How to Read a Book

How to Read a Book by Mortimer J. Adler and Charles Van Doren is a classic text. I have emphasised many times how important reading is to the Polymath, and no source is better than this book for how to read a book. I will start with the summary for those who want a top line overview or those seeking a quick refresher

Stage 1: What is the book about?

  1. Classify the book according to kind and subject matter
  2. State what the whole book is about with the utmost brevity
  3. Enumerate its major parts in their order and relation, and outline these parts as you have outlined the whole
  4. Define the problem or problems the author is trying to solve

Stage 2: Rules for interpreting a book's contents

  1. Come to terms with the author by interpreting his key words
  2. Grasp the author's leading propositions by dealing with his most important sentences
  3. Know the author's arguments, by finding them in, or constructing them out of, sequences of sentences
  4. Determine which of his problems the author has solved, and which he has not; and as of the latter, decide which the author knew he had failed to solve

Stage 3: Rules for criticising a book as a communication of knowledge

A: General Maxims of Intellectual Etiquette

  1. Do not begin criticism until you have completed your outline and your interpretation of the book. (Do not say you agree, disagree or suspend judgement until you can say "I understand")
  2. Do not disagree disputatiously or contentiously
  3. Demonstrate that you recognise the difference between knowledge and mere personal opinion by presenting good reasons for any critical judgement you make

B: Special criteria for points of criticism

  1. Show wherein the author is uninformed
  2. Show wherein the author is misinformed
  3. Show wherein the author is illogical
  4. Show wherein the author's analysis or account is incomplete

Note: Of the last four, the first three are criteria for disagreement. Failing in all of these, you must agree, at least in part, although you may suspend judgement on the whole, in light of the last point

Let us now go through these points and analyse them in more detail. It should be noted that the above list is for non-fiction books, some adjustments will be made later for fiction.

Background to Reading

Most people can now read. Everyone reading this article can read, although I suppose text to speech might make that sentence a little funny. What is the purpose of reading though? Well, of course it is to communicate information. This alone is not enough though. I'm sure we've all known someone who doesn't really "know" much but just Googles random facts or uses a dictionary definition of some word to try and settle a complex argument. Mere recounting of facts is not enough for knowledge, but rather an understanding of how and why things are they way they are.

If we are to develop this deeper understanding, we need to engage in active reading. Active reading is not just the passive absorbing of information that most people do when reading a book but an active hunt for this understanding through the process outlined in the introduction and a constant questioning.

The process laid out in the introduction is dealing with analytical reading – that is, reading one book for everything it is worth. There are several layers of reading that build off of one another.

Level 1 is elementary reading. This is the first kind of reading taught in school to children, and while it begins young, I have met even some adults who have not really mastered the art of elementary reading! What I mean by this is it is not simply sound words but to competently comprehend the meaning of texts and read them at a decent pace. While this is introduced at a young age, it's probably not until the teenage years do most of us really come to complete this level of reading.

Level 2 is inspectional reading. This is about reading a book in the shortest time possible. Not to be confused with speed reading, where you try and read faster, this is about deliberately skipping some parts of the book in order to not get muddled down in the explanatory details. This process is sometimes called "skimming" and the aim is to grasp the most important points and arguments in the shortest time possible, although with the understanding that some important details, reasoning and background will be missed!

Level 3 is analytical reading. While inspectional reading is the best you can do in a very limited time, analytical reading is the best reading possible in an unlimited time. The book is read slowly and thoroughly. At this level we don't just want to scrap some facts but a deeper understanding from the book.

Level 4 is synoptical reading. This is like analytical reading but for many books rather than one. It lets us compare and contrast, to collect arguments and form our own on the subject. In a sense, this level of reading is about trying to take a cross section of an entire field and summarise it. We will return to this in more detail later.

More on Inspectional Reading

Inspectional reading is an important skill, and it is critical to perform as a part of analytical reading. I will assume you have mastered elementary reading – that is, you can "read" as it is commonly understood, so I will not detail that now.

To perform inspectional reading, you will begin by looking over the book. Read the title and subtitle carefully. Many people do not fully read those and often miss important details and clues as to what the book is about.

Read the blurb on the back (or sometimes inside the dust jacket). This might seem like fluff but normally this will literally tell you the most key ideas of the book in a paragraph or two!

Read the about the author if there is one. This will give you a little insight into why the author wrote this book, and usually serves to show that the author is credible to be writing this text. By the way, if this doesn't suitably convince you of the author's skill, you might want to move on. My Goodreads want to read list has nearly 2,000 items on it so I am not shy to skip a book!

Read the table of contents. This will give you an outline to the parts of a book. It is preferable when the book gives a very detailed table of contents, with not only chapter headings but also subparts of each chapter. I generally consider these to be better books if they have these detailed tables of contents. Skim the index too and see if you generally think you know what the terms mean. First, be suspicious of a non-fiction book that does not have an index. Second, do not be concerned at this stage if you do not know every word or the way in which the author means it. At this stage absolutely do not use a dictionary on each word in the index!

Skim through the references and see if you recognise any. This will be good to familiarise yourself a little with the sources. Sometimes sources are all at the end, at the end of each chapter or as footnotes at the bottom of each page only. I generally prefer books that have a bibliography at the end of the book or at the end of each chapter. Books that only have sources in footnotes or in the text itself are very hard to use as sources themselves, and frankly, I often pass them by. It's amazing how the same text published by two different companies that arrange a few things like the index and bibliography differently can radically change how useful a book is! It's a testament to how important a publisher is and not just an author. I will not that there are some non-fiction books to which I don't mind there being any references. A history, science, philosophy book or so on I absolutely expect some. Although, some programming books can get away without any, especially ones that talk more generally about organising the company. Scrum by Jeff Sutherland is a good example – he talks about how he had success with engineers in the past so references aren't exactly necessary.

Finally, move on to reading little bits of each chapter. The first and last paragraphs are always good. Any lists are also worth reading. Authors who make this part easy by placing the most important information in boxes, bullet points or summary paragraphs clearly marked are to be praised! I tried to make this article like this by putting a really clear summary of the whole thing in bullet points right at the start. Not all texts can be summarised and broken down in this way sadly, but it is nice to see.

The point of all of this is to gain the "gist" of the work. Main points, its purpose, why bother with it, a "feel" for it. It is much easier to get the "gist" first and then go back for a more detailed read than to try and get the detail and the "gist" at the same time. Actually, I think only one at a time can be gathered, which is why most people spend a long time reading a book in detail and only get the "gist" by the end.

Also, this phase is short – I'd say a fairly chunky and dense non-fiction can be inspectionally read and some key notes made in 30-120 minutes. At this point, the time dedication is not huge and you will have an excellent idea if it is worth continuing with reading the book in more detail. Sometimes it just isn't.

Active Reading

To actively read we are trying to answer some questions while reading. Not just any questions but four specific ones. These will flourish into the general outline we made in the introduction. These are the four questions you must ask yourself and answer for every book

  1. What is the book about as a whole? You must try to discover the leading theme of the book, and how the author develops the theme in an orderly way by subdividing it into its essential subordinate themes or topics
  2. What is being said in detail, and how? You must try to discover the main ideas, assertions, and arguments that constitute the author's particular message
  3. Is the book true, in whole or part? You can not answer this question until you have answered the first two. You have to know what is being said before you can decide whether it is true or not. When you understand a book, however, you are obligated, if you are reading seriously, to make up your own mind. Knowing the author's mind is not enough.
  4. What of it? If the book has given you information, you must ask about its significance. Why does the author think it is important to know these things? Is it important to you to know them? And if the book has not only informed you, but also enlightened you, it is necessary to seek further enlightenment by asking what else follows, what is further implied or suggested.

It is asking these questions that separates passive from active reading. Passive reading does not allow for a deeper understanding of texts, active reading does. Active reading allows us to develop new ideas, write about what we read, and I think when practice frequently, is sure to allow us to become polymaths.

Stage 1: What is the Book About?

Stage 1 of anaytical reading is about understanding what the book is actually about. First we would want to know what kind of subject the book is dealing with. This can be quite complex, since many books deal with several subjects at once. I think the best way to go about this is to over time develop your own system of classifying books. Is it simply a "history" book, or a "British history" book, or a "European history" book. Maybe there is a separation between "British history", "European history" and "Continental European history" or maybe not. You can see this, as a Brit, is quite personal to me. I'm sure a French person would think about this in a very different way. I try and think about books in relation to other books I have read – Niall Ferguson's Empire is, in my mind, related closely to Correlli Barnett's The Collapse of British Power.

An important classification that always matters is whether the book is "practical" or "theoretical". By "practical" here I do not mean that which works but rather a book that tells you how to do something or what to do. This article can be seen as a practical work: it is telling you how to read a book. Cook books are practical books, as are car repair books. Programming books are also usually practical books. Sometimes they really show off to you that they are practical by having "how to" in the title – usually a dead giveaway. Sometimes they are written in a very practical and instructive manner (like cookbooks). Sometimes, not quite so clear cut, programming books are often like that. By "theoretical" books here I mean books that are concerned with an explanation of some phenomena, either a description or a reason, but offer no course of action because of this. Mathematical and scientific books are often of the theoretical side (although they sometimes give a suggestion of where to research next). Some books are a mixture of practical and theoretical, and that's ok.

This all concerns rule 1: Classify the book according to kind and subject matter.

Moving onto rule 2: what the whole book is about with the utmost brevity. What this means is producing a paragraph to summarise what the book is about. I don't mean what kind of book it is, we already dealt with that. By "about" I mean the most major arguments.

Rule 3: Enumerate its major parts in their order and relation, and outline these parts as you have outlined the whole. In essence here we are trying to make a summary paragraph for each of the parts of the book in the same way as we made a whole book summary. Notice how I say "parts" and not chapter. A "part" is more like a logically contained unit – sometimes it might be a chapter, sometimes there may be several chapters in part, sometimes there may be several parts in a chapter. Some books have a mix of all three of these. In any case, do not be afraid to make your own logical parts of the book that do not follow the boundaries of the chapters.

By the time rule 2 and 3 have been put into practice, the book will have been "outlined". You can make these outlines very detailed and structured – giving page numbers to where the parts begin and end with headings and so forth, although for me this outline is usually a messy collection of draft notes that I sort out later when I come to write more in detail about a book.

Rule 4 is to define the problem or problems the author is trying to solve. Now that the outline has been made this should be easy. I suggest bullet pointing these, but don't worry too much if your thoughts on this change as you read the book. You may realise there are more, or that some merge into one. Some might not even be problems that are trying to be solved. This is all ok, the point of this method is to read a book iteratively. As an example, for this article you might say something like

There are many ways you could break this down, and none are "correct" or "wrong".

By the time you have finished inspectionally reading the book, hopefully all of these should be completed.

Stage 2: Rules for interpreting a book's contents

Now we are leaving behind the inspectional reading and moving onto a more detailed look at the book. The first thing to do is to come to terms with an author, dealt with in rule 5. Now we really need to understand the book, so as we read through it in detail we will want to make a glossary of important terms and how the author uses them. Different authors use the same term to mean different things, and sometimes use different terms to mean the same thing. It is important to square your past understanding of terms with what is in the current text or you will struggle to read it.

Rule 6: grasp the author's leading propositions by dealing with his most important sentences involves us pulling out the particular sentences that are key to the claims of the author. Right now, we are not looking to understand the reasons or arguments for these propositions, just to figure out what things the author is boldly claiming is true.

Rule 7: Know the author's arguments, by finding them in, or constructing them out of, sequences of sentences deals with the next logical part of rule 6. Now that we know what an author is claiming, figure out for what reasons he makes that claim. Sometimes this will be as easy as pulling out a key paragraph or two with a nice self contained argument for each one – the rest of the book will be explanatory material. Sometimes we will need to construct the argument in full by pulling little bits from across the chapter or even several chapters.

When we have finished reading the book and know the main arguments, we can move on to rule 8: determine which of his problems the author has solved, and which he has not; and as of the latter, decide which the author knew he had failed to solve. This is pretty self explanatory: match up the list of problems we made in rule 4 to the list of arguments and see what has been addressed. Sometimes the author will simply not answer a problem. Sometimes the author will acknowledge some reason why a certain problem can not be solved. Note that if the author sets out to solve a problem, but provides reasons as to why he can not, this does not make the work bad! In fact, this could be totally valid – perhaps future authors can solve those problems? Could it be you?

Stage 3: Criticising a Book

Reading and understanding a book is not quite enough. We now need to criticise it, in some ways we already began in rule 11. But now we really want to develop our own thoughts on the book. We will definetly only be able to begin this following a detailed read of the book, and almost certainly some flipping back and forth afterwards in order to fulfil rules 6, 7 and 8.

Rule 9 begins with this exactly: do not begin criticism until you have completed your outline and your interpretation of the book. (Do not say you agree, disagree or suspend judgement until you can say "I understand"). It is very important that we know exactly what we are agreeing or disagreeing with before we begin to do so. Also, don't be afraid to say you neither agree nor disagree, or only agree in part, or that you can not say whether you agree or disagree yet. Actually, in reality, you should seldom totally agree or disagree with an author.

Rule 10 is really about disagreeing politely: do not disagree disputatiously or contentiously. What this means is we don't really want to just call an author stupid or something along those lines, but disagree in a more professional manner. If you want other people to listen to your judgement, and especially those who agree with the author you are disagreeing with, this is critical.

Rule 11 then goes further on this by stating that you should demonstrate that you recognise the difference between knowledge and mere personal opinion by presenting good reasons for any critical judgement you make. I think this is rather self explanatory: it isn't really an argument to just state an opposing opinion but you must state why.

The final four rules deal with how to disagree.

Rule 12 is to show wherein the author is uninformed. To say the author is uninformed is to say the author lacked some piece of information relevant to the problem he was trying to solve. This may have lead him to an incorrect conclusion, or to conclude that the problem can not be solved when indeed it can.

Rule 13 is to show wherein the author is misinformed. To say an author is misinformed is to say that he states something that is not true, or has an incorrect interpretation of the fact. This will almost certainly lead to an incorrect conclusion at some point in the author's arguments.

Rule 14 is to show wherein the author is illogical. To say the author is illogical is to say that the conclusions do not follow from the premises. This is to say that the collected facts and assumptions are true, but taken together do not lead to the conclusion the author made.

Rule 15 is to show wherein the author's analysis or account is incomplete. This rule deals with when a work didn't reach its logical conclusion or leave something out. Of course, no work can get to the bottom of everything, as the book would become literally infinite – but what other things could the author have reasonably done? Could he have solved more of the starting problems? Could he have made better use of the materials? Did he fail to see all the ramifications?

Fiction Books

Now we have dealt with non-fiction, let's learn about reading fiction books. We will begin with some things not to do, and then think about what to do.

While you do probably want to scan the front and back matter of literature, don't read the first and last paragraphs of each chapter! This will spoil the book!

Don't try to resist the effect that a fiction piece has on you! Let it take hold of your imagination in whichever way it might. Fiction isn't trying to convey logically interrelated arguments but rather an experience – be it in the plot or the poetics of phrasing.

Also, don't look for terms, propositions or arguments. Not only are they generally not there, fiction relies on ambiguity in the terms. Poetry especially depends on the multiple meanings words can have. This is the exact opposite of non-fiction which needs to clearly convey some argument.

Mirroring the three stage process for non-fiction, fiction works also have a three stage process but with slightly different rules

First, beginning with outlining you will want to:

You will probably want to do this after you have read the work or while doing so.

Second, you will want to interpret the work

Finally, how do we handle being critical about fiction?

Synoptical reading

Synoptical reading is for when we want to take a good view of a whole topic. This is the kind of reading that university students often need to do but are never really taught. It is also the kind of reading polymaths need to do often. The goal of synoptical reading is to understand various perspectives on a topic and how they fit together. Let's break it down into steps like the others.

Step 0: identify the candidate books.

This can be done in all manner of ways. Maybe you already read a book analytically and now want to go deeper into that topic. Its references are the best places to start – and then the references of those books. Of course, those references may all be on the side of the author, so we might need to find opposing views. Finding books in general is much easier now than it was in the past thanks to the internet – many people have compiled lists on the best books in various subject areas. Libraries are also a useful place to go here. A librarian is not only someone who maintains the library, but an expert in helping people find information. Talk to them! They classify books all day long after all, and will probably be able to list of literally hundreds of books on the relevant subjects.

I'd say by the end of this step you would probably want, depending on the size of your project, maybe 10-500 works. 10 is probably the minimum number you want to consider unless the project is very short. 500 is an upper bound for very large projects, likely when you are writing your own book on the subject. Remember, these are candidates. Not everything will end up in the final piece. I assume by this point you could analytically read each of the books, but analytically reading 500 books could take literally years to do.

Step 1: Finding the relevant passages

At this point, we will want to start whittling down the books. Not all candidate books will actually contain everything useful. Some will need to be discarded. Inspectional reading is essential at this point. In some sense this is a little paradoxical – at this early stage you might not exactly know what it is you want to write about, since you don't know the subject, so finding the relevant passages could be difficult. What might be helpful is finding some super easy, short introductory article on the subject online. This article might not end up being used as a source in the final project, but allow you to get your head around the subject matter more easily.

You will want to walk through stage 1 of analytical reading by inspectionally reading the books. Sometimes this is enough – pulling out some key quotes and ideas will be all the book can offer. Some books it will become apparent will require an analytical reading before you can write your piece. Some books will be discarded from consideration. Most books will be in the first category for this: remember that if you are to ever write your synoptical piece you absolutely can not analytically read everything.

Stage 2: bringing the authors to terms.

In much the same way as we came to terms with authors in an analytical reading, we now need to bring the various authors to terms with each other. Once we have gathered the material necessary from each work, we should become consistent with terminology across all the authors. Different authors will use the same word to mean different things and different meanings for the same word. These must be resolved by creating your own glossary of terms with your own conventions and "translate" the various authors into that.

Stage 3: getting the questions clear

This step is perhaps the hardest one, but once complete will usually make writing your piece relatively easy. If you have ever looked at a flashing cursor on a blank document and not know what to write, it is almost certainly the case that you have not completed this step. Much in the same way as we listed the problems the authors wanted to solve in inspectional reading, what problems do we want to solve? Obviously, looking at the problems the authors we are reviewing tried to solve will help but usually we want to be more "meta" in synoptical reading.

Stage 4: defining the issues

Now that we have the questions or problems we want to answer, we should think about them as issues. What would each of the authors say about these issues? Some will be pro or con, some will be silent, some will give implicit answers so you are able to make arguments like "if he had talked about X, he would say Y". Summarising these different perspectives is in itself valuable.

Stage 5: analysing the discussion

In this step we are finishing the analysis by weighing in on the discussion. By taking the perspectives of all the authors we have looked at so far we should try and answer questions like who is right? Who is wrong? Did anyone answer the question correctly? If not, what is the correct answer? Obviously, there will be a lot of nuance – authors tend to be a little right and a little wrong, some more than others. Here you are weighing in and giving your own conclusions on the state of the field thus far.

Conclusions

I hope this guide gives you an outline of how to read. Not just how to read word by word but how to really get the most out of a text. I do think this article does not do the topic full justice – I have scarcely scratched the surface and yet it is already quite long. I think in future it would be good to dive back in and show some worked examples of how to actually do this. That's all for now, let me know if you found this guide helpful.