AAS | Events | 14-15 February 2008

Janet Salisbury Janet Salisbury obtained a PhD at the Paterson Laboratories, a cancer research institute in Manchester, UK. She continued to research for a further 10 years in London, Melbourne and Canberra before the pressures of family life prompted a change. Over the past 16 years she has built up her own company, Biotext – a Canberra-based science information consultancy that currently employs nine people. Biotext supports government and academic agencies through literature research, writing and editing of science information for practitioners, policy makers and the general public. In recent years her work has covered topics such as evidence-based medicine, infectious diseases (human and animal), use of antibiotics, GMOs, cloning and embryo research, air and water quality, and sustainable development. Janet has had involvement as technical writer for the NHMRC and with the Legislation Review Committee (Lockhart Report on human cloning and embryonic stem cell research).

Thank you very much to the Academy of Science for inviting me to come and speak here today. It is a great pleasure to be able to speak at a forum such as this.

I was given the title Addressing specialised audience, students and the public. I decided that I might adapt the title slightly when I came to write my talk, so I have renamed this rather broad topic that I have been given, Getting your message across.

As has been mentioned, I have a company, called Biotext. I have been doing this sort of work for about 17 years now. Before that I was a research scientist. I started out life doing a PhD in a cancer research institute in the UK, and then I did postdoctoral research for six years in London, then worked in Melbourne at the Department of Medicine and then at the John Curtin School [of Medical Research]. But finally my big-picture side got the better of me, and I found that I was a bit more of a big-picture person and I needed to get out and find out about how disciplines overlapped and what bigger picture there was out there about all this science. Also, at the same time I did have a young family, and a career as a researcher is quite a difficult thing, as I am sure the women among the audience, especially, would know.

My company does some literature research; we write original material sometimes; we do a lot of work with technical expert committees, writing up reports about quite major national issues from time to time, where we are the technical writers working with the committee, putting together the information and writing it; we also write a number of other summary documents, public discussion documents, just general science information, brochures, documents, reports, anything that you can think of – and we take it, as I say, sometimes right from the research stage through to the end publication, and any part of that continuum along the way.

Today I want you to think of three different scenarios that are barriers to getting the message across. Two of them are very standard things that people talk about in science communication, and one of them is a bit of a bee I have in my own bonnet.

In the first scenario, you make a great discovery, you write a paper about it, but no-one takes any notice. In the second scenario, you are developing a new technology which you think is great, it is going to help save lives. You speak at a public meeting about it, but the people there are very hostile towards you. You wonder why. In the third scenario, your research studies provide some great new evidence for a scientific issue. You tell the newspapers about it, but the story sparks off a lot of scientific argument that you weren't expecting, and you wonder why.

Barrier 1 is that no-one takes any notice. Bob Williamson touched on this this morning when he said that he had, the other day, heard a scientist talking on the radio – they had their 'grab' on the radio – and he could not understand a word they were saying. I am more involved in the written word than the spoken word, and this speaking is not my normal medium so excuse me if I don't sound quite natural in this environment.

I am just going to show you two things which I have taken from the John Kirkman Communication Consultancy. John Kirkman runs excellent courses for people who write in the health sciences and medical sciences in the UK and in Europe. If ever you are in Europe, and especially if you are in the UK, and you are interested in this area of research, I suggest you get yourself signed up to a John Kirkman workshop. They are really excellent.

This is 'Brown's version'. John Kirkman has been doing a little bit of his own research in polling using this version.

In the first experiment of the series using mice, it was discovered that total removal of adrenal glands effects reduction of aggressiveness and that aggressiveness in adrenalectomised mice is restorable to the level of intact mice by treatment with corticosterone. These results point to the indispensability of the adrenals for the full expression of aggression.

Then the next one, which is called 'Smith's version', is an attempt to write the same information in significantly plainer English than Brown's version.

In the first experiment with mice, we found that the mice became less aggressive when we removed their adrenal glands. When we treated the mice with corticosterone after we had removed the glands, they became as aggressive as the intact animals. These results indicate that the adrenal glands may control aggressive behaviour in mice.

John Kirkman has been polling people for some time, and I would like a show of hands here on each of the following questions: Which version is easiest to understand? Who thinks that Brown's version is easiest to understand, and who thinks that Smith's version is? We seem to be fairly unanimous on that one.

Which version is more precise? Is it Brown's version – one or two people maybe think Brown's version is more precise – or Smith's? And who is not sure between those two?

Which writer inspires more confidence in you? Who thinks Brown's version inspires more confidence? One or two do. Who thinks Smith's version does? Most people here think so.

This is the clincher. Which style is more appropriate for scientific texts? Is it Brown's version? Yes, a number of people think so. Or is it Smith's version? And I see that some people don't know.

What John Kirkman has found from doing this sort of polling over the years is that although most people say that Smith's version is easiest to understand, there is a little bit more of a split about which version is more precise. Normally most people do think, even, that Smith's version is more precise, but there is always a good percentage of people that think that Brown's version is more precise. Which writer inspires more confidence in you? Again most people say Smith's version, but some people say Brown's version. But the question of which style is more appropriate for scientific texts is usually split fifty-fifty. So, although most people can see that a plain English version is easier to understand, there is still some resistance to the idea that that version is more appropriate for scientific texts, and people are worried about the loss of precision.

My personal feeling is that people who are even at the very pinnacle of their career and are highly expert in a field, and who certainly would be able to understand Brown's version, are still very busy people. If you write something in very plain English and have as few words as you can have – or less words, anyway, or more easily understand words – then it helps those people as well as the people who are not experts in the field. So I am a huge fan of Smith's version. I think that if people were to write their scientific papers in a very plain style, then they would be more likely to grab the attention of their audience.

We probably could have a debate afterwards about those issues of loss of precision and whether something is more appropriate in a scientific text, but I think the advantages probably do greatly outweigh the disadvantages.

I am going to talk very quickly, because I don't have a lot of time, about five common problems of science writing: use of many complex terms, imprecise use of jargon, excessive premodification, overuse of the passive voice, and overuse of what we call weak verbs and abstract nouns. Usually, all these things combine together to make scientific texts extremely dense.

I am going to go through this quite quickly, because this would be the subject of an all-day workshop, with lots of exercises and things to do, if you could get yourself out to one of those, but I am just giving you a flavour of this.

Instead of saying the 'early puerperium', which is a word that I can never even get my tongue around, it is quite good to say 'immediately after childbirth'. Then everyone knows what you are talking about. 'Splenomegaly' denotes an enlarged spleen; 'hepatectomised' refers to an animal that has had its liver removed. And there are many other examples.

I should apologise for my bias towards the life sciences, because I have realised today that there are a lot of astronomers and chemists in the audience. On the other hand, you have just had a large dose of astronomy from the previous speaker, so I don't feel too bad about getting in with the medical stuff here.

We come now to jargon. Some jargon words have been dealt with very well by people like Don Watson, who was the speechwriter for Paul Keating. He has written a couple of books in the last couple of years about the misuse of language in the bureaucracy and in public life in general, and about how we have lost the ability to craft really good speeches and prose as a lot of jargon, fashionable words, has come into the language. In fact, such words are not terribly helpful, because sometimes you are not quite sure what they mean.

For instance, does 'enhance', which is a very commonly used term in all sorts of areas, actually mean 'increased' or 'improved'? When you are talking about scientific precision, which some people were clearly worried about when we talked about Smith and Brown, then it is important to know whether you mean 'increased' or 'improved'.

Does 'parameters' mean 'variables', 'values' or 'upper and lower limits'? Again presumably it is quite important to know that, and so perhaps 'parameters' is not a very good word.

'Inhibit' is the other one I am going to mention here. Does it mean 'stop' or 'slow down'? You go into the Macquarie Dictionary and you look up 'inhibit', and it says this is a word that can mean 'stop' or 'slow down'. If you are a chemist or a biochemist, for instance, I imagine, talking about a chemical reaction or an enzyme reaction, when you are talking about 'inhibiting' it is probably important to know whether you mean 'stopping' or 'slowing down'.

So just because you use a word that sounds fashionable and is used by your colleagues around you, and you think that is the one that fits the bill, it is probably a good idea to stop and think what it actually means.

Excessive premodification is beloved of scientists. I have got an example here: 'a series of fixed duration sequential constant rate infusions'. If you are familiar with this field and you have read this little combination of adjectival premodifiers many times, you probably pretty much know what it is about. But if you put yourself in the position of a poor punter out there reading this for the first time, trying to work out whether it is fixed duration rate infusions, or rate infusions, or sequential infusions that are at a fixed rate, or whatever, you realise that it would probably take them two or three times of reading over this sentence before they worked out what you mean. You could put hyphens in, but that still makes the reader have to stop and work out exactly which hyphen belongs where and what it means. This one is better cast as 'sequential infusions of constant rate and duration', which is much more helpful for the reader.

My second example, 'genetically modified insect resistant early ripening sweet corn', could have two or three different meanings. It could be sweet corn that is genetically modified to be insect resistant, or sweet corn that is genetically modified to be early ripening and that also happens to be insect resistant, or it could even be something to do with genetically modified insects that were in some relationship with early ripening sweet corn. So, clearly, even if it meant using one or two more words, it would be better to spell out what you mean and not have all these words bundled up, one on top of the other, all modifying the same noun, so that people have to try very hard to work out what you mean.

I thought that passive voice had more or less passed out of the standard scientific way of writing, and I was horrified a few years ago, when my son was at a local high school, to find out that in their science classes they were still being taught to write totally in the passive voice. When they wrote up their experiments they had to say, 'The Bunsen burner was lit,' and, 'The beaker was placed on the Bunsen burner' – this whole way of writing. Do people all write up their experiments like that in their papers? I think this myth has been largely debunked, and most people now do accept that all of the major high-impact journals accept papers that are written in the active voice and not in the passive voice.

That is not to say that there is not a place for passive voice. In fact, it would be very boring if a whole technical paper or report was written totally in the active voice, so you do have to mix it up a bit.

In this case, instead of, 'The humidity of the cabinet was tested,' you can put, 'Jones et al/we/I/the researchers' – whoever it was – 'tested the humidity'. Sometimes it is actually critical, because otherwise you don't know who did it. In the second example, we have 'Smith et al … found that …' and then later on in the same paragraph, 'However, it was found in a recent experiment that …' Who found? Was it Smith? Was it the author? You really don't know. So it is very important to put who it was that did what.

To turn to weak verbs: it is amazing how many things in science laboratories are 'carried out'. Very few things are ever 'carried in', but a lot of things are 'carried out'! In this case, the experiments 'were carried out to measure the rate of water uptake', but the 'were carried out' is really just a weak verb that has been put in there to stop you from using the real verb in the sentence, which was 'measure'. That is what was happening: you were measuring. The rate of the water uptake 'was measured', or 'We measured the rate of the water uptake'.

Another weak verb is 'performing'. I think this comes particularly in surgery, because surgeons love to 'perform' things. So again there is a lot of 'performing' going on around science labs and other places, and this is exactly the same sort of thing: 'Injection of the dye is performed' can be just as easily stated as, 'We injected the dye', or 'The dye was injected'.

Abstract nouns present a very similar problem. They occur when the verb is actually made into a noun. And then, because the verb has been made into a noun, you have to add another verb – in this case 'was made'. The example is, 'A comparison of the two techniques was made'. This could have been, 'The two techniques were compared'.

I have gone through that really quickly because, as may not be news to many of you, these are the standard techniques that are taught in good-writing classes. But scientific texts are littered with this sort of thing and it does make it very hard for the reader. Once you know these few rules, you can home in on these things in texts, weed them out and very quickly get something that is much more comprehensible.

In this regard, to get your writing sharpened up so that when people read it they will sit up and take notice, what do you do?

Most institutions, I suspect, have courses that you can attend. For instance, here at the ANU, the Centre for Public Awareness of Science and the Centre for Continuing Education both have science writing courses, and I know that it is the same at other institutions across the country. You could talk to the science communicator in your institution and ask them to recommend courses for you to go on.

You can look at the professional organisations. For instance, the Australian Science Communicators run workshops and conferences, often with very good workshops on just this sort of thing: plain English writing and ways to improve your communication. Something like the Brisbane Writers Festival would have workshops you could attend if you happen to live in Brisbane. The Australasian Medical Writers Association has a whole professional development program that you can do and get a Professional Development Certificate, and that includes writing, clear writing, writing for the web, a number of different courses on how to improve your writing skills.

The Institute of Professional Editors will help you with editing texts and refining techniques of improving the consistency of your texts, use of terminology, units, and all the other things that editors do.

There is also a number of private providers around the country. My organisation offers training in science writing, and there are many others. It is not difficult to find help.

There are many books, too. If you go to your university bookshop or library you will find many books on science writing, how to write in plain English, science communication. You just need to get in and find these resources, go to some courses. You will find that there is a very simple skill set you can learn that will improve your communication.

The second barrier that I mentioned is that you go out and speak to people and they are hostile.

This barrier is related to public perceptions of science. In the last 20 to 30 years there has been increasing public concern about emerging technologies. Whereas at one time scientists might have been put up on a pedestal and everyone thought that they were wonderful, now the public has become much more sceptical about science. Radioactive waste, the mad cow disease outbreak in the UK, the possible effects of mobile phones on health, GM food, which particularly was a huge issue in the 1990s, and embryonic research – these sorts of issues, to name but a few, are making people more concerned about research and about the general research agenda, and about what scientists are really doing and how that will affect our society.

What has been done about this? In the 1980s and '90s it was felt that a lot of the public concern arose because people did not really know enough about science – they had not done much science at school, maybe. There was an educational deficit, essentially. So the way to fix it would be to spend lots of money setting up science centres, setting up science clubs, sending out scientists to give lectures in schools and in the community. And it was assumed that once people knew more about science they would not be concerned any more, because now they would understand how good and wonderful it was, just as the scientists thought that it was good and wonderful.

Unfortunately, despite the expenditure of masses and masses of money – millions of dollars, particularly in the UK – by the end of the 1990s the situation was no better. In fact, it was worse. There was a crisis in Europe over GM food, and it was really holding back the research agenda in that area.

A huge inquiry by a select committee was commissioned in the UK House of Lords, and the committee issued a very famous report called Science and Society, making a number of recommendations.

Some of the key findings were around the facts that so-called 'scientific issues' involve many other factors besides science, and that framing the problem in a way that excludes the other factors, such as moral, social, ethical issues, invites hostility. Also, what the public finds acceptable often does not correspond with the risks understood by science. It may relate to many things, many of the other values and ethical things that are brought to the table, but also in terms of risks it is partly to do with the amount of control people may feel they have over their lives and their choices.

One of the spin-off reports following the Science and Society report made this often-quoted statement: 'Science is too important to be left to the scientists. Their knowledge and their assessment of risks is only one dimension of the challenge for society. When science raises profound ethical and social issues, the whole of society needs to take part in the debate.'

That has been taken up very much by the science communication community, who have, subsequent to this report being released, done an enormous amount of work looking at ways in which the public can be engaged in discussions about science that are not just about teaching, not just the deficit model. This new model has come to be thought about as a public engagement model, where the scientists are in a dialogue with the public. So, rather than just telling people about their science, they are engaging in a discussion.

To come back to these two models: firstly, there is the public understanding model, which is the deficit model or the education model. Really, that aims to increase public sympathy for science by telling people more about the science. It comes from the premise that the science is good and people just need to understand it better, and it tells people rather than listening back to what the concerns are. It places the ethical values and 'feelings' in a narrow context with the science, and by definition, since the scientist is out there talking enthusiastically about their work, it tends to exaggerate benefits and play down uncertainties, and doesn't allow a huge forum for fear to be discussed. Also, it places the ownership of the science with the scientists.

The public engagement model, on the other hand, aims to stimulate more debate and it does not make assumptions about the outcome. To truly have a dialogue among any number of people, all who are going into that dialogue need to do so with an open mind and be prepared to listen to all other sides of the discussion. So this model encourages feedback and debate. Also, the ethical issues and values are placed within a much broader social context, and in this context benefits can be sometimes more realistically discussed and, as well as the uncertainties, people can express their fears. It is more inclusive.

Where would you get training in terms of these aspects of science communication that you may need to know about in order to go out and talk to a sometimes sceptical public? There are plenty of books and articles to read about it, you can attend workshops and conferences – again such as those put on by the Australian Science Communicators. One of the things that I think it is good to do is to read reports, including government reports, about how scientific research is translated into policy. Often governments release documents for public consultation before policies are made, and these documents discuss different sides of the issue, discuss the ethical issues. I was for a couple of years involved with a review of Australia's legislation on embryo research. As a result of that, a report was published which has come to be known as the Lockhart report and looks at all sides of the issue. It includes quotes from the people that came to the public hearings and who put in public submissions. It is really worth while getting some of those reports and reading them, and seeing what all these other values are that people bring to the table.

It is good to talk to other scientists and science communicators about their experiences of talking to the public about the sorts of issues that cause public disquiet.

Barrier 3 is the one that is a bit more of a bee in my own bonnet: people argue over what your results mean.

I think science evidence is a bit like this, by now quite well-known, quote from Donald Rumsfeld at the time of the Iraq war. Clearly, in science there are 'known knowns', 'known unknowns', 'unknown unknowns', things 'we don't know we don't know' – and all those things. And so this thing that we think of scientific evidence, which is talked about in the newspapers, scientific papers and public discussions, is actually a little bit grey.

So how much evidence is enough? When you think about it, you realise that our language allows us very many adjectives that we can describe anything with, and we can describe evidence with all these different adjectives. We can go from 'no evidence' through 'weak', 'poor', 'limited', 'moderate', 'consistent', 'clear', 'good', 'strong', 'convincing', 'compelling' and 'excellent' to 'unequivocal'. So, if you read in your newspaper that there is 'good' evidence for something, how do you know whether that is better than 'clear' and not as good as 'convincing' or 'compelling'? We use these adjectives without normally thinking about them, but there is a kind of an implied scale, so that when you write that this is 'strong' evidence and you have come out of your lab and you have done the experiment, you know what experiment you did that made you think that it was strong. But the person reading the article, once it has been summarised down into only a few words in the newspaper, doesn't know, and so they try to think about what that means – and it may mean many different things to different people.

For instance, here are two examples from the press a year or so ago when I gave a talk similar to this somewhere else.

The first statement, 'There is clear evidence for a link between diabetes and colon cancer', does not really give you any idea of what the studies were that went into this 'clear evidence'. It could have been any one of a number of different things. (I will come to that a little bit more in a minute.) People usually just read a statement like that and assume something from it, but what they assume from it does not necessarily have much to do with what actual research studies that 'clear evidence' refers to.

The second one, which is from the draft report of the last IPCC report, is quite interesting because it talks about 'unequivocal' evidence and then it goes on to say that human activity has 'very likely' been the driving force in the change. It was interesting, because the previous IPCC report had 'likely', and not 'very likely'. So one of the major changes between the two reports was that 'likely' had gone to 'very likely'. It shows how important these adjectives are.

So, if you see something like, 'There is strong evidence in favour of' something, what does that mean? What studies could there have been? Could there have been one very good-quality study which had a big effect? Could there have been several poor-quality studies with big effects? Could there have been several good-quality studies with small effects? Or could it have been something else altogether? You don't know, just from the statement.

Similarly, and even more worryingly, if you see a statement, 'There is no evidence to support' something, what does that mean? Does it mean that there have been several good-quality studies that show no statistically significant effect, several poor-quality studies that show a small effect, or several studies that show inconsistent results – or that no studies have been done? I promise you that I have read that term 'no evidence' used in important scientific papers where it did mean any one of those. You had to delve back into the results section to find out what it meant, but in the conclusion or in the abstract where it just said 'no evidence', it hadn't distinguished between those different meanings.

It is almost no wonder, because we rely so heavily on this word 'evidence', that often when something is published in the paper and it is about a contentious issue which involves evidence, people start arguing about whether this 'good evidence' means this or that or the other. Evidence, in the dictionary, is the grounds for believing or not believing something, but in the law courts in our legal system, which I think is very much embedded in the psyche of people, jurors are asked to only convict someone when there is no possible doubt. It is very rare, I suppose, in science that there is no possible doubt, so when we talk about evidence there is always going to be some cause for further discussion and argument about it.

I think actually it is better to try and avoid the use of the word 'evidence' and only talk about the studies that you have done. So instead of saying there is 'strong evidence', you can say that there have been several good-quality studies done – or something that gives the reader a bit more of a clue about what this 'strong' is.

Just very briefly: another way that language can derail the message is by putting in some bias when it is not warranted. So if you say there have been no studies that support a beneficial effect of polyjuice potion – or anything – for back pain, really if there have been no studies (and it does mean no studies here, because it says 'no studies' and not 'no evidence') why are we implying that the effect has been beneficial? If there have been no studies, then we don't know anything about the effects of polyjuice potion on back pain.

It is very easy to introduce bias if you don't constantly try and think about what you are saying.

Summing up, you should get as much training as possible in clear science writing. Try and be aware of broader issues, in science communication, about engaging with the public.

Be inclusive of your audience when addressing the public and listen to people's views. Don't dismiss their concerns. Also, be prepared to say if you don't know, because scientists often do paint themselves into a corner by not being prepared to say that they don't know, and then that causes loss of public face down the track when they have to admit that they were wrong.

Remember that scientific 'evidence' is much talked about but difficult to quantify, so try and avoid further confusion by giving concrete rather than abstract information.

Discussion

Philip Kuchel: Thank you, Janet. I remember when I first started writing a paper, my boss said to me, 'You've got a split infinitive'. I really didn't know what that was.

Janet Salisbury: It's all right, you can have a split infinitive.

Philip Kuchel: Well, the next thing I did was to race up to the Commonwealth Bookshop and buy the Complete Plain Words. I learnt all about it there.

Jenny Graves: I wonder what is the role of the mentor in helping early career researchers to write well. I am always faced with the problem, as I am sure everybody else would have been, of whether I write it myself because it is due on Friday, or send people to read Fowler's Modern English Usage. Or is there some other way that you can do two things – mentor for the science and mentor for the expression as well?

Janet Salisbury: I suppose I was very lucky, that I had a PhD supervisor who was very pedantic, not so much about the absolute detail of grammar and split infinitives and that sort of thing, but about getting the information really well organised. I suppose I took that for granted, but I think that is a critical role.

Philip Kuchel: Something which you didn't mention is the increasing use of teleological arguments. You find it in the press: 'The runner is feeling pain. His muscles want more blood supply, need more oxygen'. Teleology, to us as science journal writers, is a no-no.

[ Home | Contacts | Search | Index ]