Barbara minto the pyramid principle pdf download vk

Data is just a bunch of numbers packed into a spreadsheet. What makes it valuable are the learnings within. So, in your reporting document, ask your team to discuss which factors impacted their performance the most and how they can act on the data at hand. For most of us, the Minto Pyramid Principle is the opposite of what our high school English teachers and college professors taught us.

But you must remember that they were paid to read our writing. In the real world, the only currency that people pay to read blog posts and internal wiki articles is their attention. So, start collecting that attention — by cutting to the chase. Skip to content Home.

Search for:. The Pyramid Principle will show you how to communicate your ideas clearly and succinctly. Barbara Minto reveals that the mind automatically sorts information into distinctive pyramidal groupings.

However, if any group of ideas are arranged into a pyramid structure in the first place, not only will it save valuable time and effort to write, it will take even less effort to read and comprehend it. Take the example from slide 5 and imagine you try to convince Marry that she has to learn English if she does not want to stay the rest of her life in Vienna. When talking to her, you might want to start with the premises before stating the conclusion.

You start with a problem, make a hypothesis and try to prove it. If it wont work, you have to be honest to yourself and change the hypothesis. Tim Berners-Lee might not have tried to prove the idea of the WWW when he first worked on his personal hypertext system Enquire in This writing style contrast to the chronological style of fiction, in which a story is told in the sequential order in which events occurred.

To illustrate, suppose you are holding a meeting for your field salesmen, at which you are planning to teach them how to present a new selling technique to chain grocery stores. However, in order to do so effectively you need some information from them on a particular problem chain in their local area.

How would you structure the introduction? In this case the question would be implied rather than stated, since the flow of the writing would not require it to be spelled out. Nevertheless, it is absolutely essential that you spell it out for yourself before you begin to write. Otherwise, you run the danger of not being absolutely sure of your question.

Note also that the Complication and the Answer are reversals of each other, since the Answer is the effect of carrying out the actions, which of course would solve the problem. From time to time needs updating. And again you have another question that would be implied rather than stated in writing.

Requests for funds Another very common memorandum type is one requesting funds. What data? In what form? Only the Situation would be stated in the writing, and then the pyramid structure would tell the reader to approve the purchase for some set of rel- evant reasons. For example: We should approve this request because: The cost will be more than offset by the projected savings It will greatly increase the Group's productivity It will create new opportunities for service. Step 1 Step 2 Step 3 However, the introductory structure varies slightly depending on whether you are telling the reader how to do something he has not done before or whether you are telling him how to do properly what he is already doing.

The trick here is to begin your thinking by literally laying out the present process as they do it now. See Exhibit Then lay out the process as you think it should be done. The differences between the first structure and the second tell you what the steps on your Key Line must be.

You may assume that you know pre- cisely what they are, having been working on them for so long. But unless you lay them out and compare them, the chances of leaving something important out are very great.

You cannot be sure your thinking is complete and that you have not left anything out unless you make the actual step- by-step comparison. I have seen many examples of incomplete thinking in this area that I make a special point of mentioning it. Indeed, we had an example in the Big Chief memo in Chapter 3. Letters of proposal These documents are the lifeblood of consulting, and have thus had a good deal of thought lavished on them over the years by consulting firms.

The Answer to the Question is always 'yes,' of course, followed by a 3-part structure explaining that: 1 We understand the problem 2 We have a sound approach for solving it 3 We have good people to work on it.

This is perfectly logical in concept. However, let me point out that, struc- turally, the only section that reflects thinking is the approach section.

The first section is an extended description of the Situation, playing back to the reader in organized form everything he has told you about how the prob- lem arose. Similarly, the final section, while true, is simply a listing of the qualifications of the people who will be assigned to the team. The major thinking is done in the approach section, where you specify the steps you will take never more than 5 to solve the problem.

These steps must be stated in end-product terms as explained in Chapter 9, and will serve to define the major phases of the study, with the specific end products to be expected at the end of each phase. It is on the basis of the approach that the client should make his decision to hire, although alas that is not always the case.

These are usually the formal com- munications one schedules with the client at the end of each phase of the study, leading up to the final report.

After the first one, the structure is always the same. The first one will say something like this: S You people have X problem. C We told you in our Letter of Proposal that we would do Y first to solve it. We have now done Y. Once this presentation has been made, the client will have a particular reaction. Perhaps he will ask you to investigate an anomaly you have uncovered in your work.

Or he may approve what you've done and tell you to move on to phase two. At the time of your next progress review, then, you might say something like this: S In our last progress review we told you that you had a capacity problem C You said you thought this would not be a problem long because you believed your competition was shortly going out of business. You asked us to investigate whether that were the case.

We have now completed our investigation. What must be apparent by now from these examples is that the pivot on which your entire document depends is the Question, of which there is always only one to a document. If you have two, they must be related: 'Should we enter the market, and if so, how?

On occasion you will not be able to determine the question easily just by thinking through the introduction. In that case, look at the material you intend to include in the body. Whenever you have a set of points you want to make, you want to make them because you think the reader should know them. Why should he know them? Why would that question have arisen? Because of his situation. So that by working backward you can invent a plausible introduction to give your question a logical provenance.

Transitions between groups 0 nee you have wn. In doing so, however, you want to make your progress from point to point seem smooth and nonmechanical. Thus, you do not want to say such things as: This chapter haslooked atthe need for priorities. The next chapter looks at how these priorities should be s. In other words, you do not want to relate what two chapters or sections do, you want to relate what they say - their major ideas. And you want to do it in such a way that you seem to be looking in two directions at once - back to what has been said and forward to what is to be said.

If you make this pause at the beginning of a chapter, section, or subsection, you should use the technique of referencing backward. If the chapters or sec- tions are long ones, then you will probably find it clearer to pause at the end and make a summary before going on. Referencing backward The technique of referencing backward consists simply of picking up a word or a phrase or the main idea of the preceding portion of the pyramid that you are linking, and using it in your opening sentence.

You are probably familiar with the technique in transitions between paragraphs. For example: No single executive has full-time responsibility for directing Group affairs, The absence of necessary leadership and coordination for senior operating and staff executives results in The problemsstemllling from lack offul!

Suppose you had just finished a chapter telling the Ritz-Ryan hotel chain that it was not taking full advantage of its common ownership of many hotel, restaurant, and catering operations. You are about to start a new chapter outlining the changes that would have to be made in the top executive structure if the Group is to be in a position to take advantage, and you have a pyramid like that shown opposite.

Ryan can take advantage' bfits G:ofnbineq resources. J shqrt, clear lines of authprity andresporisibHify. The point is to make the transitions unob- trusive yet clear, primarily through picking up the key word or phrase and carrying it forward. You are, of course, carrying it forward to connect with the major point of the next section, which has already been intro- duced briefly in the 'explanation' part of your original introduction. Thus, here you need not lead up to it with a 'story' as you did previously, since your reader now presumably has as much information as he needs to understand the points.

You do, however, need to introduce the grouping of ideas to come under this section, and explain how they support its major point. An example of doing this is at the end of the first section on page 49, where the conclusions about introductions are summarized. Here is the summary that appeared at the end of the Ritz-Ryan chapter we have just been discussing. In summary, the top-level executive structure recommended in this chapter consists of the Ritz-Ryan. Board and Chairman, a Group Managing Director, and three key executives reporting to him, each in charge of one of the Group's major businesses.

These positions and reporting relationships provide a strong framework for long- term leadership and control of Groupwide operations.

Only by streamlining the structure to provide this degree of control and accountability can the Group realize the improvement of opportunities identified elsewhere in this report. Concluding summaries of this sort are not difficult to write if you keep in mind that they are meant to restate, as adroitly as possible, the principal matter and tone of the preceding text.

Since you have these in front of you in your pyra- mid, all you are doing is pulling them together again for the reader. In all of this positioning, the intention is to make the job of thinking required of the reader as easy as possible.

He is, after all, rarely trained in analysis and reflection, and can have nowhere near the understanding of the subject you have even if the subject is his own company. You and he are not peers in interpreting your thinking on the subject. Thus, you must expect that his mind will not be precisely where you want it to be, in terms of understanding, as you finish one lengthy group of points and prepare to go on to the next.

Your transitions are meant to grab his mind, as it were, and pull it back to where it belongs if he is to comprehend what you are trying to say.

This is essentially an exercise in good manners, provided it is done gracefully and only where needed. Concluding Theoretically, if you write a proper introduction and structure the body of your document to obey the pyramid rules, you should not need a con- cluding statement.

You have, after all, clearly stated your reader's question at the beginning, and answered it fully with impeccable logic. The tendency to end short memos by saying, 'If you have any further questions, please do not hesitate to call,' no doubt reflects this need.

The obvious, perhaps too obvious, procedure at the end of a longer doc- ument is to signal the end by putting a line of asterisks in the middle of the page, which is sometimes called a 'sunset. Rather, you want to find a compelling set of words that not only sums up for the reader what you have been saying, but also produces an appropri- ate emotion in him about it. At least, that is Aristotle's advice about what to do in a conclusion.

That there is an 'appropriate emotion' for the end of a business docu- ment may be open to question, but I should think the major feeling you want to leave with your reader is that of a need and desire to act. Consequently, you want to give him some indication of what he is to think about or is able to do with the new knowledge he now possesses as a result of his reading.

This can take the form of either a philosophical insight or a prescription for immediate action. An airline president, for example, would probably be offended by strongly emotional statements when being urged to adopt a new planning system. In general, however, if you insist on appending a conclusion, you will want to write something that puts into perspective the significance of your message.

Here, for example, is the concluding paragraph of a report whose message was that it is technically possible to create a European- wide system for rapid retrieval by computer of technical literature. You will also have created a common market for information, one that makes available the full range of existing sources, not just national collections, to all users.

This could lead not only to advances in standardization and harmonization, but also to the development of totally new standards. We find the prospect exciting, and are eager to work with you in launching the pilot project. Simple pragmatism dictates that you do without. However, there is an occasion on which you will definitely need a concluding section, and that is when you are dealing with future actions.

Sometimes you will write a very long document that recommends a course of action that you think the reader is likely to take. If he takes it, there are some things he ought to do Monday morning to get things in motion. To house these activities, you create a section called Next Steps.

The only rule is that what you put in this section must be something that the reader will not question. That is, activities must be logically obvious ones. For example, suppose you are recommending that the client buy a com- pany, and you think that he is going to do so.

After 30 pages of explaining why you think it is a good idea, you assume you have him convinced. If, on the other hand, they were points that did raise questions in his mind, then you would have to include them in the body of your text, and make certain they fit horizontally and vertically with everything else you're saying.

Properly organized, these ideas will always form a pyramid, with the vari- ous levels of abstraction sorted out and related under a single thought. Ideas in the pyramid relate in three ways - up, down, and sideways. An idea above a grouping summarizes the ideas below, while these ideas in turn explain or defend the point above. At the same time, the ideas in the grouping march sideways in logical order. What constitutes logical order differs depending on whether the pyramided group was formed deduc- tively or inductively.

These two forms of reasoning are the only patterns available for estab- lishing logical relationships between ideas. Consequently, an understanding of how they differ and what their rules are is essential to being able to sort out your thinking and express it clearly in writing.

Briefly, the difference is as shown in Exhibit Deduction presents a line of reasoning that leads to a 'therefore' conclusion, and the point above is a summary of that line of reasoning, resting heavily on the final point. Induction defines a group of facts or ideas to be the same kind of thing, and then makes a statement or inference about that sameness. The deductive points derive from each other; the inductive points do not. These differences are really quite enormous, as the next two sections will demonstrate.

Once you have digested them, you should have little difficulty in recognizing or sorting out either form of reasoning, or in choosing the one that appropriately permits you to say clearly what you mean. Deductive reasoning. Deductive reasoning appears to be the pattern the mind generally prefers to use in most of its thinking, possibly because it is easier to construct than inductive reasoning.

But while it is a useful way to think, it is a ponderous way to write, as I shall hope to show. How it works First, let's understand what deductive reasoning is. It is usually described as taking the form of a syllogism - an argument in which a conclusion is inferred from two premises, one major and one minor.

I find these terms confusing in explaining how deductive thinking works in writing, and so I will not use them again. The second statement relates to the first if it comments on either its subject or its predicate.

Exhibit 18 shows several deductive arguments, each of which can be seen to do precisely these three things. And in each case the point at the top should roughly summarize the ideas grouped below, resting heavily on the final point. These are examples of deductive arguments in which each step of the reasoning has been included.

But sometimes you will find yourself wanting to skip a step and chain two or more deductive arguments together, since to put in every step would take too long and sound pedantic. This chaining of arguments is perfectly permissible, provided that your reader is likely to grasp and agree with the missing steps. Exhibit 19 gives an example of a chained deductive argument that should probably go something like this: We produce enough used newspaper to meet our own demand. But we have done more than meet our own demand.

Therefore we have a shortage. A shortage of used newspaper causes a shortage of newsprint. We have a shortage of used newspaper. Therefore we have a shortage ofnewsprint. They are boring, primarily because they make a mystery story out of what should be a straightforward point.

When to use it This leads me to urge that, on the Key Line level, you try to avoid using a deductive argument, and strive instead always to present your message inductively.

Because it is easier on the reader. Let's look at what you force the reader to do when you ask him to absorb a deductively organized report. Suppose you wish to tell him that he must change in some way. Your argument would look something like this: You must change Why?

Next you ask him to repeat the process, this time tying the first A of what is going wrong to the second A of what is causing it, and hauling the whole cartload to hitch to the third A of what to do about it.

And the same with the Bs and Cs. Not only do you make the reader wait a very long time to find out what he should do Monday morning, you also force him to re-enact your entire problem-solving process before he receives his reward. It is almost as if you're saying to him, 'I worked extremely hard to get this answer, and I'm going to make sure you know it. And now, while you may indeed have deductive arguments at the lower levels, still you have answered the reader's major question directly, with clear fences in your thinking between subject areas, and all information on each subject in one place.

To explain it another way, at the end of the problem-solving process, yo,u will have come up with a set of ideas that can be sorted onto a Recommendation Worksheet like that shown in Exhibit This permits you to visualize the fact that you have gathered findings that led you to draw conclusions from which you determined recommendations. In writing to recommend action, you will never give findings that do not lead to conclusions, nor state conclusions that are not based on findings.

The conclusions are, in fact, the findings at a higher level of abstraction. Nor will you have conclusions that do not lead to recommendations, nor recommendations that are not based on conclusions. One conclusion can lead to several recommendations, and several conclusions can lead to one recommendation, but there must always be a connection. The conclusions generally state the problem that the recommenda- tions solve. Consequently, the effect of the recommendation is to solve the problem you concluded was there.

The effect of the recommendation is to make our product competitive. The issue here is whether it is better to tell the reader why he should change and then how to go about it, or that he should change and why. As a rule of thumb, it is always better to present the action before the argument, since that is what the reader cares about, unless you face one of those rare cases in which it is the argument he really cares about.

Note that I am talking only about the Key Line here, and not about lower levels. Deductive arguments are very easy to absorb if they reach you directly: Birds fly I am a bird Therefore, I fly When, however, you must plough through 10 or 12 pages between the first point and the second, and between the second and the third, then they lose their instant clarity.

Consequently, you want to push deductive reasoning as low in the pyramid as possible, to limit intervening informa- tion to the minimum. At the paragraph level deductive arguments are lovely, and present an easy-to-follow flow. But inductive reasoning is always easier to absorb at higher levels.

Inductive reasoning d. In inductive reasoning the mind notices that several different things ideas, events, facts are similar in some way, brings them together in a group, and comments on the significance of their similarity. In the example of the Polish tanks cited in Exhibit 17, the events were all defined as warlike movements against Poland.

Hence, the inference that Poland was about to be invaded. If, however, the events had been defined as preparations by Poland's allies to attack the rest of Europe, a quite different inference would have been in order. How to do both things with precision is explained in considerable detail in Chapter 7. But at this point you need only understand the rudiments of how it is done to be able to distinguish the process from deduction.

This word will always be a plural noun a because any 'kind of' thing will always be a noun, and b because you will always have more than one of the 'kind of' idea in your grouping. And in each case again you can see that none of the ideas in any of the three groupings is a misfit; each one matches the description of the plural noun. The next step is always to check your reasoning, and this is done by questioning from the bottom up. For example, if you see a man who wants to found a city in which only Latin should be spoken, dig a deep hole in the center of the earth, etc.

Yes, you can, or at least you could when the statement was originally written. By contrast, consider the two examples in Exhibit If you see man- agers who don't face reality, won't countenance criticism, etc. Certainly not, it's sloppy reasoning and writing. What about the next one? If productivity is low, overtime high, and prices uncompetitive, can you infer that you have a profit-improvement opportunity?

Perhaps, but I can think of three or four other things that could also be labelled indicators of a profit-improvement opportunity. In that case, you know the overall point is at too high a level of abstraction in relationship to the three points grouped below, since it does not make a statement specifically and only about them. The low productivity led to the high overtime, which led to uncompeti- tive prices.

Whenever you have only one piece of evidence for anything, you are forced to deal with it deductively. Thus, the point implied at the top is something like 'Our prices are high because our productivity is low.

Remember, if you are thinking deductively, your second point will always comment on the subject or predicate of the first. If it does not so comment, you should be able to classify it by the same plural noun as the first, to test that you have a proper inductive grouping. To demonstrate, I recently ran across two so-called deductive fallacies in a logic book, which went as follows: All communistsare proponents of socialized medicine Some members of the administration are proponents of socialized medicine.

Therefore some members of the administration are communists. All rabbits are very fast runners. Some horses are very fast runners. Therefore, some horses are rabbits. In both cases, I'm sure you will instantly be able to see that the second point does not make a comment on the first point, so these ideas cannot be deductively related. What the second point does do in each case is to add another member to the classification plural noun established in the first point.

Placing ideas in classes is defining them by a plural noun, and you know that that is induction. To test yourself, suppose I say to you: o Japanese businessmen are escalating their drive for the Chinese market. Can you pick which of the next two points relates inductively to this, and which one deductively?

Note that with inductive ideas you generally either hold the subject con- stant and vary the predicate, or hold the predicate constant and vary the subject. This expectation of the mind for deductive and inductive arguments to be completed often leads the reader to project his thinking ahead, to formulate what he thinks your next point will be.

If his is differ- ent from yours, he can become both confused and annoyed. Consequently, you want to make sure that he will easily recognize the direction in which your thinking is tending by giving him the top point before you state the ideas grouped below. In doing so, you will naturally reflect the hierarchical structure of the pyramid, as shown in Exhibit You can reflect this hierarchy in a variety of ways, the most common of which are headings, underlined points, decimal numbering, and indented display.

Feelings run high about which of these is the 'best' formatting device. I myself lean to the use of headings as described below. However, in deference to what are excellent reasons given by proponents of the others, I discuss them as well. Whichever format you choose, remember that your objective is to make comprehension easier for the reader.

This means that the format must be applied properly to reflect the levels of abstraction in your argument. To give the desired appearance without the proper content can cause confusion. To this end, you want to make sure you thoroughly understand the rules before you begin application. Headings Essent1a. II y, t he tee hn1que. Iy subordinate ideas ever further to the right of the page Exhibit Thus, major ideas are capped with major section headings at the left-hand margin, divisions of these major ideas are capped with subsection head- ings, divisions of those with numbered paragraphs, and so on.

Of course, the style of headings you choose need not necessarily follow this particu- lar form, but whatever the form, each heading should represent a division of thought. Because your headings will represent divisions of thought, their use should reflect the relationships between the ideas inherent in the pyramid.

Since the headings indicate levels of abstraction in the pyramid, you can never have only one item at each level. Thus, you can never have only one major section, or one subsection, or one numbered paragraph, or one dash point.

Put more plainly, you shouldn't just stick in a heading because you think it would look good on a page, the way newspapers and magazines do, to break up the printing. A heading is meant to call attention to the fact that the idea it represents is one of a group, all of which are needed to understand the overall thought they support.

Since all of the ideas in a group are the same kind of idea, you want to emphasize this sameness by using the same grammatical form for the wording of each heading, etc. As you can see, because the subsection headings in the first group begin with the word 'To' does not necessarily mean that those in the second group must do so as well. Remember that there are invisible fences imposed between the ideas in each major section. Thus, the parallelism to be emphasized is between ideas in the subsection group, not between groups of subsections.

The headings are meant to remind, not to dominate. Thus, you want to make them as concise as possible. You would not want, for example, to make the first major section heading above read 'Appoint a full-time Chief Executive to provide clear central authority.

Headings are for the eye more then they are for the mind. As a result, they are not often read carefully, and you cannot depend on them to carry your message. Accordingly, you need to make sure that your opening sentence under a heading indicates that you are turning to a new topic.

In fact, your entire document should be able to be read as a smooth-flowing piece without the headings. By the same token, you should never use the headings as part of the text.

For example: Appoint a full-time Chief Executive This action will go far toward clarifying the day-to-day responsibilities of This rule, of course, does not apply to numbered paragraphs, which are meant to be read as part of the text. In doing so, you want to state the major point that the grouping will explain or defend, as well as the ideas to come.

To omit this service is to present the reader with a mystery story, since he will then not be able to judge what the point is you are trying to make in that section until he gets to the end - and by then he may well have forgotten the beginning. The paragraphs imme- diately following a chapter heading or title should express the major idea clearly, as well as supply whatever other information the reader requires to ensure that you and he are 'standing in the same place' before you make your point and tell him how you plan to develop it.

Subsequent chapter headings should be written in parallel style. A section can be further divided either into subsections or, if the points are short, into numbered paragraphs. This is a Subsection Heading These, too, should be worded to reflect the principal thoughts they cover, and expressed in parallel style. If you wish further to divide the thought in a subsec- tion, you can use numbered paragraphs.

This is a numbered paragraph. The first sentence or opening phrase can be underlined to highlight the similarity of the points being numbered. The point to be made may require more than one paragraph, but you should try to limit the development of the point to three paragraphs. Stars can be placed three in a row, in the center of the page, to indicate that a concluding comment to a long section is about to follow see above.

The paragraph point I can be used to set out lists when the number of items to be included is less than five for example, for the section headings listed above , or to call attention to a single paragraph that contains a point to be emphasized.

This is perhaps the most important rule of all. You want to use headings only if they are going to clarify your meaning - if they are going to make it easier for the reader to keep the subdivisions of your thought in his head. Often it is not necessary or useful to have any divisions below the major section headings. If you formulate your headings properly, they will stand in the table of con- tents as a precis of your report - another extremely useful device for the reader in trying to come to terms with your thinking.

Underlined points Another popular approach is literally to show the hierarchy of ideas by underlining the entire statement of the support points below the Key Line level Exhibit Lower level support points are also stated in their entirety, but distinguished by form and indentation. The purpose of this format is to provide speed and ease in reading. The theory is that the reader should be able to speed through if he wishes, reading only the major underlined points, and in that way comprehend the entire message.

Points below must directly answer the question raised by the point above, and no more. There is no room in this format for graceful liaisons of language or attempts at amplification.

Such things destroy the clean, stark presentation of the logic. If you must amplify or give background, you will have to do so in the introductory or concluding paragraphs. It destroys the ease with which the logic can be comprehended if the reader must wade through 30 words before he grasps the point.

If you find yourself with more than a dozen words, or more than one subject and predicate, think again. Most people ignore this requirement and end up simply listing points, without regard to the niceties of either induction or deduction. If you find yourself going beyond that, the likelihood is that you have overlooked an opportunity to group, and should rethink what you are saying. Again, if the section will be longer than seven para- graphs, state the points, centered, on the lines below, and then: 1.

Number in Parentheses. If the Document is Very Long. Number Without Parentheses. While this is lovely for the reader, it can be a bit difficult for the writer, because it imposes some strict rules on him. Many companies, and most government institutions, like to use numbers rather than headings to emphasize the subdivisions of a document, and some go so far as to number every para- graph. This approach is claimed to have the advantage that any single topic or recommendation can be easily and precisely referred to.

However, frequent index numbers do tend to interrupt the reader's concentration on the content of the document, or on any section of it, as a whole. If you decide that you prefer to have numbering because of its value as a quick guide, you would probably be wise to use it in conjunction with, rather than as a replacement for, headings. The headings have the value of enabling the reader to pick up the gist of the ideas quickly as he reads.

And they are quite useful in refreshing his memory if he finds he has to go back to the document several days after his initial reading. In addition, you will usually find that saying 'In Section 4.

Wh9t numbering system should you use? This one is very common: I. There is no other animal that will suffer to the death to aid its master as will a dog 1. Other animals will run when danger nears a.

The dog will remain i. Even though it might mean death. This one is probably simpler to use: 1. These examples show the relationships of the numbered levels to each other, rather than the actual form they should take. This is a constantly recurring question, and one to which more people come up with the wrong answer than any other. To start with, let us agree that the best talker is the most natural. He is easy, fluent, friendly, amusing and free from the fetters that seem to bind others to small pieces of paper.

He is just talking to us in the most natural way in the world: no script for him - how could there be? He is talking only to us and basing what he says on our reactions as he goes along. Such a talk cannot by definition be scripted. Our tongues are not so honeyed, and our words are less winged. And even for those who can on occasions touch those. Visuals A brilliant talker does not need visual aids to stop the audience from falling asleep, but the subject of a presentation very often demands them.

And if you have them, it can be fatal to depart from the prepared order in i which they are to appear.