest thing you can do in the OECD is measure something and make a ranking. So if you could develop a scale - 'cause I'm cheering a working group on or part of a working group on this - it could be inserted in because nothing could deal with this better than the countries being shamed by having an index.

Prof Jakubowicz: It's an interesting question. We tried to work out how you might actually do this because part of the problem is - there are two different ways of answering this. Part of the problem is a lot of people do not recognise what they see as racism, so if you ask them, "Are you racist?" They say, "No." In fact, the people with the most fierce racist values will tend to be the people who most deny being racist. And when this was tested, this was tested by the Australian Government about 15, 16 years ago, maybe more, nearly 20 years ago now, the question was put in terms of basically whether they agree or disagree with a set of [parameters]. Now when the clusters came back from the interviews and they did 7[000] or 8000 of these interviews, the people who were most strongly committed to racism would then go onto identify that racism only applies to behaviour which results in violence. People who were less racist, the more committed, if you like, to egalitarian values, would identify racism as being about a lot more language discrimination et cetera, et cetera. So it is possible to do it because I mean it's more or less what you did. We went in, we had 2000 or so panel respondents, they did well over 100 questions about 35% of them responded that they had an encounter with racism in the previous 6 months, that is they were able to identify the stuff as racist. And we could then look at where they might sit on that scale. We also had the other 70% who said they hadn't encountered racism and we had a picture of what they looked like. So we were able to see what those relationships might look like, so it's not impossible. The same figures as we came up, the same proportion that we came up with came up in a survey in Germany about a year ago around 30% that I think probably now it might be higher (05:51:11) what's going on there all the time. And the US one which looked at both racism and sexism and other forms of hate speech produced around 30%. Now I don't know how accurate that is, we've been asked within our survey (05:51:31) about three or four years ago now, whether we

think this has gone up or not. It's hard to tell, the number of people on the internet has expanded dramatically over that period, the amount of stuff people encounter, the structures of Facebook and so on all changed like Facebook changed its default switch from public to private. One of the effects of that was that people who were in groups were no longer so easily exposed to attack. So there's lots of different questions about all of that stuff but it's quite feasible to do.

Unidentified: Last question.

Emma Johnson: Just a quick question, so down the front, hello, Emma Johnson.

Prof Jakubowicz: Oh there you are, sorry.

Emma Johnson: Hi. A quick question. Cyber sexism appears to be as virulent and growing as well. Can we draw any analogies or parallels between the two types of bias and the resilience and what we can do to combat it?

Prof Jakubowicz: Cyber sexism, particularly misogyny (05:52:39) is experienced at a much higher rate and reported in a much higher rate. Also structures of resilience and the support strategies that women have developed, particularly in the US, are much more active and effective, even though there have been some horrific sort of cases trolling particularly of women activists in areas like online gaming that's being raring up (05:53:11). And there are a number of - if you like - sort of hatcheries for misogyny and racism in the States, which are very well protected, the main one of which is the place called 4Chan, which has an Australian channel, I just went into it yesterday to see how it's doing and it is getting - I think the point where they started suggested - because I'm a Jewish background I should 'fry' and that would be a good way to get rid of me (05:53:39). It's reaching, if you go there that's what you expect. But there's also a sort of dark web component in it where they really work up how to be really nasty and they test each other so they'll try something and if the other boys in the game say, "Oh that's great" then they'll launch. Right? So there's a really intense of area of work and that was the stuff that was coming out during the Trump period and so on in the United States in Russia. There are networks where particularly in the States I think where women who have been [trolled] or they've been particularly harassed are (05:54:27) into that network and ask for support and then there's a lot of support that's active over through those networks. The race one is much more complicated and [difficult]. But the project that - I'm just trying to remember the name of - the project that really set all this up was a thing called Gamer [Gate], which was the one where the programmer, the woman who was a programmer, tried to develop a game around depression and dealing with depression and she just got zapped really badly, including what's called doxing. Do people know about doxing? Doxing is when a

group of trolls decide that making your life hell online is not sufficient and they want to make your life hell online in the real world, so they publish your address, they identify your children's school they go to, in the simplest terms you start getting very large numbers of pizzas turning up on your doorstep, the fire brigade turns up. But more than that the nasties turn up and stand around and it's really not a very nice experience and in the States, of course, particularly after [Ferguson] when the whole issue of racism has become really, really important during the Trump regime, these issues are just growing and growing. And in the US despite the constant reiteration by people like Zuckerberg for instance from Facebook that they want to try and do something about it, they really are in a sense - as Tim Berners-Lee suggests - sort of hoisted on their own petard, they simply can't crack it because they can't program the stuff to stop doing what it's doing or it all falls over and then there's nothing there for them. I mean there was one - I think one comment I saw in the last couple of days that we shouldn't all really get too comfortable with the idea there's only one Facebook because it's quite possible in the future that Facebook will basically explode and self-destruct and then there won't be a Facebook anymore, there'll be dozens, hundreds, millions of these things floating around, which may or may not be a good thing, I don't know.

Unidentified: On that sobering note, thank you.

Unidentified: So we have a very short coffee break, about 10-12 minutes out on the veranda and then we're back.

AFTERNOON TEA BREAK

Prof Griffiths: Okay, I should have known at this stage of the afternoon a 10-minute coffee break was not going to happen. We’re now running 10 minutes late, so we’ll just run the whole schedule 10 minutes late. Quieten down for our next speaker. Introducing Nick Enfield, the Professor of Linguistics at University of Sydney; many of you will have seen in the media or perhaps on the Facebook, I’m sorry to say after what we’ve just heard, many of the events that Nick’s been organising recently. Nick is head of the new interdisciplinary centre in the faculty of Arts and Social Sciences, the Sydney Social Science and Humanities Advanced Research Centre at Sydney, but he’s also currently leading one of our university-wide Research Excellence Initiatives, the Post-Truth Initiative. There have been a large number of very interesting event speakers, and you probably have seen media pieces generated out of that program, so I think nobody better suited to tell us about mind and language in the post-truth era. Nick?

Prof Enfield: Well, thank you very much. I hope you can hear me. It’s a great pleasure to be here. I’m really very pleased to have got the invitation, and I was just mentioning to Simon what a hard act to follow all of the incredible presenters have been so far, so here we go: The Post-Truth Initiative, if I can just sort of direct you to where there’s information about the Initiative, it’s something that we’ve been running through this year at the University of Sydney, and if you’re interested in the sorts of things we’ve been doing, on the website we’ve got a lot of videos, podcasts and publications from the kind of stuff we’ve been doing. So a lot of what I’ll talk about now will come partly out of the discussions that have been had there, but really very much from my own perspective.

As Paul said, I’m a linguist by training, but that takes me quite a lot into cognitive science, and so a lot of what I’m going to be talking about is really going to sort of try to complement a lot of what’s been said, thinking more from the perspective of the individual and some of the issues that were raised much earlier this morning.

I’ll just start with this cartoon which some of you might have seen; “These smug pilots have lost touch with regular passengers like us. Who thinks I should fly the plane?” Now, some of you might have seen this in the US election cycle last time around and it was a really great illustration of part of the problem of what we’ve been talking about today is this sort of notion that somebody might put up their hand and say, “Well, look, I know better than these scientists or these experts or these people who say they can fly planes” and so forth, and the wonderful thing about this particular cartoon is how it really puts its finger on how absurd that position is and we certainly wouldn’t expect the people to be really raising their hands in that kind of a context. But unfortunately, in the real world we see things which are disturbingly kind of similar to that.

So an example that I think resonates with this quite well is from a recent news cycle around tobacco, and this is the news from the Vatican that Pope Francis banned sale of cigarettes at the Vatican on health grounds. And as you can see here, the motive is very simple; it quotes “The Holy See cannot be cooperating with a practice that is clearly harming the people.” And so accordingly the World Health Organisation tweeted their support of this saying, “World Health Organisation welcomes the Vatican’s decision to ban the sale of cigarettes as of next year,” pointing out that tobacco kills millions of people every year and adding this infographic based of course on the science of what we know about tobacco, and we’ve got Lisa here who knows very well, has been in the thick of that.

Now, I want to draw your attention to somebody who retweeted this and made a little comment on it in their tweet, and this is somebody who’s a public figure and has of course a lot of sway, and so this is Nigel Farage writing, “The World Health

Organisation is just another club of ‘clever people’ who want to bully and tell us what to do.” There’s something seriously wrong here of course. This is quite an astonishing little bit of public discourse and it’s I think a classic instance of the post-truth discourse that we’re talking about. So there’s really something going on here and one has to wonder what is the motivation of somebody saying this kind of thing.

We also have seen, some of you may remember from the Brexit campaign there was a lot of coverage obviously of what Farage did, and this is a tweet from when he was in the middle of that campaign. A journalist asks him why has he taken up cigarettes again, and you can see in the picture, I hope, he’s holding a cigarette in his hand. He says, “I think the doctors have got it wrong on smoking.” That’s incredible really because in this case at least he’s putting his money where his mouth is and actually smoking and saying this.

Question: Do you think we should encourage him to smoke?

Prof Enfield: Yeah. So it’s either mad or it’s stupid or it’s something else, and it’s an interesting question as to wonder what that is. So as we just saw, there’s all manner of evidence out there that tobacco smoking is very dangerous and it’s scientifically grounded, and so it is kind of incredible to think why would a person be promoting that as a view to people who would be voting for them and so forth. It says something about the environment in which these kinds of statements are possible.

Now, the reason why we find this crazy is because the reality of the matter is that what the scientists have shown is that this is a harmful practice. And when I talk about the word “reality”, I like to quote the author Phillip K. Dick, his definition of “reality”. So many of you will know his scientific – his science-fiction writing. I think it’s actually a brilliant definition of it really because no matter what Farage says is happening, the smoke is having its effect on his lungs, on his bodily system, and that’s really the essence of what scientific work is all about, is trying to figure out – we started out earlier in the morning with that sort of definition for that first level of reality, if you like, was the real physical world that doesn’t go away.

Of course, scientists have been really trying to insist on this point and we’ve seen the March for Science which happened all around the world, including here. We saw – many of you may not know about this recent Initiative – the Pro-Truth Pledge, which I think’s a very interesting Initiative and one that you may want to go and have a look at. You can go to this website and you can pledge your earnest efforts to share, honour and encourage truth, and I think it’s actually a nice response to some of the sentiments that we’ve heard today, a way that we can actually go out and make a pledge. It’s quite interesting. You pledge to a number

of things, and one of the important parts of it is that you pledge to be accountable to that pledge as well so that you agree for people to hold you to account for having transgressed this pledge at some point. So the idea there is this is really a serious pledge. So here are just a few of the things that you pledge to when you take this pledge: For example, to fact-check information to confirm it’s true before accepting it and sharing it; distinguish between my opinion and the facts; re- evaluate if my information is challenged; retract if I cannot verity it.

These are hard; right? We know that these are the sort of things that are central to science and what we would like to see happening, but it’s actually very difficult to hold ourselves to those standards. And what I want to talk about is some of the reasons why that’s the case, and that is we heard this morning about the kind of research that’s been done on cognitive biases and I want to go into that and talk about some of the reasons that have been established in work on cognition for just why it is so difficult to stick to these kinds of aspects of the pledge.

Here’s an example; this is the Cornsweet illusion. Now, just look at this object. It’s a kind of classic perceptual illusion type of case. So you look at this, you ask people to sort of ignore the shape of what they see and the fact that there’s a landscape and all that kind of thing, and just talk about and just say what is the difference in colour between the top panel and the bottom panel of the object that’s in the middle of the screen.

And oftentimes people will say what they think the colour difference is or the shade difference is, and they’re very surprised to see. What I’m going to do now is obscure everything except just the top and the bottom part of the shape in the middle and you’ll see that the colour is exactly the same, so I’ll just go back. And of course, what’s happening here is that through our perception we’re putting all sorts of interpretation around what we’re seeing and we are compensating for our perception of where the light is coming from, what we believe about the colour of the object itself and so forth. So again, I’ll just obscure that, so just keep your eyes on the top and the bottom. They look quite different in shade, but you see that the shade is actually identical. So it’s a very sort of low level sense in which what we perceive in reality, our beliefs about what we see, don’t correspond necessarily to what’s actually there, and we heard a comment about that with respect to it was also an example about colour.

Let me go into some other kinds of examples which go, let’s say, a little bit higher into our cognition, and I’ll go into this example of – I just want to talk about three different biases and drill down a little bit into what they are.

Question: Are?

Prof Enfield: So yes, are. This is the letter R and it’s a nice example of – it’s just to sort of stand for the example that I’m going to give which is from an experiment on what’s called the availability heuristic. The availability heuristic is something that we use in our thinking which biases the kinds of conclusions we might come to. So if you say to people, “Okay. We’re going to talk about the English vocabulary. And what I’m going to is I’m just going to pull out words randomly from the English vocabulary and I’m just going to ask you to imagine what proportion of these words are going to have the letter R in them, but more specifically is it going to be more likely if I randomly pull out a word that it’s going to begin with the letter R or that’s it’s going to have the letter R as its third letter?” And when you put people under time pressure and you ask them to come up with a prediction which is going to be the more likely one, they say the one that begins with the letter R. So they can quickly sort of think of words that begin with the letter R.

If you give them a fair bit more time or if you really systematically go about testing this question, then you find that actually there’s tonnes of words in English that have the letter R as their third letter and if you go out and systematically count them up, as people have done on various different occasions, you find that actually the ratio is about 2:1 in the direction other than what people think. It’s a very simple example of the explanation being that, from the point of view of the agent who’s trying to solve the problem, it’s simply easier to think of words in terms of what they begin with, rather than in terms of what their third letter is. And so that’s what we call the availability heuristic; there’s something about the way we think that directs us in one direction and makes us confident about conclusions which actually aren’t supported by the empirical data, so that’s a kind of classic reason why do science in the way that we do it. My point here is to say that there’s something about our cognition, this is one of the biases that actually pushes us in that direction. So that’s to say when we apply the scientific method, we’re struggling against ourselves in a certain sense.

Here’s another example: The famous confirmation bias. This goes back to one of the earlier papers by Wason where the confirmation bias was sort of one of its – people have always known about the confirmation bias that what you already believe is easier to accept and it’s obviously a great problem at the moment - but this is just going back to one of the original experiments here and what Wason did was he said, “Okay,” and this is in a psychology experiment, “here’s a triple of numbers 2, 4, 6. So there’s a rule that generated this triplet and I know what it is; I’ve got it written down on this card here. And you can just give me more triplets and I’ll tell you whether they fit the rule or not, and then when you think you know what the rule is, tell me what it is.” That’s all the experiment was. So people would say things like, “Okay, what about 8, 10, 12? Yes, fits the rule. What about

20, 22, 24? Yep, fits the rule. Fine.” So then they would say, “All right. Well, I think I know what the rule is; right? You just add 2 every time.” Okay. Another person might say, “What about 3, 6, 9? Yeah. What about 20, 40, 60? Okay. Well, I know what the rule is.” and they write it down.

Now, what he found was that people very seldom really seriously test the rule. All they’re really doing is saying, “I’ve got a hypothesis and I’m just going to spit out things. I’m looking for confirmation. I’m not actually looking for falsification.” And we know that falsification is what we try to look for in the scientific method, but it doesn’t come naturally to people. So if you started out with something like this, 3, 4, 57, and any other series of ascending numbers, Wason would say, “Yes, that fits,” ‘cause the rule was every number is bigger than the previous number; that’s all it was. And if you cottoned-on to that you’d say, “What about 4, 3, 60? No, that doesn’t fit because there’s a descending-----

Question: That’s not an algorithm.

Prof Enfield: It’s just a rule. And this is the confirmation bias that millions of studies have been done on what that is, but it’s one of the most powerful drivers of some of the problems we’ve been looking at today. I want to give an example of this; there we go. So I want to turn to an example from the big wide world. We’ve seen Trump already and so we’re going to see him again. I want to show you a little video clip that was circulated online just recently. Some of you may have seen it. I’m just going to show you this video clip. Trump is hosting some people at the White House and I just want you to focus on down in the bottom right-hand part of the screen, you’re going to see the hand of a little boy who is in a wheelchair and Trump is just shaking hands with people and leaving the room. So I’ll just play the clip and let you look at it. The kid’s coming into view now in the bottom right. The little kid is there in his wheelchair. There he is; Trump walking past. Okay, Trump; off he goes. Bye everyone. All right.

Now, here’s J.K. Rowling retweeting it and saying, “Trump imitated a disabled reporter. We all saw that in the election cycle. Now he pretends not see a child in a wheelchair, as though frightened he might catch his condition.” Boom, circulated online, you’ve got 12,000-plus retweets, you’ve got 50,000-plus likes. It was a huge thing. I saw it, it just came up on my feed and I went, “Oh, gosh, Trump. What an awful guy.” Then the next day I saw what happened in the moments immediately prior to that scene and I’ll display them to you. This is the beginning of the scene. I’m not a fan of Trump, but this was a really important little moment, I think, in showing, and in the end J.K. Rowling came out and said, “I apologise.” And I think the key thing here was that this was a bit of confirmation bias in action saying, “I hate Donald Trump. I want to do anything I can to show the reasons

why. I see this clip that is not showing the whole truth; it confirms my belief.” Boom, I send it to all my very, very many followers.

So this is obviously a very powerful bias and it’s one that creates problems just of this kind in social life, but also in terms of trying to use this kind of biased mind to do the kind of science, the classical scientific method of trying to falsify, rather than confirm. That’s what makes it difficult and that’s what often gets in the way of it.

Now, I want to talk about one more bias before going on and that is not one that’s really been labelled and, if you look at the literature on cognitive biases, it doesn’t quite come up in the list that has these labels like availability heuristic and confirmation bias, but it’s one that I’m just going to call for the purpose of argument today an identity bias. And that is where the identity of the people involved somehow brings in a whole constellation of beliefs or a whole constellation of ideas that people should be signing up to and it takes them as a kind of package, the point being that if you have one view, then the problem is you’re going to be taken to be having these other kinds of views. Or if you go against one of these views, then you are taken to be going against all of the other because of who you might be.

An example might be if you go to the US and you meet some open carry people, you might be able to make pretty good predictions about a lot of the views that a person has who would subscribe to open carry; right? So here’s some people just buying some chicken with their AK-47 and shotgun and stuff, so that is a view there’s a belief behind that practice, and interestingly it’s associated with a whole range of other beliefs which don’t actually correspond with them in any necessarily coherent kind of way. And yet there’s a promotion of if you have one of these views, like I think open carry of a firearm is a good thing, then you should probably have these other views, like you’re sceptical about climate change or you don’t believe in abortion or whatever the case may be. And this is an issue that is really creating a lot of problems for rational discourse because it’s what’s responsible for shutting down a lot of discussion.

And what I’m referring to is something that’s very highly active in the US where North American campus life these days – some of you might have seen the controversy around this young woman Lindsay Shepherd who is at a university in Toronto and she – I don’t know if you’ve seen the news about Lindsay Shepherd, but she recently was teaching a class at university to undergraduate students and she was talking about the issue of language use and the fact that in English we have these two pronouns, third person singular pronouns he and she, which are coded for male and female gender, and there’s an issue with people who don’t identify with either of those genders. And we know the issues around people

insisting on not having those used and so forth. So there’s been a controversy around that; some quite controversial figures in academia who are saying, “I refuse to follow the rules about avoiding these pronouns,” and so on.

Well, Lindsay Shepherd was giving her students a tutorial about this and she played a YouTube clip which was a debate between someone who was pro the rules around gender-neutral pronouns and someone who was against it. It was a recording of a public debate. And she got hauled across the coals. She did a secret recording of it and revealed this, and you can go and look it up online if you’re interested, but it was a very harsh kind of hauling across the coals by her senior colleagues for traumatising her students essentially. It’s complex in all sorts of ways, but fundamentally that’s what it was; she was exposing her students to hateful ideas and traumatising them. She was told that this was the same as unapologetically playing a video of Adolf Hitler giving a speech. Literally; I kid you not.

So it’s a phenomenal thing that’s going on right now and the fact that she’s a person who identifies as a politically liberal person and she’s now getting all of this hate – I don’t know if she’s been doxed yet, but she’s certainly been majorly hated online as being an alt-right figure. So that’s this identity bias in action and the reason why I raise it is because it stifles debate and it stifles basically ideas, and it stifles the voicing of opinions that wouldn’t fit with otherwise the rest of the package that a person is supposed to embody.

The point of giving these examples is to really say that as individuals we are all sort of subject to all of these cognitive biases and it’s quite an interesting puzzle as to why we have them. There’s a lot of interesting recent literature on it which I won’t go into, but it is fascinating because we have to ask, “Well, why do we have this kind of faulty cognition? Why do we have these bad ways of thinking?” You were saying earlier that this was a pessimistic view of human thought and actually we’re capable of other ways of thinking. There are interesting questions around whether actually maybe these so-called biased forms of thinking do have advantages maybe in our evolutionary context, but probably not in today’s context.

Whatever the case, I think what’s fascinating about our cognition is that we are able to focus on our own biases and sort of outsmart them, if you like. So here’s another website I just want to draw your attention to - it’s recently been set up at Harvard - called Outsmarting Human Minds and there’s a whole range of projects that are happening within there. But basically, they’re essentially promoting this view, that others like Daniel Kahneman and Gerd Gigerenzer and other people have been promoting, which is that with effort we can see the operation of these

problematic biases in our own cognition and we can overcome them, we can outsmart them, we can do better basically.

And so I subscribe to that view and what I want to do in the last little bit of time that I have is to say something about the role of language in all of this. Now, language is my area of specialisation and I think that language plays a very important role in all of this. It was mentioned earlier on; His Excellency said something about language at the very beginning and so I want to now turn to language in the last part. The sort of slogan that I’ve been formulating when I’ve been talking about this recently is the idea that facts have to go through language to get to us.

Now, that’s a strong way of putting it, but in fact it’s not really inaccurate, I don’t think. Most of the time when we’re talking about our evidence when we’re trying to support our arguments with evidence, what we’re essentially doing is taking that evidence and formulating it into words and producing assertions in an interactive kind of setting. And so the essence of the argument that I want to make is that through doing that we inadvertently bring in a whole bunch of other aspects of what language is and does when all we’re really trying to do is to make an assertion to put forward a bit of data, but we drag in this other stuff.

So what is it that we drag in? Well, language does a lot of things, but here are some of the three very basic things that it does: Let’s just take a simple sentence like “It’s sunny outside.” Now, with language, if I make these sounds, what I’m doing is I’m coding a proposition. If I say, “It’s sunny outside,” it’s sort of not really, but imagine it’s sunny outside. I’m coding a proposition and I’m making a statement about the world which you could test to be true or false, et cetera. But what I’m also doing through saying it is, I’m telling you certain things about myself; for example, my accent tells you where I’m probably from, other aspects of how I speak tell you something about who I am and perhaps who I am to you. And another fundamental thing about what I’m doing is you will never just hear that as a completely independent, standalone, disembodied statement; it’s always going to be perceived by you as a reason for action, and it’s a little bit like that illusion that I showed you earlier on. You’re always going to project structure onto that and you’re going to compensate. What you’re going to do is never just take it at its pure face value. If I say to you, “It’s sunny outside,” you’re going to assume that I have a reason for saying it. So it’s actually quite a powerful thing that we project onto language.

Why is he saying that? Maybe he’s suggesting we should go out and enjoy the sun. Maybe he’s saying it’s still too hot and we can’t go out, whatever is relevant. And in fact if you study linguistics you learn about these principles of relevance, the central idea being that the listener to the message assumes it’s relevant and

actually looks for the message that underlies it. So we’re very much tuned into not taking words particularly literally, but looking for the message that’s behind them, and that’s part of the reason why, when we’re offering evidence, it’s never just taken without being taken as a reason for something.

So I’ll give you a few examples of a very good one in the post-truth sort of - it’s going back a little way, not that far – it’s from early this century. So here’s Colin Powell holding up a mock vial of anthrax in the United Nations General Assembly during the time when they were worried about chemical weapons and weapons of mass destruction in Saddam Hussein’s Iraq. And they had weapons’ inspectors across the country looking for weapons of mass destruction, reporting that they didn’t find them, reporting that they didn’t find them. Well, the Bush administration made this assertion that our intelligence says there are weapons of mass destruction in Iraq.

So you don’t just say that; right? It’s a reason for action and of course action followed, so we saw the shock and awe, and we saw everything that followed from it in Iraq. And the reasons now we know that that was false; there weren’t weapons of mass destruction. Many people knew it was false or said that it was false at the time. And then of course after the fact we can ask ourselves what other reasons might there have been. Well, some people said there’s some element of revenge or just simply duty to respond to what happened on 9/11; other people suspect that reasons behind it would be to do with wanting to control the oil or whatever it is, there are many other possible reasons that this might have been done.

The point here is that you don’t act without reasons, so sometimes you’ve got to produce them and pull them from somewhere, and this is something that really infects a lot of what we do of course and it drives things like conspiracy theories. Something gets stated, for example, in the news and people look for reasons behind why they would be stating it.

Just one last example then would be the example of this little boy. Do any of you recognise him? No reason why you should, but this is Noah Pozner who was slaughtered along with 20-plus other kids at the Sandy Hook incident in the US. And this is his mother, Veronique Pozner, who is giving testimony around what happened. And if you have read up at all on conspiracy theories, it’s a very sad story so I don’t really recommend you go and look it up, but if you did, you’d find that there are a number of people, there’s a whole Sandy Hook hoax movement. And so this woman, among many others actually who are parents of the children who were slaughtered, they have literally had death threats. To this day they’re

getting harangued and harassed, and people have gone to jail for threatening their lives.

But the reason why I’m bringing this up is to just draw your attention to this. This is from some of the hoaxer kind of stuff online here where they’re trying to build the case that this woman is actually not who she says she is, that her son never existed, she’s not who she said she is, and here it says that there’s some doubt about who she really is, that actually she’s from an embassy or something like that, but here the key thinking long-time gun grabber; okay? And so there’s a lot packed in behind that little bit of language there, but the key thing behind it is that it says, “Here’s the reason why she’s saying this stuff that her son was killed. The only reason is because she wants to take away our guns.” So any statement that’s getting made is being interpreted as a reason for action and that’s why I want to say that facts have to go through language to get to us.

The last little bit of what that actually means that I want to mention has to do with the words that we choose when we report those facts. Some of you might have noticed Freddie Gray. The problem is we don’t remember who all these people are. Young, black guys getting killed by police; it’s just another one. Freddie Gray made the news. He was killed in police custody in Baltimore and there was widespread unrest, and one thing that was interesting - so some of my colleagues in linguistics studied how this was reported in the news - so if you looked at the right-leaning news, the imagery you see there is much more about people throwing Molotov cocktails, and the left-leaning news was more about the community disruption and so forth. But if we look at the language that was used, this is a comparison of how much of the time the word “riot” was used versus the word “protest” was used. So the red is the word “riot”; the blue is the word “protest”. So you see Fox is using the word “riot” to cast their behaviour as very different from the way that let’s say – CNN are still using the word, but you can see what’s going on here in terms of the language being used. And here from print media very marked differences between these three outlets and the nation, which is studiously avoiding the word “riot”.

And so it’s very much there’s this very strong correlation with how are we portraying these groups of people. Some people respond to this by saying, “Oh, well, it really makes no difference; we’re just reporting the facts. Here’s the stuff that happened.” But actually, we know that these things do make a difference and again we can go back to the cognitive psychological work that’s been done in the past.

So this is from a famous study that was done by Elizabeth Loftus, who some of you may know. She’s a memory expert and she’s done decades of incredible

work on false memories and the implanting of memories and so forth. And this is from one of her first experiments where she shows a little film of these two cars. You see what happens. She plays the exact same video to two different groups of people and she asked them to watch the video very carefully, try to remember everything they can about it. Later when she asks them a question she uses two different words. In one case she says, “How fast were the cars going when they hit each other?” And the other group she says, “How fast were the cars going when they smashed into each other?” And what she gets is that the smashed into group consistently estimate that they saw that the cars were going faster than they actually were.

Much work has been done on this effect and what it shows is that the words we use can actually overshadow our beliefs, our memories and, in a sense, our truth, but I’ll just say our beliefs about what actually happened, underlining the importance of objective measures of these things. So because of all of these reasons my way of thinking about critical thinking around truth is that when we see a statement being made, what we want to do is ask what motivates a person to make a statement. They’re not just making a statement, but they’re doing it for a reason. How do they package that statement? What words are they using and what words could they have used? And then of course going back to the point about reality being there whether we like it or not, never doubting that there is this sort of unanchored kind of relativism behind all that, that there is a brute reality that provides sound reasons for action.

So the last thing I want to say is just to finish on an optimistic coda, and this came up in a question that came to me in a meeting a little bit like this not long ago where we were talking about fact checking and this kind of thing and the problem of all of this misinformation online, and somebody said, “Well, who’s going to be the gatekeepers?” And I am convinced by those people who say even really great fact checking is not going to stop all of what’s going on online, but what we can do is do what J.K. Rowling didn’t do, and that is follow the Pro-Truth Pledge that we as individuals can be the gatekeepers – and obviously this is easier said than done – but the reality is if we did actually apply those principles of fact checking before we click “Like” or before we share or whatever, then we would be able to avoid or lessen the effects of many of these problems. I’m sure that’s all I have time for. Thank you very much.

Prof Griffiths: Thanks, Nick. We will take one or two questions. It’ll have to be quick ‘cause we’re really starting to run further and further behind time. So just from up the back?

Question: Hi, thanks for that. It just brought to mind – I’m an engineer and I’m from [interstate]. I work on a road system project in the Philippines and it was voted – it was an Australian aid project actually, quite a large project – it was voted by Columbia University as the best aid project they’d seen in the world.

Prof Enfield: Congratulations.

Question: And it was to bring farmers into a situation where they could sell their goods and everything was very positive. Not too long afterwards a Walkley Award-winning journalist in Australia wrote the project up so it was the worst project you’d ever seen. It was military infrastructure to subdue the rural community. Is that a post- truth sort of story?

Prof Enfield: Well, I couldn’t possibly answer that, but of course the question goes to what’s the evidence to support the claims? I’ve been involved myself; I work in Laos and I work in rural areas. And I’ve been involved myself in work in an area where a huge hydroelectric dam has been built, and there are two very different stories about whether this dam project was a success or not. I’m afraid with those very big projects there are so many factors and so many stakeholders that I don’t even think it’s wise to try to cast them in those very simple terms, not to say that you’re not deserving of the prize or the award that you got – that’s fantastic – but I think the point is that those kinds of projects and those kinds of social phenomena are so complex that you’d really have to break them down if you seriously wanted to test the truth of those kinds of claims, I think, but they’d be good examples of language optional sort of castings of the facts, yeah.

Prof Griffiths: Take one more question over here and then we’ll be sadly (indistinct) of a forest of hands.

Question: Hillary Clinton in What Happened calls Fox News and Richard Murdoch a mega foe to the lives and post-truth behaviour of Trump, and I was listening to your talk and I really impressed that I hadn’t quite seen how much the language the Murdoch papers give. I go to the Coolongolook Café to get a coffee on the road and I read the Daily Mail, is it, headlines and they’re full of abuse, and I wonder to what extent a lot of our problems are actually related to the almost totalitarian ownership of newspaper media today, and if that language was defused and that politicisation of nationalism and hard-right politics was taken away from us, whether we wouldn’t have a more truthful world?

Prof Enfield: Well, it’s an interesting question. I think that it’s clearly the case that there’s very heavy bias in many media outlets, and some people argue that the free internet and the information age is actually a great antidote to this. The problem of course is who dominates. And from my point of view the only thing I can say is that the

good news from the linguistics side is that you can analyse that and it becomes an empirical question, and you can say, “Here’s the same story being reported in this, this, this, this and this outlet,” and take that scientific approach of saying, “I’m not going to tell you which one you should read, but I will tell you here are the words that that outlet uses and here are the words that this outlet uses,” and then these things become less subliminal and more in the centre of our awareness.

Prof Griffiths: Thank you. Well, our last speaker of the day is somebody whose work you’ll all be familiar with, Mr Ross Gittins. And I can’t really put it much better than in your program; Ross Gittins has been the Herald’s economic editor since 1978 and is one of Australia’s most influential public commentators. He has a reputation, it says – and this is an understatement – for helping readers make sense of complex economic issues. Certainly, a bit of (indistinct) I look forward to many days in the Herald. So thank you very much for agreeing to speak and we look forward (indistinct).

Mr Gittins: Well, thank you. We’ve had a lot of interesting and quite varied contributions today on the topic of truth, rationality and post-truth which I know from what people have said to me during the breaks you’ve really enjoyed and got a lot out of; I have, too. Opening the proceedings was Don Hector who asked us what had happened to reason, then told us that the post-modernists and relativists were in the ascendency, rejecting established sources of reason and accepting that belief should have equal sway with fact, thereby putting an open, free society in great danger. I should say that in my summary of what people said I am trying to fit it into this overall theme of truth, rationality and post-truth.

Simon Chapman, hero of the long-running battle against the tobacco companies to get restrictions on smoking and harm it does, told us about his latest crusade against the unfounded fear of wind turbines. Here, rather than battling powerful industrial interests, he’d been battling uninformed individuals whose fears have been taken far too seriously by a conservative government containing many climate change deniers.

James Wilsdon told us about the Brexit experience with its many fanciful claims and rejection of evidence and the views of experts. He quoted the leading Tory Brexiteer, Michael Gove’s - what’s been described as a spinechilling line - “People in this country have had enough of experts.” Well, I must say that if I was a Brexiteer and all the evidence was on the other side, that’s what I’d say too if I was a politician. And it doesn’t surprise me to hear a politician in the heat of battle making that kind of silly statement and it doesn’t convince me that experts are on the way out.

Emma Johnston said we were in a post-truth era of virulent attacks on science and online trolls in which the truth can be virtually impossible to distinguish from the fake news. As a profession, scientists need to shore up their standing in the community, asserting the importance of their work in contributing to evidence- informed decision making. They need to help the public recognise credible scientific knowledge within the new information free-for-all. They need to change the culture that discourages scientists from speaking out. Genuine partnerships with communities, businesses and industries could go a long way to re- establishing trust in science.

Lisa Bero from Pharmacy took a different, more professionally self-critical tack, reminding us of the way conflicts of interest arising from financial gain can reduce the influence of research evidence in policymaking, but then asked whether we should be paying more attention – this was the bit about the horse and the cart – to the way conflicts of interest can bias the design, methods, conduct, interpretation and publication of research. We need to make our research trustworthy. She concluded, and I concluded, that she’d provided a lot of ordinary people with evidence to say, “Just because a scientist says it, do I have to believe it? Where did this scientist get his money from? What personal biases might the scientist have inserted into his proof?”

Then we come to Peter Gluckman, I think, who spoke about the role of evidence and expertise in policymaking. I thought it was a very realistic and very enlightening presentation for someone who chats to the New Zealand Prime Minister a lot of the time and has some real-world experience that I think many of us haven’t had, and putting this whole wonderful notion of evidence-based decision making into a much more realistic context. Science is not the only source of evidence. Scientists can only be advisers. They’ll only be part of the advice. The advice of scientists – and of course there’s also advice of economists; I’m not sure whether you count them as scientists, but I don’t think I do either – but actually their advice and their evidence probably gets taken more seriously than the advice of real scientists. But then of course politicians, in making their decisions, will take account of public opinion. Why wouldn’t they? Isn’t that why we have a democracy? They’ll be influenced by ideology, which is how we’ve structured the traditional two-party system. They will be open to compromise. Politicians stay in government by putting together coalitions of interest, shifting coalitions of interests.

Then I think we get to Andrew. Andrew Jakubowicz explained how the internet facilitates the spread of racism and reduces trust, damaging the functioning of multicultural societies, and then he discussed various ways to reduce the problem. I have to say that I think we’re describing a problem to which there will be a

reaction from advertisers who don’t want their advertisements appearing next to racist things, from governments, from the companies themselves who fear that, if they don’t lift their game, governments will come in with a heavy hand. I can’t believe that the present state of affairs can just roll on without anybody thinking that it needs to change.

And then we come to Nick Enfield. He argued it was not remotely in the community’s interest to dismiss expert testimony from scientists, in the process diminishing our trust in them in this post-truth era where we feel free to substitute alternative facts, rather than simply criticising things anonymous people say on the social media. He singled out Tony Abbott’s assertion that coal is good for humanity, which he seemed to think was a self-evidently wrong, wrong, wrong thought, when the overwhelming majority of people, who are professionally qualified to evaluate scientific evidence on the matter, know otherwise.

Well, they know it now, but they didn’t know it 200 years ago when we decided to use coal to power the Industrial Revolution. And when we deny that coal has been good for the community, what are we saying? Are we saying, “Isn’t it a terrible thing that 200 years ago we started using fossil fuels? Wouldn’t we be much better off today if we hadn’t?” Or are we saying just looking forward we now understand how much damage coal does, not just to the emissions, but also to people who live near coal-fired power stations. Fine; surely we should be doing something about that. But in disparaging my good friend Tony Abbott’s remarks about the virtues of coal, are you saying that we should stop burning it as of this moment now that we know how much damage it does? Are we saying that there are no benefits along with the costs? Economists believe everything has costs and benefits and you have to weigh them up. They don’t believe that coal has only ever had costs and it’s all a terrible thing.

Anyway, then Nick made a very pertinent contribution, reminding us, as did Don Hector, of the findings of the psychologist Daniel Kahneman who won the Nobel Memorial Prize in Economics for his role as the founder of behavioural economics. Kahneman demonstrated that most of the time humans are unthinking, emotion- driven, non-rational animals notorious for their poor reasoning, even though those humans can at times reach the heights of rational reasoning we see our scientists attaining in, for instance, Newtonian Physics and Einstein’s Theory of Relativity. That’s the strange thing about humans; what Kahneman called type 1 and type 2 thinking. Type 1 is all instinctive and driven by emotions, but we can also do type 2 where we force ourselves to think very clearly and rationally, and sometimes some of the more brilliant among us come up with amazing discoveries.

So what are my thoughts about all this? Well, sorry, but the journalistic scepticism, which is my substitute for scientific scepticism, leaves me unconvinced by much of what was said today, even though of course there was a lot I agreed with. As a journo would put it, I think it’s a beat-up. I can understand how frustrating scientists must find it to discover there are uninformed people who simply reject the scientific evidence of global warming and are impervious to counter-argument. Indeed, psychologists tell us that the more dire the scientist’s warning about how little time we have left to prevent hugely damaging climate change, the more the deniers are reinforced in their denial. That’s not exactly how the rational thinking model is meant to work, but it is the way humans work, and you have to go to the people who are right on the edge of science, the most respectable end of social science, the psychologists, to get them to explain a few things to us.

I can understand how shocking many scientists find it to be told to their face that they’ve not believed, they’re not telling the truth, but are making up crises to get more research funding, which of course as you know is a constant refrain of the climate change deniers. But I don’t find this evidence-defying, unreasonable, irrational behaviour, this refusal to use one’s brain all that surprising. I’ve lived with it every week of the 40 years that I’ve been commentating on economics. You think they believe all the things that I tried to tell them? You think they don’t just reject it out of hand if it suits them, or if what I’m saying doesn’t suit them?

It strikes me that hard scientists know a lot about how the physical world works, but not a lot about how humans work, nor do they seem to know much about how the political game is played, although we had some contributions today that I thought ought to have been fairly enlightening for a scientific audience. Did you know, for instance, that people are given a vote regardless of how uneducated they are, how unthinking they are, how willing they are to give free rein to their instant emotional reactions to developments and their refusal to use their grey matter for anything other than enhancing their encyclopaedic knowledge of cricket scores and reality television? Unfortunately, it’s a free country and if they want to do that, they’re allowed to, and they get the vote.

Did you know that humans are prone to tribal behaviour? That politicians, for their own venal reasons, have turned climate change into a tribal issue where your tribe believes in it, but my tribe doesn’t? That I can close my mind to your incomprehensible arguments and simply refuse to accept that your professed expertise means you know the truth but I don’t, and I can do that for no reason other than that me and my tribe don’t believe that shit.

I’m not convinced we live in a post-truth era as we’ve heard so many times today. The Oxford Dictionary defines “post-truth” as circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief. And this is something new, is it? We used to live in a world where rational analysis reigned supreme, where no-one ever used facts selectively, no- one quoted a fact that needed checking and all the policy decisions politicians made were based strictly on evidence, where anything said by someone wearing a lab coat was accepted without question, but then along came the internet and social media and suddenly all respect for the truth and facts, and evidence and experts went out the door. Really?

I think we’ve always lived in a world where a lot of people are pretty dumb, where many choose not to use their brains for the purposes that scientists think they should, where they much prefer to give their emotions free rein, where anti- intellectualism is common. To me this isn’t something new; it’s a description of the human condition. To attribute it to the ascendency of postmodernist intellectualising, rather than the prevalence of mug punters, is to engage in intellectual delusion. I don’t think the punters needed the postmodernists to tell them that they didn’t have to believe it if they didn’t want to. What’s changed is that the internet and social media have given the anti-intellectuals and tribalists and racists a microphone through which to broadcast. One effect of this is to make our tribe far more aware of the terrible things other tribes have always thought and said about us while out of our hearing. Now, we’re aware of what they’re saying about us.

This does mean that there’s now a lot more scope for people to be shocked and hurt by the new knowledge of the terrible things other people think and say about them. The internet and social media have also made it far easier for disparate members of particular tribes, including the science tribe, to find each other and engage in orgies of confirmation bias to rev each other up. As has been observed today, social media has facilitated the development of many and varied echo chambers. What’s less obvious to me is how much real difference this upsurge in preaching to the choir makes. It probably does contribute to the other forces making our politics and our community more polarising.

Many speakers today have implied there’s been a big increase in the community’s anti-intellectual attitudes and behaviour. This may or may not be true. Ironically, I don’t remember anyone producing any hard-statistical evidence that there has been such an increase. One alternative explanation for the trends we think we see and attribute to the digital revolution, but which hardly rated a mention today, is the longstanding decline in standards of political behaviour by the mainstream

parties which is prompting increasing number of voters to flirt with various strains of populism.

I think I detected a fair bit of tribal rah-rah thinking by the science tribe in what was said today. Science and scientists have been disrespected as never before and we must lift our game and fight back. I suspect I heard echoes of nostalgia for the good old days when the pronouncements of scientists were accepted with respect and without question, much as people in olden times wanted their priests to tell them what to do and not to do to live a moral life.

Well, let me remind you that our population is better educated than it’s ever been, and one of the things they try to teach you at uni is to think critically about the pronouncements authority figures make, even those who tell you they’re experts. Don’t just nod when your doctor tells you something; put the doctor through their paces. I think you’re getting a touch of your own medicine, the stuff that universities have been inculpating for generation after generation; don’t just accept the experts, question them, query them, make up your own mind whether or not you’re persuaded by their evidence.

Okay, I’ve annoyed you enough, so thank you very much.

Prof Griffiths: Are you okay to take a few questions on that?

Mr Gittins: Yeah.

Prof Griffiths: Let’s go to the floor for five or 10 minutes before we have our final thanks to the speakers and so forth. Who’d like to go? You’re all terrified. I never thought I’d discover that post-truth and (indistinct) are the same idea. That’s good.

Question: That’s a comment rather than a question, but you mentioned Tony Abbott talking about the coal and so on. Time is short; I’ll mention the second thing you talked about. It’s the question of Nigel Farage and smoking, and how he comes up with this weird statement. When the correlation between smoking and lung cancer was established in England in the 50s, this data was presented to the Chief Statistician in England, a very famous statistician called Fisher, and he was asked to interpret the data. It didn’t take him very long. He said the answer’s very simple; lung cancer causes smoking. The irritation and pain and so on of the lung cancer causes a desire and a need for some sort of easing of it and you find that in smoking. So that was a rational explanation.

Mr Gittins: One of the findings of behavioural economics is that everybody has a self-control problem and each of us has a self-control problem, maybe more than one, but your control problem might be different to mine. You can see what mine is. I think Nigel Farage said that, not because he genuinely believed it, but because he was

doing something that he knew was injurious to his health, but wanted to do it anyway, had been caught doing it and so he gave this flip answer to which we shouldn’t attach too much weight and not take it as proof that somehow we live in a post-expert’s age. I don’t think we do.

As for evidence, evidence and public policy decision making is a pretty new idea. Evidence-based decision making has, as you know, been pioneered by the doctors who are making decisions about what pill to give you on a host of bases that were not evidence-based, what my old professor at uni told me you should do, what my mates all do, and so the medical profession to this day is struggling to make medicine evidence-based. They are way ahead of the rest of us, but the fact is that one of the side benefits of the digital revolution is that it’s much easier and cheaper to collect evidence. That’s what the whole big data thing is about, but also just the fact that you have computers and that as businesses do their business, it’s not a great imposition on them to say, “Well, while you’ve done all that, just ask the computer to fill out this form for us ‘cause we need the data.” Evidence-based decision making is just starting and has got a long way to go, and that’s because it’s much easier and cheaper to collect evidence.

Question: Mine might be the last question of the day, Ross. Is the beat-up, as you say, from a journalist’s perspective killing off traditional media?

Mr Gittins: Well, yeah, in short the obvious answer is yes. The less obvious answer is if you want to be able to read stuff that you can trust; guess what? You have to pay for it. And unless the serious media can persuade enough people to pay for their services, those services will decline and you’ll be left with everything that’s free, but you can’t be sure whether to believe it.

Prof Griffiths: I think that’s a very good note. It’s 4.30. So we’ll have-----

Mr Gittins: That was my commercial message, by the way.

Prof Hibbert: Oh, dear, we were doing so well, weren’t we, till he stood up? Could I just say I think that was a fantastic way of finishing it off because it did bring us all back to earth and my job is not now to have yet another commentary on what we’ve heard today, but just to say a rather large number of thanks. Around, as I’ve been talking to people, they asked, “Well, how did you ever come up with us? Who chose all those speakers?” Well, the answer is Don Hector, Brian Spies, Richard Sheldrake, Heather Goodall, Paul Griffiths, Des Griffin, Max Crossley, Ian Wilkinson and me. Don Hector came up with nearly the title; it was somewhat amended by Paul Griffiths, so you’ve got to thank Don for the whole concept, perhaps, and I think it was Des Griffin and me who suggested Ross Gittins, so I apologise for that, too.

So many thanks. We’re at the end of the day just about on time, which is a first for these forums, so I’d like to thank the speakers and the audience for actually participating. And I think it worked; it certainly worked for me sitting where I was. And I think this year will turn out to have been one of the best years that we’ve had in these fora. Things did seem to come together and again at the end I thought that all worked and we all now have a bit of an idea of what we were on about.

Could I just ask Bob Marks to stand up. He’s our editor of our journal and proceedings. So for all of the speakers, this is the man that’s going to be harassing you from now until whenever you can hand in the hard copy of your talk. So there will be an issue of the journal and proceedings of the Royal Society of New South Wales which will contain the full papers, and I’m very pleased to have heard several of our speakers refer to their full papers as though they had indeed started writing them.

I have to thank again the Office of the Chief Scientist and Engineer because they’ve given us very good support; Mary O’Kane in particular. And as I say, we’re all sorry to see her going at the end of the year. In terms of just putting it all together, this doesn’t happen, as it were, by magic, as many of us don’t believe in magic, so I’d just like to thank the AV people, the caterers from Government House and all the staff of Government House who have liaised with myself and Ian Wilkinson and so on over the last couple of months. You couldn’t get nicer people and they really have pandered to us when we’ve changed things at the last minute and so on, so I’d really like to thank the people at Government House and also TAMS.

Which of course just leaves me one ever-so-big thank you to end the day and that’s to His Excellency for having us here in his home for three years running. I think it works, I hope you’ve been, shall we say, entertained, educated? Have you learnt something? But really we thank you very much because this is a really unique experience and I can’t see it happening anywhere else in the world, let alone Australia. We’ve all got a lot out of the day and, once again, thank you very much and thank you for coming.

END OF TRANSCRIPT