Macdonald-Ross & Waller - The Transformer Revisited

How can we organise complex information for the benefit of the reader? This question isimportant, for in a modern society people cannot always meet face to face: often we must get the information we need from documentarysources – pieces of paper or computer files.The Open University is a case in point: we use correspondence texts as our main teachingmedium. The effectiveness of these texts is naturally of great concern to the University;and by our research we try to help theUniversity improve the quality of its printedteaching material. However, the issues we raisehere are not peculiar to our own situation. The school system has its text-books, theindustrial corporation has its field servicing,the armed forces have their training systems,Government departments must inform thepublic: all these organisations depend on information, and this information is usuallycommunicated in text form. We think the principles that apply to us apply (with varia-tions) to these related cases.

One person talking to another – that everyday sight we take so much for granted –can usually make himself understood. Talkingface to face people deploy complex techniquesalmost instantly, without self-consciousthought. But, how much more difficult it is to make oneself clear to a reader one cannotsee, and who cannot ask questions. So, as onemight expect, communication breakdowns arequite common, and can have the most seriouseffect. Recently farmers were advised to cutdown trees suffering from Dutch Elm disease –but were not told to burn the wood! And howoften have we met forms we could not fill in orregulations we could not properly understand?(The Government’s ‘brief guide’ to vat was asmall encyclopedia packed with officialese.)This bungling incompetence occurs often andis too serious for us to accept complacently.We must do something about it.

177

authors’ addressesMichael Macdonald-Ross2 Church StreetWheathampsteadHertfordshire AL4 8AP

Robert WallerInformation Design Unit79 High StreetNewport PagnellMK16 [email protected]

Information design journal 9/2&3(2000) 177–193

The transformerrevisitedMichael Macdonald-Rossand Robert Waller

Written in 1974 while theauthors were with the OpenUniversity, this paper firstappeared in the 1976Penrose Annual.

The original abstract, writtenby the Penrose editor, read:Break down the barriers in theinterests of the reader. Takeresponsibility for the successor failure of the communica-tion. Do not accept a label ora slot on a production line. Bea complete human being withmoral and intellectual integrityand thoroughgoing technicalcompetence. This is the mes-sage of this article by twohighly professional communi-cators at the Institute ofEducational Technology of theOpen University, MiltonKeynes. It examines the rangeof complex problems involvedin putting the expert’s mes-sage in a form the ordinaryperson can best understandand use.

It is reprinted here with minorchanges that mostly reflectthe current unacceptability ofthe pronoun ‘he’ used generi-cally. The authors have alsoadded a 2000 Postscript.

Michael Macdonald-Ross & Robert Waller · The transformer revisited

Information design journal 9/2&3 (2000) 177–193

Like most organisations of our age, the Open University has a production-line system.Many different people help create the teachingmaterial: authors, educational technologists,editors, designers, illustrators, photographers,television producers. All these work in separatedepartments, often quite unco-ordinated onewith another, and all too often mutually suspi-cious rather than co-operative. No one personis responsible for mediating the whole creativeprocess. With so many fingers in the pie it fol-lows (of course) that important things are leftundone and no-one accepts responsibility for thesuccess of the texts as instruments of communica-tion. In our opinion the University works notso much because we are expert at distanceteaching, but because, in the main, studentsare keen, intelligent and hardworking.

We suggest that these large organisationslack a crucial process – transformation – and a person to carry the process out – the transformer. This is the skilled professionalcommunicator who mediates between theexpert and the reader. The transformer’s job is to put the expert’s message in a form thereader can understand, and to look after thereader’s interest in general. For example, anyreasonable query the reader might have shouldbe thought about and catered for in a propermanner. Put this way the role seems simple,but it is not. A transformer needs a good general education and a wide range of particu-lar technical skills. Transformers will findthemselves dealing with all kinds of experts,most of whom do not know what they want to

say, or how to say it; they will discover how little is really known about the readers, and thetechniques they need are scattered around inthe most unlikely places.

We would like to have you think that theidea of the transformer is our own, freshlyminted. However, it is not. The name was firstcoined by Otto Neurath when he developedthe system of graphic communication knownas Isotype. Though we extend the idea beyondits original scope, Neurath’s basic idea wascorrect, important, and ... ignored for half acentury!

At last there are signs that Neurath’s long-dormant idea might get the attention itdeserves. We see with interest how some nowcall themselves ‘editorial designers’ to say thatboth text and graphic elements must be han-dled together, as a unity. Also we see coursesin ‘Information graphics’ springing up. Such alabel says to us that design can be more thanexternal decoration, it can help the way the message itself is con-structed. And we see that some professions ofrecent origin do have one person in charge ofthe creative process (creative directors inadvertising agencies and producers of televi-sion programmes). Even our own Institute isthinking of changing its name to the Instituteof Educational Development. The word devel-opment suits the notion of transforming quitewell. The word ‘transformer’ has not yet beenbagged by any of these groups, and so maybecan be accepted by them all.

The transformer starts with what to say, andthen resolves how best to say it. Naturally, thisdistinction must not be overdone. What youwant to say does partly determine how you sayit and, in return, the content of a message isalways altered to some extent by the way it isput over. Nevertheless, the distinction is a useful one. First one discusses the content ofthe message with the experts; later one works

178

It is the responsibility of the ‘transformer’ to understand the data, to get allnecessary information from the expert, to decide what is worth transmittingto the public, how to make it understandable, how to link it with generalknowledge or with information already given in other charts. In this sense,the transformer is the trustee of the public. Marie Neurath (1974)



out the exact form of the message with the helpof illustrators, photographers, printers andother technical staff. The skills needed and theproblems faced differ at each stage. The cycleis complete when the transformer discoverswhat effect the message has on the reader.

The transformer is overseer of the wholeprocess of communication – what is said, howit is said and what its effect is. They work withcolleagues whose skills are more specialised tomake sure the message gets across and toreduce the chance of communication break-downs. Transformers act on the reader’s behalfas best they can, sorting out the kind of issuesa reader might raise if they were present in person.

What to say

Transformers must be as closely involved withthe content of the message as they are laterwith its presentation. They must understandthe subject at hand so that their later judge-ments are informed judgements; but there ismore to it than that. The transformer will usually need to help the expert get his or herideas sorted out so that the subsequent act ofcommunication has some chance of being successful.

The transformer is the partner of the subjectmatter expert and not their slave. Our experi-ence suggests that experts are not always asexpert as they seem at first sight, and evenpublished statistical data may be unreliable ormisleading. Moreover, the expert’s natural tendency is to think of their subject rather thanto think of the reader. For reasons of this kind,the transformer should not accept an author’sinstructions without critical thought. Theymust question and analyse until they can putthe author’s intentions in proper perspective.

The idealistic young communicator may besurprised to find their source vague and con-fused instead of clear and authoritative, but the expert is only human! They may not haveorganised their ideas yet; they may feel uncer-tain and anxious about issues that they havenot sorted out but which cannot be ignored.The transformer helps as a sympathetic listenerwho gently refuses to go away until the confu-sion is sorted out. At any rate, you must stickto it, for no-one can make a good communica-tion out of muddled thinking.

The transformer sets about their task withthe help of two insights: a good communica-tion is selected for a purpose and has a sound logical structure. These two insights lead to thetechniques which a transformer can use in discussions with the source:

Purpose and objectivesAll human communication depends on artfulselection, since one can never say all that isknown. The simplest basis for selection is aclear statement of purpose. The dialoguebetween expert and transformer may start withvague statements like: the reader should havesome idea of this or should appreciate that.The transformer works to derive objectiveswhich are as precise as the particular situationrequires. For example, they may ask, what willa successful outcome look like? What will thereader be able to do if they have interpreted themessage correctly? After clarifying what theexpert wants to say and why they think it worthsaying one can discard irrelevant material andidentify the prospective pay-off for the reader.

Tasks and errorsSometimes the work on the objectives must besupported by the kind of professional studyoften found in military or industrial training. If the reader is to perform a well-specifiedfunction or activity, then that activity can be

179



analysed into its constituent tasks, and typicalmistakes can be collected and analysed. Oftensuch a study starts with the ‘master performer’– an experienced person who exhibits all theskills and know-how necessary to meet themost demanding job standards. Such a studyjustifies the selection of objectives by connectingthe message with the reader’s world.

Organising principlesIn education we make less use of the notion of tasks and errors and more use of organisingprinciples. All subjects consist of facts, argu-ments, theories, problems and procedureswhich (to a greater or lesser extent) are unifiedby central themes or organising principles.Sometimes these principles are quite grand,covering vast domains (the theory of evolutionwould be one such example). Sometimes theprinciples are more practical, for instance,computers can only carry out those operationswhich can be exactly specified in a pre-determined code. Sometimes (as in the socialand human sciences) one finds competingworld views which one must understand inorder to interpret the opinions of the experts.Such world views are to some extent mutuallyexclusive and a transformer must realise thatthey lie behind and largely determine what

experts say and do. The principles help usorganise a map of the terrain.

FactsIt is a fact that 8 per cent of men (but only 0.4per cent of women) have defective colourvision. It is a fact because experts agree onwhat ‘defective colour vision’ means: there arestandard tests to identify it and the results ofkey investigations are not in dispute. On theother hand, if someone says ‘whites are moreintelligent than blacks’ then the status of sucha statement is much less clear. The statementis not known to be true; but it is not known tobe untrue – it is in dispute, contentious,unsayable, and offensive even to research.Therefore the job of a transformer is to inter-cept opinions or interpretations masqueradingas facts. They must identify the status of anykey statements made by the expert.

ArgumentsPeople differ in their ability to deploy argu-ments in an appropriate manner. This is toobig a subject to be fully treated here, but soimportant that it must be mentioned. Thetransformer’s dream is to see all argumentsvalid and all propositions as they claim to be(true, untrue or uncertain as the case may be).Although this is the kind of quest which neverquite ends, even a moderate amount of successdoes a good deal to ease the reader’s problems.

BiasMost people welcome a second opinion to help get the balance of an article right, butthoroughgoing deliberate bias is difficult tocure. The transformer may be resisted becausethe author, or even the whole organisation, hasa vested interest in maintaining the bias.Companies may wish to hide defects in theirproducts or in their finances; Governmentdepartments may not wish the public to know

180

And the next morning they issued not twenty ounces of bread – the prison-er transport ration at the time – but no more than nine ounces ... in theevent, only one loudly asked the guard distributing the bread:

‘Citizen chief! How much does this ration weigh?’

The whole car fell silent. Many waited before beginning to eat their ration,expecting that theirs, too, would be reweighed. And at that moment, in allhis spotlessness, the officer appeared ...

‘Which one here spoke out against the Soviet government?’

All hearts stopped beating. Solzhenitsyn, The Gulag Archipelago, 1973



their full rights under the law; lecturers nowa-days do often allow their political opinions tocolour their courses. In such cases transform-ers may get caught by their dual loyalties. If they cravenly submit to their colleagues then they are just lackeys (running dogs); ifthey side entirely with the reader then maybethey can no longer influence their own organi-sation (paper tigers!). The transformer shouldplay for honesty and integrity, qualities whichdo tend to pay off in the long run. OttoNeurath used to say that ‘the transformer is the trustee of the public’. That is a pretty fairway to put it.

How to say it

The transformer’s job now is to plan the presentation of the message. Since text can onlyuse printed language and graphic devices toput over the message, to do this they need skillas a writer and editor, and as a typographerand graphic communicator. Of course, they do not have to write or draw everything them-selves – but they must be skillful enough toensure that the presentation enhances the message.

LanguageNo transformer can hope to succeed unless heor she is a skilled wordsmith. Experience andexperiment teach us that simple sentences withactive verbs and familiar words can be readand understood by most adults. As the wordsget less familiar and the construction gets morecomplex, so more and more readers fall by thewayside. Different target populations have dif-ferent levels of reading competence, and it isthe job of the transformer to know their audi-ence, to make sure the language is tailor-madeto their requirements. Reading tests can be

used to assess the competence of the targetpopulation, readability measures can be usedto predict the difficulty of the text and thereare many books which show how to write clearEnglish. We are particularly fond of the booksby Harold Evans and Leslie Sellers*, whichshow how much we can learn from those FleetStreet sub-editors who daily put complex ideasin a form which can be read and enjoyed bythe general public.

LinksNew or difficult ideas can be grasped by thereader, but only if you make the right connec-tions with his familiar world. By his explana-tions and choice of examples the transformercan make difficult ideas seem easy becausethey are linked to things the reader alreadyunderstands. In this way good writing makesuse of a reader’s own knowledge and experi-ence. Technical terms can, if necessary, be putso that we can all understand, yet at the sametime the transformer must not debase the origi-nal notion by sloppy thinking or unclearexpression. Explanations can and should bechecked with the expert, and reference booksconsulted.

181

The English Language becomes ugly and inaccurate because our thoughtsare foolish, but the slovenliness of our language makes it easier for us tohave foolish thoughts ... Millions of peasants are robbed of their farms andsent trudging along the roads with no more than they can carry: this iscalled transfer of populations or rectification of frontiers. People are impris-oned for years without trial, or shot in the back of the neck or sent to dieof scurvy in Arctic lumbercamps: this is called elimination of unreliable elements. George Orwell, ‘Politics and the English Language’ 1946

* see Evans, H. (1973) Newspaper design. London:Heinemann; Sellers, L. ( 1968) The simple subs book.Oxford: Pergamon Press.



Typography and graphicsTo attract and keep interest and to ensure legi-bility are the classic aims of the designer andthe typographer. These aims are still impor-tant, but the transformer must have a muchbroader vision than that. Often the graphic ele-ments carry a crucial part of the message, soone cannot regard design as just external deco-ration applied to an existing message. Rather,we go along with the ideals of the NewTypographers: design is an integral part of thecommunication process. Good typographyhelps readers plan their reading strategy, tellsthem where they are and helps them to findtheir way about; good graphic design allowsone to say in words and illustrations whatcould not be said in either form alone. Weshow three examples of graphic communica-tion: an algorithm, an Isotype chart and atable. They illustrate perfectly the kind ofproblems a transformer may face daily, and arewell worth careful study. There are other spe-cialised techniques (especially in mathematics,science and technology) which the transformermust collect as the need arises.

182

Real Isotype diagrams require the process oftransformation. Marie Neurath of the Isotype Instituteproduced the pictorial chart from the table above• notice the clarity and simplicity of the language and

symbols• the information load has been cut to leave the

message clear. Only alternate years are shown, andthe change of scale means fewer man-symbols

• the prominent main caption is important – it definesthe context of the statistics

• Isotype charts do not just display statistics – they aimto show what the statistics say – and to give visualimpact to their message.



A criticism of the rather baffling table above can help usto design a better alternative:• ‘country of last permanent or intended future

residence’ is hard to understand.• the monotonous typography makes similarly worded

items of different status look alike.• evenly spaced columns and old-style numerals make

the figures hard to compare.• insensitive use of space means that ‘total’ rows of

figures are often visually associated with the wrongsections.

• typographic confusion also results from an attempt tosave money by using only one size and style of typethroughout the publication.

A transformer needs educational, editorial andtypographic skills and also a questioning spirit to producethe result pictured top right.• obscure wording is simplified.• typographic and spatial coding is used to associate like

with like.• the ‘total’ figures have been grouped together to avoid

confusion and to provide a summary of the table.• one of the aims of the course is to help students

develop skills in interpreting tabular statistics. Apictogram would thus be inappropriate, althoughpossibly more effective in other ways.

• when we traced the statistics to source, we found thatthe figures were derived from quite small-scale surveysof air and sea passengers. Thus we decided to cut outthe decimal point.

183



184

This example was writtenfor the Open Universitycourse ‘The HandicappedPerson in the Community’(P853). Ten years agoBrian Lewis and PeterWason showed that thestructure of rules andregulations could bedisplayed better asalgorithms (logical trees).But to do this the detailsof the design must beright. Although theauthor’s basic idea isreasonable the originalversion is linguisticallycomplex, Iogicallyconfused and graphicallychaotic.• Ianguage. Most readers

simply cannotunderstand long,complex sentencespacked with conditionalclauses – read a fewboxes and see!

• Iogic. The logicalstructure ‘Are you A or Bor C or D?’is ambiguous– what is the readersaying ‘Yes’ or ‘No’ to?

• graphics.Where to lookfor the next step? Thatis quite a problem here.

The new version on thenext page makes thesechanges:• supplementary information removed to notes• one question at a time• simple language, addressing the reader directly• Iogic clarified• graphic flow simple and predictable: outcome boxes denoted by

heavy border, No and Yes in consistent direction, number of boxesminimised and length of connecting lines kept as short as possible.

This is only one way of doing it. It may well not be perfect but, as yourown eyes tell you. it is a huge improvement on the first version.Twelve drafts and several long phone-calls lay between the first andlast versions.



185



ActionIf you want the reader to do something afterreading the text then these points are crucial.We assume that what the readers should do isalready decided (by task analysis or whatever);now we want to know if the text enables themto perform. The three key questions are: willthey understand what you are asking them todo? Can they actually do it? (is it physicallyand psychologically possible?). Will they knowwhen they have completed the task success-fully? The simple Highway Code* rule, ‘REDmeans stop’, meets all three criteria simultane-ously in three words. But what about ‘at 70

mph shortest stopping distance is 70 ft think-ing distance +245 ft braking distance = overallstopping distance 315 ft’? This is useless infor-mation. In the first place the accuracy is spuri-ous, as the Code admits (wet and slipperyroads, different vehicles, poor brakes and tyres,tired drivers ...). But in any event, how onearth are drivers meant to translate those num-bers into appropriate action on the road? Theanswer is, they can’t and they don’t! Gooddrivers may learn to judge stopping distancereasonably well, but no thanks to the HighwayCode for that. The infamous Green CrossCode is another case in point, so complex lin-guistically and logically that the young childrenit is meant for cannot grasp and use it with anyfacility.

Organisers and signpostsThe army saying goes: ‘First I tells them whatI’m going to tell them; then I tells them; then Itells them what I’ve told them!’ People do needto know where they are, where they are goingand what the prospective pay-off is. As youproceed they need to know what the status ofthe message is. Are you giving them main orsubsidiary information? Are you asking a ques-tion or giving an instruction? The form of themessage must show its status and function.Texts are full of devices which help the readerfind his way around: objectives, prefaces, intro-ductions, contents lists, headings, questions,instructions, numbering systems, glossariesand indices are typical examples. If the com-munication aims at specific goals then theorganisers are strongly directive (objectives,instructions); if the reader is allowed morefreedom then the permission-giving organisersbecome more important (contents, headings).Thus the choice and emphasis depends on thepurpose of the communication and the situa-tion of the reader. Organisers are not isolatedbits and pieces; they must fit together. Thecontents page, glossary, index, numbering sys-tem, headings, diagrams and main text aremutually interdependent; what you do in oneplace affects all the others. The design and lay-out of a two-page spread can become a littlecameo of transformation; all kinds of informedguesses are made about reading behaviour sothat readers can best access the informationthey need.

Feedback Since you cannot be there to answer questions,readers may often wonder whether they haveunderstood you properly. So you must allowthem to check their understanding by givingthem feedback. You must tell them how torecognise when they have executed an instruc-tion or answered a question correctly or under-

186

Our charts give as little detail as possible. In this they are the opposite of a naturalistic picture, and especially of a photograph ... It is certainly not necessary for a chart to be poor in details, but all details must have teaching-value. Otto Neurath, 1936

* The Highway Code is a UK government publication fordrivers that encapsulates the law and good practice. TheGreen Cross Code, mentioned later, contains road safetyinformation for children.



stood a key idea. The feedback should takecare of the most frequent kinds of mistake andshould be placed immediately after the stimu-lus (question, instruction or whatever). It isworth mentioning that feedback can be quitesubtle, for example repeating the same idea indifferent form gives the reader a second map-bearing which checks their first interpretation.‘Of making many books there is no end; andmuch study is weariness of the flesh’(Ecclesiastes 12.12).

TestingThe transformer cannot get everything right;you will be lucky to get things half-right at first.Yet the final version may need to be very goodindeed if it is to succeed. How do we face thisproblem? We say there is a time for guessingand a time for testing. The testing should bedone during the transforming process on arough but complete draft version. The object is (obviously) to gauge whether the purpose ofthe communication is being achieved, and toidentify and collect sources of misunderstand-ing. Any text which is important enough tohave money spent on it is important enough to be tested, and the transforming processmust allow time for the lessons learned to beput into effect.

The later stages

Earlier we saw how the transformer workedwith the expert source; now we see them work-ing with two quite different kinds of person –the skilled craftsman (or technician) and theresearch worker. We deal here in less detailthan earlier, though in the long run the trans-formers’ relations with craftsmen andresearchers will be just as important as theirrelations with sources.

ProductionTransformers depends upon illustrators, pho-tographers, compositors, film editors and otherskilled people to help him realise their vision.They can make a real contribution. At theirbest they can teach the transformer a good dealabout the possibilities and limits of their arts,and about the costs and implications of variousproduction methods. In some cases (scientificphotography, for example) a skilled operatorcan help the growth of the subject-matter aswell as its subsequent presentation. The trans-former should deal with these specialists inperson, not through any kind of intermediary.Without the transformer the final productoften reflects the production process ratherthan the needs of the reader. How often havewe seen the fussy house-style applied unintelli-gently, or the diagram which has lost its mean-ing because the illustrator did not understandit! Often a freehand sketch by someone whounderstands works better than it does whenredrawn ‘professionally’ by an illustrator. The last decade has brought us strike-on com-position, dry transfer lettering and simplifiedphoto-litho processes; modern print methodssuch as these cut out many of the old con-straints and mysteries of printing. Under-ground papers take advantage of those newtechniques to bring the communicator closerto the reader. We should think long and hardabout such ideas.

187

Where is the argument? Traditional visual material, oftenmarvellously presented, does not always reckon with thisessential point. The life and habits of an animal may beshown in a film, but if we ask why certain features werepresented, the answer is frequently quite inadequate. It may merely be that the photographer saw the chanceof a good or amusing shot. That may be excellent enter-tainment, but - where is the argument? Otto Neurath’ 1944



Evaluation‘Discovering the effects’ is the last part of thetransforming cycle. This means going out intothe field and looking at all the consequences ofthe communication that can be found. Thereally gross errors should be picked up earlierin developmental testing, but many subtleeffects only show up later in the field. Theinformation gained can be used for subsequenteditions, but is in any event useful since similarproblems and situations are just bound tooccur. Transformation is an active process thatlearns, not a passive dispensing of recipes.Evaluation can be a complex technical matter.Unless the transformer has research trainingthey may well need the services of a profes-sional researcher to them to ways of collectingdata. The budget for evaluation depends onthe cost and importance of the communicationand the prospects of similar publications in thefuture: we do suggest that there should alwaysbe such an item in the budget.

Recurrent problemsAs time goes on the transformer will noticecertain problems coming up time after time.Sometimes they cannot be solved simply orintuitively; they need detailed work in depthfor a long time. This is the job of the basicresearcher, who may well be in-house if theinstitution is large enough. The relationshipbetween these two colleagues is often uneasy,with the transformer wanting solutions yester-day and the researcher limited by what it ispossible to do, given present knowledge and

resources. A similar case is the relationbetween general practitioners and medicalresearchers. The GP has patients now in thesurgery, but the researcher knows that somequestions just do not have easy answers. For all that, the art of medicine advances by thismixture of co-operation and conflict, and soshould the art of transforming.

Finale

Our message is humanistic: break down thebarriers in the interests of the reader. Takeresponsibility for the success or failure of thecommunication. Do not accept a label or a sloton a production line. Be a complete humanbeing with moral and intellectual integrity andthoroughgoing technical competence. Be atransformer!

188

He who knows best what to omit is the best teacher, and he who can omitnothing should not be a teacher at all ... The Vienna school postulates: toremember simplified pictures is better than to forget accurate figures.

Otto Neurath, 1925



2000 Postscript RW

The Penrose Annual (subtitle: ‘the internationalreview of the graphic arts’) was published fromthe 1890s to the early 1980s. It started as ashowcase for printers and manufacturers ofprinting equipment, but it soon covered otheraspects of printing’s cultural context, includingtypography, illustration, and printing history.The 1976 volume, in which ‘The transformer’appeared, included not only articles of obviousinterest to printers (about imaging science andphotosetting), but also mapmaking, postagestamp design, private presses, magazine designin Poland and corporate identity.

When this paper was written I was a recentrecruit to the Textual CommunicationResearch Group (TCRG), set up by MichaelMacdonald-Ross, which operated from themid seventies until the mid nineties. A gradu-ate of the typography course at the Universityof Reading, I joined a group that soon includeda cognitive psychologist, an information scien-tist and a linguist (although he did not staylong, and that side of the work was not sup-ported by a specialist). Michael Macdonald-Ross’s background was originally in zoology,and subsequently training technologies, and hehad been one of the consultants employed tohelp set up the Open University. The group’sgoal was to draw together whatever usefulapproaches it could find in order to inform thedevelopment of better distance educationmaterials.

The group was part of a general comingtogether of graphic design with its neighbour-ing disciplines.Other programmes in the UK

included Herbert Spencer’s group at the RoyalCollege of Art, the work of James Hartley andPeter Burnhill, and Michael Twyman’s curriculum at the University of Reading. And although it was not multi-disciplinary inthe same way, any account of the period

should mention the immense contribution of Patricia Wright and others at the AppliedPsychology Unit in Cambridge.

Michael Macdonald-Ross’s major contribu-tion to this process was to draw links with amuch wider range of approaches and disci-plines than most designers were previouslyaware of. There had long been links betweenpsychologists and designers, but at TCRG wewere exposed to other fields and sub-disci-plines such as content analysis, task analysis,bibliography, taxonomy, cybernetics, theoriesof learning, and instructional design.

The Institute of which we were part had thejob of helping academic specialists communi-cate clearly to students studying at home. So Otto Neurath’s transformer role was anobvious model for us to adopt. And when wewere asked to contribute an article to PenroseAnnual, it seemed a good opportunity todevelop our own interpretation of the role.

Twenty-five years on, many of the pointsmade here continue to be relevant and areworth debating again. Reading it again aftermany years, I was struck by the moral dimen-sion expressed in its conclusion: ‘Be a com-plete human being with moral and intellectualintegrity and thoroughgoing technical compe-tence. Be a transformer!’. This moral dimen-sion has permeated information design, to theextent that some appear to see it as almostentirely an altruistic (and by implication, pub-lic sector) activity. Indeed people reportingcommercial projects at information design conferences are sometimes made to feel theyare prostituting their skills. This worries me:my company works for private and public sec-tor clients and we find equal measures ofintegrity in each. And who thinks pensionsplans, insurance policies and bank statementsshould not be equally as clear as governmentforms or educational texts?

189



What of the term ‘transformer’? Offprints ofthis paper were much requested at the time,and the term was explicitly adopted in at leasttwo organisations that we know of, although itdoesn’t survive today. One was the OpenUniversity’s community education division,where we tried to use techniques (and pro-cesses) from journalism to help build readable,effective courses on parenting, health, and similar topics. The Open University’s Schoolof Business also used the transformer methodin some courses. Transformers were alsoappointed at the Natural History Museum inLondon, which in the late seventies movedaway from specimens in glass cases in favour ofinteractive experiences.

Although the article talks of the transformeras a single person, we were also describing aprocess, and perhaps a team. We used todebate this and I think that over the years,Michael and I have probably reversed posi-tions. He used to argue strongly that only asingle individual could keep track of the differ-ent facets of a particular job, whereas I used tothink it impossible for a single person to haveall the skills and qualities to be a transformer –understanding the content as well as thedesign, challenging the expert’s motives, stand-ing up for the reader, overseeing production.But now I think it equally challenging for ateam or institutionalised process to do thesethings effectively. Good transforming is oftendone instinctively and quickly, but teamsrequire issues to be articulated and minuted.And the reality of deadlines and budgets meanthis hardly ever happens.

At Information Design Unit we recruit peo-ple to be designers or writers. But the designersthink about the words, and often write, and thewriters use layout as an integral part of theirwriting. We don’t expect people to completelycross over, but it is always our hope that peo-ple will not stay entirely in the specialist role

assigned to them but will ‘break down the bar-riers in the interests of the reader...’ and ‘takeresponsibility for the success or failure of thecommunication.’ In other words, to be trans-formers.

Robert Waller

190



Postscript MM-R

Our article on the transformer certainly raisedissues – probably far too many! How influentialwas it? Hard to say. We probably struck chordsin readers who were already aware that tradi-tional roles such as designer and editor tendedto divide a process which required integrity.

The question as to how a person can developthe skills needed to be a transformer is crucial,yet hard to answer. There is no completely sat-isfactory kind of education yet available, andmuch of what we learnt was through our ownefforts. There is no end to the ways in whichpeople come to understand ideas, and sothere’s no end to the need for transformers togrow as communicators. In addition to linguis-tic and graphic skills, the capacity to think andanalyse is highly important. Looking back I cansee how the lack of some kind of standard education and training (and the lack of a per-sonnel category for transforming) has been an obstacle to inserting our ideas into largeinstitutions.

The article, written in our optimistic youth,scarcely touched the world of committees andorganisations, and imagined the transformer tobe a polymathic individual who had a kind ofover-riding control of the communication.Well, frankly, one is rarely in such a privilegedposition! The tendency for large organisationsto work by committee is certainly a greatobstacle to clarity and unity of purpose.Despite its limitations, the article made somekind of a splash in its time and contained someacute observations and examples. It was indeedstimulated by our work in the British OpenUniversity, which set out to deliver quality dis-tance education with a minimum of face-to-face contact. Since the OU succeeded (when somany expected it to fail) I might claim ourideas were right on target. But I won’t makeany such claim. The OU works because the stu-dents make it work. Perhaps the essence is toget people to think about their message fromthe viewpoint of the recipients.

Michael Macdonald-Ross

191



The Isotype chart in ‘The Transformer’

There is a story behind the Isotype chart weincluded in this paper. One of the few psycho-logical studies to address the effectiveness ofIsotype charts was Vernon (1946). Figure 2shows the chart she used in her tests, describedby her as an Isotype. It takes the table weshowed in ‘The transformer’, and simply sub-stitutes symbols for numbers, in centredcolumns that prevent the eye from making sim-ple comparisons. Not surprisingly, she foundthat the chart was less effective than a simpleline graph.

We felt that this study did a huge injustice toIsotype, failing as it does to represent either thegraphic quality of real Isotypes, or the qualityof graphic argument that the Neuraths rou-tinely built in to their work. This might nothave mattered, but the study is also featured inVernon’s very popular textbook The Psychologyof Perception (1962).

We traced the original government statisticsused by Vernon, and approached MarieNeurath, at that time retired but still workingin London, to design a chart from the samedata. Figure 3 was the result, and we repeat ithere for easier comparison. She simplified thedata and used a cohort of symbols movingacross the chart to show the main argument –that during wartime, the work force shiftedfrom less essential industries to the armedforces and war-related industries. We testedthe result successfully, although did not pub-lish the results – but we think the charts speakfor themselves.

Neurath’s chart illustrates both the strengthsand the limitation of the Isotype approach.The visual argument takes precedent over allelse. Because only three years’ data are neededto show it, the other years are omitted.Because she wants us to see (not just to under-stand) the way that employment shifted, she

forces the divisions between employment typesto move. So if the argument were different, anentirely different chart would be required. It isquite possible, then that the equivalent statis-tics from another era would be presented in away that would not be comparable with these.The transformer is not just presenting thestatistics, but telling us what to think. So thereis a thin line between Isotype and other kindsof distortions we are used to criticising in sta-tistical presentations (for example, the use ofinappropriate scales to flatten or exaggeratetrends).

The key issue for me is whether the readerknows the status of a chart – Isotypes do claimobjectivity and a statistical base, but becausethey discard the scientist’s white coat theymake their status clear to the reader, and invitecritical engagement. They are clearly argu-ments, not primary data.

Vernon, M.D. (1946) Learning from graphical material.British Journal of Psychology, 36: 145

Vernon, M.D. (1962) The Psychology of Perception.London: Penguin

192

Figure 1. Thegovernmentstatisticsused as thebasis ofVernon’s‘Isotype’chart.



193

Figure 1,reproducedfrom Vernon(1962).The caption,which was notan integral partof the chart,read ‘Chartshowingnumber of menaged 16-64years invariousoccupations’.

Figure 2. The same data diagrammedby Marie Neurath of the IsotypeInstitute.

Some things to note:• language is considered alongside

design – the large simple captiondefines the context in the chart isread

• each figure represents 1 million men– not 500,000. Isotype chartsnormally use a scale of 1:10, 1:100etc in order to make interpretationeasier

• the word ‘Year’ omitted because it isunnecessary – what else would1939 be in the context of World WarII?

• only three years’ data are needed toshow the argument, so the rest isedited out.

Macdonald-Ross & Waller - The Transformer Revisited

Documents

Transcript of Macdonald-Ross & Waller - The Transformer Revisited