The influence of Isotype
This is an edited version of a very long conversation with Sue Perks in Zeist, Holland, March 2009
Where did you first learn about Isotype and who were the key people involved in introducing you to it?
The first person who mentioned Isotype to me (except that I don’t think he used the word Isotype)—I think he probably mentioned Otto Neurath—was Richard Yeend who was working at The New York Times
as an art director in the business section, in about ’82 or ’83. I was just beginning to work on my first book about charts and diagrams. Yeend said I should look at Otto Neurath’s work. At about the same time I went to England and met Michael Twyman in Reading.
And what inspires you most about it?
Simplicity. Some of the things I was doing at Time
were very complicated – illustrations on top of a diagram on top of statistics. This was because Time
was a magazine for busy people who weren’t necessarily interested in the subject matter. I was very much encouraged by the art director Walter Bernard and by editors to change the look of Time
’s charts. Since I trained as an illustrator that seemed like a natural fit. I’d done work of this nature for the Radio Times
in England before coming to America.
The first pieces I did at Time
were rather complicated. Later I learnt that I could be pictorial but simpler, and to use one of Otto’s phrases, to be ‘statistically accountable’. Statistical accountability was what Neurath preached – he wanted readers to be able to count the numbers in his charts and to see what the subject was: the charts weren’t just illustrations with a number atttached to them (Although they actually were like that when Otto first started making charts,) I liked the idea of paring things down to the absolute minimum and keeping the charts simple but also pictorial. I got into a lot of trouble with academics such as Tufte, who was very critical of my work in Time
. He said that if you have to dress up the numbers then they are the wrong numbers; people should be intrinsically interested in the numbers themselves. This is a very hard academic viewpoint, and quite wrong if applied to a general-interest magazine like Time
, which was edited for busy readers who needed to be encouraged to read the sometimes rather dry stories, for instance about economics.
I think Tufte also visited the archive at Reading with his wife.
In Tufte’s second book he reproduces something from Time
but does not say who did it (nor did he get permission to use it!). He writes snidely that Time
hired an English illustrator and he terms the resulting work ‘chart junk’. I can talk forever about the silly situation between us, but in fact much of the time we’re not that far apart. He comes from an academic point of view, I’m a commercial artist. I design charts and diagrams so that people are able to understand sometimes difficult concepts. My audience is the general reader. Tufte is talking to other academics.
What inspired you to write the article in the Information Design Journal?
Karel van der Waarde asked me to do it. And I think my friend Paul Mijksenaar might have suggested me.
Does Paul Mijksenaar acknowledge any influence of Isotype in his work?
I think so, on a rather—I don’t think he would mind me saying this—on a rather superficial level; the look of the things.
What principles of Isotype have been most useful to you as a designer?
The main one is the idea of repeating an icon to represent a greater quantity, rather than making the icon bigger. But people always say wouldn’t it be more fun if you just made this person bigger than that person to show the difference? Often I find myself having to tell people: ‘no, making a person bigger is a geometric enlargement: the person has got fatter, and thicker too… you’d be surprised how small the visual difference would be between those two people, if I really enlarged one of them in an honest way.’
On that point, The New York Times
has, I think, overused the idea of representing values by the area of a circle. It’s a good charting method if there are huge differences in a data set. (You still have to include a key that represents each circle’s quantity.) You mentioned Rosling’s Gapminder earlier, and he uses that principle all the time. To me it’s a poor way to communicate numbers and meaning. Gapminder charts look precise and you can interact with them and insert different coordinates and different countries for whatever subject you’re dealing with. But you can’t actually extract precise numbers from them. William Playfair was right when he said charts are pictures of numbers, but they aren’t the actual numbers. You can see the trend, and that’s better than a load of numbers in a table. I think Neurath said that as well; you will remember a picture but you won’t remember the numbers. You can always go back and look at the original data. What was great about Neurath was that you could see what the subject was as well as the numbers. Hans Rosling has done a lot of good work but his charts all look the same. I come from a pictorial background, I want people to be interested and to identify with the pictures and not to have to read my carts as though they were text. Of course, this approach can be dangerous; the data must be there, and if I hide the data with pretty pictures then I’m just doing illustrations, not information graphics.
One of your latest books is ‘Wordless Diagrams’. You obviously work sometimes with words and sometimes exclusively with pictures
was an attempt to see whether I could explain a lot of things without any words at all. It was a kind of exercise.
What effect did Isotype have on your work intellectually?
It had an effect later. The first principle it confirmed was that I could use pictures and statistics together. But really I just used the surface look of some of the icons that Rudolf Modley included in his popular book of pictorial symbols. I looked at that book a lot. Some of the symbols came from the Netherlands, some were actual Isotype symbols (where Modley worked briefly), and some were Modley’s own icons. It was a very influential book for me because it was all black and white and the icons were all very easy to copy!
But after a few years at Time
I thought that we might be trivializing some of information we were showing. The period of getting people interested enough to read the articles was over and it was time to take the reader a bit more seriously. Some readers wrote to me saying if ‘you want to be a cartoonist, go off and be a cartoonist, but don’t try to mix facts and illustrations together’. I didn’t agree with them, but certainly that’s what Tufte was saying.
Ultimately I started to get more interested in Neurath’s work and I thought Time
should calm things down a bit, visually. I was rising in the ranks at the magazine and I got all the editors together and suggested we should change the direction of the way we were doing charts; we should make them calmer and maybe strip out some color because they were getting too colorful, splashy and illustrative. They were also getting ripped off by people like USA Today,
and some awful work was being published, and I felt bad about that. But the editors hated the idea of changing the charts: ‘Oh, so now we’re going to go back to boring charts!’
That was the beginning of the end for me at Time.
The top brass there were trying to appeal to a younger audience, but I (and many others on the editorial side) believed that younger people weren’t going to buy the magazine anyway, whatever different design it was clothed in. We'd just annoy the real readers. However, once they grew up, those young people probably would want to read a serious magazine, with charts that were in keeping with the tone of the articles.
The magazine gave me a sabbatical in 1993 and I decided within a week that I wasn’t going to go back. It was just when 3D drawing programs and Photoshop were becoming popular. The magazine replaced what I was trying to do—to make the charts calmer—with something which was, to my mind, much more overdone than the original graphics I’d been doing.
Before this you were working with Radio Times with Peter Brookes?
He was very big on humor. Is that something that you picked up on?
Yes, I liked him a lot and we got on very well together, David Driver, art director of the Radio Times
put us together.
I think David knew instinctively that I was good at organising things, and of course Peter was brilliant at drawing. We did some Olympic and World Cup pieces together. For instace, for a piece about Olympic show jumping, Peter would draw a horse jumping (over nothing) and I would draw a big directional arrow and a diagram of the fence being jumped showing how high it was. The two styles worked really well together. When we first started collaborating, the Radio Times
was all black and white. Later there a few color pages were added. Sometimes when we were working on a spread there was color on one page and black and white on the opposite page. The color pages had to go to the printer a week in advance of the black and white, so we had to plan the whole thing very carefully, because we couldn’t go back and change anything on the color page. We did a Soyuz/US link up diagram in that format; Peter drew all the machinery of the space hardware and I did an explanatory diagram that went with it. Of course now we have computers and processes aren't so complicated.
And the simplicity of your work, along side the fact that it can be produced on a computer, means that any magazine/newspaper processes can be done quite quickly now.
Yes that doesn’t help me because people just leave it later to ask for the work!
But your work did get much simpler
That was partly just an intellectual thing–I wondered what I could take away from a diagram or chart and still have it hold together. I also started to work just in black and white. I only added color when it was necessary to help to clarify the message. Color was never decorative (or I hope not!). One good example of this idea is Harry Beck’s wonderful map of the London underground. The use of color there was strictly informational.
Going back your work, it’s known for communicating complex ideas simply. Where does this come from?
It started with Brian Haynes who like me had been at the Royal College of Art. He got me an internship at the London Sunday Times Magazine
. He recognized that I wasn’t a very good illustrator. I’m still not much of an illustrator–I’m a designer who uses pictures.
Haynes saw something in what I was doing. He was an entrepreneur and an art director. He was a very good art director. He taught me more in the few weeks that I did in internships with him than I did in all the time that I was at Hull Art school and the Royal College of Art.
Brian was a great mentor, and he was able to pay me! He gave me some big jobs when he was art director at Women’s Mirror,
such as the Buckingham Palace and Wimbledon projects. We did all sorts of things together at other places that Brian worked: one project I enjoyed was about how to unload containers from huge cargo ships, for the cover of the Observer Magazine.
And we did a piece about Guy Fawkes where I collaborated with the illustrator Adrian George. That was a terrific period of learning; I got my tight, and simple illustrative style together during that time. That was a style that made the move to computers very easy. I could draw naturally, freehand, relatively well, but the prospect of sending a freehand drawing to print was somehow scary, so I deliberately tightened up the drawing and simplified it with no freehand work at all; I drew every line using French curves and oval templates and so on.
So that formed your visual language?
The other thing I learnt from Neurath and Arntz was from a wonderful loose-leaf book that they put together. It had pages and pages of little icons that they had used for their charts. It was a sort of visual dictionary.
That made me think about using my own icons again and again. It saves time, but much more importantly, it forms a personal visual language. If you look carefully through my work you’ll see the same little icons appearing in different graphics. It’s not because I’m lazy, it’s because it speeds up the work, and I don’t have to think about making another icon for a car, or a person running, or a kangaroo. I thought hard about all of them when I first drew them. So this person sitting at a table, or that person sitting at a computer (which I will update, of course) can stand for every person sitting. In a certain sense, this means that I now have a library of my own visual alphabet. Perhaps that’s why I always want to control both the words and the pictures in the jobs I do. Because to me, the words and the pictures are all the same thing. You might say some things in pictures and some things in words. In Wordless Diagrams
I tried to do it all in pictures. But most of the time it’s a better diagram if there is a real marriage of words and pictures.
It’s your own set of symbols that work for you, and which you update and add to,which actually is what Neurath did.
He wanted his visual language to be universal. Do you want yours to be totally universal?
A universal picture language? Yes, I want lots of people to see be able to ‘read’ it, but actually that’s almost impossible because people have many different visions of what things are. I actually like variations in spoken languages, and also in styles and forms of visual languages. Russian looks different to Japanese. I don’t want the whole world to be the same because I think that would be very boring. The world is already too homogenised. That might be good for communication, but I think differences in cultures should be celebrated, not diminished by making them invisible.
Going back to Isotype, the big departure between what Neurath was doing and what Modley was doing was all about universal picture language. Neurath was happy for things to remain consistent, whereas Modley wanted to adapt it wanted it to add a personality to it.
You say ‘personality.’ Do you view Neurath’s work as not having a personality?
No, no, that’s not what I meant.
He wanted his work to have a different personality?
He wanted it to have a human personality but it was universal. Whereas Modley wanted to take that work and change it for different audiences and different purposes. Modley was more commercial in his outlook as well. Whereas Neurath was not commercial.
I think that happens. I often counter Tufte’s criticisms of my work by saying that I am a commercial artist. I do try to apply good statistical principles to what I do, but the older I get the less commercial I get. But in the end I have to make a living!
You’ve admitted that Gerd Arntz is a hero of yours is it because of the style of drawing?
I think it is.
Did you become more like him or were you already working like that and you thought ‘wow he’s drawing like me!’
I think actually it was a bit of both, I was both disappointed and thrilled to discover his work. I was disappointed because I thought I had invented something! You absorb things and don’t realise you’ve absorbed them and then you think you’ve invented them. But the more I looked at Arntz’s work the more I saw him to be a real artist. The icons he had to draw could have been very ordinary, but he tilted a head slightly so this person looks a little puzzled, or that person might be stooping slightly rather than standing up straight. Perhaps he’s out of work. There’s humanity in the drawings. And that’s why I think that Neurath’s work has lasted so long. However strict the principles that he set down are, they are softened by the fun that you get out of Arntz’s work, looking at exactly what he’s done with the person who’s out of work. He is facing to the left and his shoulders are hunched slightly, it’s so simple but it is so real. You know it’s the simplest, iconic way to draw something, but the person depicted is alive.
When you are commissioned to produce a diagram or a chart, are you given all of the information as text?
It is different for every job. Some magazines or clients know exactly what they want (or think they do!). In that case they are just hiring me to carry out their ideas. I’m not very interested in that.
So then you’re an art worker rather than a transformer?
Yes, that’s exactly right. The best clients are people that have a good research department that I can work with, and have a problem that we have to solve, or a story to tell.
So the best brief for you would be a question?
And a list of the people that you would be working with in the research department?
Yes, and increasingly that’s a list of links on the web. Nowadays we talk about speeding things up. Google is the most amazing place to use, but I have to be very careful. I will often search a subject in two or three different ways. That way I’ll get different answers. I’m increasingly finding that the deeper down I go on the list of possible answers, the more I find. The answer I want might easily be further down; it just didn’t come up to the top because not enough people asked for it.
Did you use the principles of Isotype in the film Vampire Power?
I didn’t know that I was using them but I guess I just absorbed it.
That film was for Good
magazine, but I made an earlier one for TedX (the 10th Ted conference) I had done a print piece for Richard Saul Wiurman’s Understanding USA book first (he introduced at TedX) and he invited me to do a presentation at the conference, so I decided to make a film of the piece…in other words, reinvent it for a different media experience: a stage with a live audience.
It was the same with the vampire piece. Good magazine was launching a website that included videos based on ideas from the printed magazine. I broke the print piece down into a narrated story so that instead of being one picture, it now became a linear story. The drawing of the vampire comes in much later in the film than it does in print. I drew a number of stills, then my son Rowland animated the whole thing, working with the stills and my narration.
The narrative works very well. It’s a great little film.
What do you consider to be the legacy of Isotype?
I think it’s the principle that I mentioned at the beginning: statistical accountability. You can see what the chart is about–the subject–but you can also count the numbers. Neurath wanted to help people see the numbers and the subject at the same time. That’s what I was doing initially at Time
when I would do a drawing with a chart on it. The pieces told you what the subject was, and
what the numbers are. Making information graphics this way is basically no more than doing an illustration with a fever line or bar chart, or pie chart on top of it. I did a lot of charts like that at Time
and they became very popular. But using the Neurath (Vienna) method is much better because it allows the pictures themselves
to have statistical accountability. To my mind, the best way to explain something is by drawing a chart or diagram that contains both pictures and numbers, as long as they are in the right relationship to each other.