Syntopical reading is reading several books on the same question side by side, so that you compare what each author says instead of absorbing one author's view. Mortimer Adler named it the fourth and highest level of reading in How to Read a Book. You pick a question, find the passages in each book that answer it, translate the authors into one shared vocabulary, and map where they agree and where they really disagree. The goal is not to decide who is right on the spot. It is to understand the whole argument well enough to hold your own position in it.
What are the four levels of reading?
Adler first published How to Read a Book in 1940 and rewrote it with Charles Van Doren for the 1972 edition, which is the one most people read today. It describes four levels, and each one contains the ones below it.
- Elementary reading. Knowing what the sentences say. This is what school teaches.
- Inspectional reading. Getting the most out of a book in limited time. It has two parts: systematic skimming (title, preface, contents, index, the opening and closing of key chapters) and a superficial first read straight through, without stopping at the hard parts.
- Analytical reading. Reading one book as thoroughly as you can: classify it, outline it, find its key terms and propositions, reconstruct its argument, and only then judge it.
- Syntopical reading. Reading many books on one subject and building an analysis that may not be in any of them.
Most summaries spend their space on level three, because it has the most rules. Level four gets a paragraph, often with a note that it is "beyond the scope" of the summary. That is a pity, since it is the level that matches how adults actually learn a topic: from several sources at once, over weeks.
How is syntopical reading different from analytical reading?
In analytical reading the author is the teacher. You work to understand the book on its own terms before you are allowed to criticize it. In syntopical reading the relationship flips. You bring a question, and the books serve your question, even when it is not the question their authors set out to answer. A chapter you would skip in an analytical read can be the only part of a book that matters syntopically.
That shift is why Adler insists you inspect every book on your list before reading any of them closely. Skimming tells you which books, and which chapters, deserve the slow read. Without it you end up reading one book analytically from cover to cover, which is a fine thing to do, but it is not syntopical reading.
What are the five steps of syntopical reading?
Adler lists a preparatory step and then five steps proper.
Before you start: build a bibliography and inspect it. Collect the books that may touch your question and skim each one. Drop the ones that turn out to be irrelevant. The rest is your reading list.
- Find the relevant passages. Not the whole book, only the parts that bear on your question.
- Bring the authors to terms. Authors use different words for the same idea and the same word for different ideas. You set one neutral vocabulary and translate each author into it, rather than adopting any single author's language.
- Frame neutral questions. Write a set of questions that most of your authors answer, openly or by implication. When an author gives no clear answer, record that honestly instead of guessing.
- Define the issues. An issue exists where authors answer the same question in opposing ways. Sometimes they only seem to disagree because they understand the question differently, and part of the work is telling those apart.
- Analyze the discussion. Order the questions and issues so the whole debate becomes visible, and quote each author in their own words so you can check you have been fair.
Adler adds a standard he calls dialectical objectivity: present every side without quietly taking one, through tone, emphasis or the order you put the answers in. You will not fully manage it, but trying keeps you from reading only for agreement.
What does syntopical reading look like on a real question?
The rules read abstract until you run them on something small. Say your question is does reading make you smarter?, and your sources are three books and two podcast episodes.
- Passages. One book has a chapter on vocabulary and general knowledge. Another has a single section on reasoning. A podcast guest spends ten minutes on it in the middle of a two hour interview. Those are your passages; the rest of each source is out of scope.
- Terms. One author means "smarter" as IQ, another as knowing more, the guest as being better at thinking through a problem. You fix three neutral terms: knowledge, verbal ability and reasoning, and translate each claim into one of them.
- Questions. Does reading raise knowledge? Does it raise reasoning? Does the format, print or audio, matter?
- Issues. On knowledge everyone agrees. On reasoning, two sources say yes and one says the evidence is thin. That is the real disagreement, and once the terms are fixed it is much smaller than it first looked.
- Discussion. You write half a page: where the sources agree, the one point they dispute, and what evidence each side offers.
That half page is the product. If you want to see where the real evidence on this question lands, we laid it out in does reading make you smarter, and the split between knowledge and reasoning is the heart of it.
Isn't syntopical reading too much work for normal readers?
At Adler's scale, yes. His own syntopical projects were institutional. A Syntopicon, published in 1952 as an index to the Great Books of the Western World, sorted the canon under 102 "Great Ideas" and took a team of readers years. Adler then founded the Institute for Philosophical Research, whose two volume The Idea of Freedom appeared in 1958 and 1961. That is not a model for someone reading on a commute.
The method scales down well, though. A syntopical read of three sources on one narrow question can fit into a few evenings. What makes it expensive is not the reading. It is step one, finding the passages again once you have finished the books, and that is where most people quietly give up.
Can you do syntopical reading with podcasts and audiobooks?
You can, and for many topics the best sources are spoken: long interviews, lecture series, debates. But nearly every tool Adler relies on is made of paper. Inspectional reading uses the table of contents, the index and the last pages of a chapter. Finding passages assumes you can flip back to a marked page. Audio has none of that by default.
Here is how each step translates when the source is something you hear.
- Inspect with what the show gives you: show notes, chapter markers, the episode description and the first and last five minutes. Skip a guest you can tell is off topic.
- Mark passages while you listen. You cannot underline, so you need a timestamp the moment a relevant claim is made. We wrote up how to take notes from podcasts without stopping for exactly this.
- Bring authors to terms on paper. The translation step needs writing in any format. Keep the neutral terms on one page and note which speaker said what, in their words.
- Quote, then check. Spoken arguments are looser than written ones, and memory rewrites them. Before you put a guest on one side of an issue, go back to the exact moment and listen again.
The habit underneath is the same one behind a commonplace book: passages collected by question rather than by source. Syntopical reading is what you do once enough of them pile up on one question.
What mistakes do people make?
- Starting without a question. "Read five books about leadership" is a reading list. "What do these authors say makes people follow someone without authority?" is a syntopical project.
- Reading everything analytically. Inspect first. Most of your sources only matter in a few chapters.
- Borrowing one author's vocabulary. Once you use the terms of your favorite book, every other book sounds wrong. This is the step that does the most work and the one summaries skip.
- Treating the result as a verdict. The output is a map of the debate, plus your reasoned place in it. If you finish with the same opinion and no new questions, you probably read for agreement.
Where Graspd fits
Graspd is an AI companion for nonfiction listening, and it covers the part of syntopical reading that audio makes hard: getting back to the passages. You set a mark with one tap while the episode plays, and the transcript around that moment is saved with it. Later you can search your marks across every podcast and book you have imported, by meaning rather than exact words, so "follow without authority" can find a guest who said "influence without a title". The steps in between, fixing terms and framing questions, stay yours. That is deliberate: the comparison is the part that builds understanding.
One limit we will not hide. Graspd works with podcasts and DRM free audio files. Audible and other DRM protected audiobooks cannot be imported by a third party app, and no workaround changes that.
FAQ
What is syntopical reading in simple terms?
Syntopical reading means reading several books or sources on one question and comparing them, instead of reading one book on its own. You find the relevant passages, put every author into the same vocabulary, and map where they agree and disagree. Mortimer Adler called it the fourth and highest level of reading.
What are Adler's four levels of reading?
The four levels in How to Read a Book are elementary, inspectional, analytical and syntopical reading. Elementary is understanding the sentences. Inspectional is skimming and a quick first read. Analytical is reading one book as thoroughly as possible. Syntopical is comparing many books on one subject. Each level includes the ones below it.
How many books do you need for syntopical reading?
There is no fixed number. Adler's own projects covered whole libraries, but the method works with three to five sources on a narrow question. Two sources is a comparison, not a discussion. The limit is usually your time for finding passages again, so start small and add sources only when they answer your question.
Is syntopical reading the same as a literature review?
They are close relatives. A literature review also compares sources on one question, but it usually reports what the field says. Syntopical reading asks you to translate the authors into neutral terms you choose, frame questions they may not have asked, and build an analysis that none of the sources contains on its own.
Can you read syntopically with audiobooks and podcasts?
Yes, but you have to replace the paper tools. Use show notes and chapter markers instead of an index, and set timestamped marks while listening instead of underlining. Before citing a speaker on one side of an issue, listen to the exact passage again, because memory of spoken arguments drifts quickly.