John Lewis Gaddis’s “The Landscape of History”: Book Review, Notes + Analysis

Poor Ash’s Almanack > Book Reviews > History / Biography

Overall Rating: ★★★★★★ (6/7) (standout for its category)

Per-Hour Learning Potential / Utility: ★★★★★★ (6/7)

Readability: ★★★★★★ (6/7)

Challenge Level: 2/5  (Easy) | ~150 pages ex-notes (208 official)

Blurb/Description: Historian John Lewis Gaddis methodically asks – and answers – the question of how historians should think, and why it’s useful.

Summary: A friend recommended this book to me and I wasn’t sure what to expect; history has been a hit-or-miss category for me because for every Christopher Browning or Stephen Greenblatt who delivers profound and usable insights, there’s a Sir Walter Isaacson who gets bogged down in trivial, irrelevant details.

The Landscape of History turned out to be a delight: it’s not only a far superior introduction to history than the trite and predictable “Lessons of History” by Will and Ariel Durant, but it also, surprisingly, ends up doing what I wanted Peter Godfrey-Smith’s “Theory and Reality” (TaR review) to do: provide an overview of the thought process by which good historians analyze the past and bring the story back to the present.

Gaddis is a very clear and thoughtful writer.

Highlights: Gaddis is a big fan of the idea of “consilience,” which is basically another word for being multidisciplinary, and, in Gaddis’s words, remaining open to what insights from one field can tell you about another.” 

The Landscape of History is a delightful integration of a number of topics ranging from complexity theory to agency tocounterfactuals and probabilistic thinking; it’s very thought-provoking.

Above all, it manages to do that rarest of things: bridge the gap between being short and not too repetitive without losing the nuance of a longer, more thorough tome.  Very well done.

Lowlights: There’s little to dislike about this book – it’s very concise and thoughtful.  I do think at a few points Gaddis undersells the ability of the social sciences to discover enduring aspects of human behavior (like cognitive biases and our need for social connection).  Additionally, the end of the book sort of makes a sharp left turn into a new idea – history as being about oppression – that is taken for granted and not addressed in nearly the thoughtfulness of other topics.

Mental Model / ART Thinking Points:  disaggregationstorytellingschematrait adaptivityculture,tradeoffsconfirmation biasagencyrationalityprobabilistic thinkingfeedbacknonlinearity, emergence, utilitymargin of safety, a/b testing, mental modelsbase ratememoryn-order impactsbottleneckideology,

You should buy a copy of The Landscape of History if: you want a thoughtful, concise, thought-provoking exploration of how history intersects with other fields of study.

Reading Tips: none

Pairs Well With:

The Making of the Atomic Bomb by Richard Rhodes (TMAB review + notes).   The Making of the Atomic Bomb is one of my favorite history books for a lot of reasons; it’s long but worthwhile, providing a detailed look at not only scientific and engineering approaches, but also doing the “zooming in and out” thing Gaddis discusses to provide just the right amount of historical context.

Ordinary Men by Christopher Browning (OrdM review + notes).   Ordinary Men is probably the best history book I’ve ever read, and certainly shorter than TMAB: it epitomizes Gaddis’s dictum to show empathy but not sympathy, as well as using the right level of detail to create a map (while remaining counterfactual).

Scale by Geoffrey West (SCALE review + notes).  Gaddis consistently discusses complexity theory and emergence, noting power laws, fractals, and self-similarity; Scale is the best introduction to these topics I’ve encountered: it’s remarkably clear and West does a phenomenal job of helping the reader learn, rather than confusing them.

Superforecasting by Philip Tetlock (SF review + notes).  Gaddis discusses the idea of counterfactuals and probabilistic thinking quite a lot; I think Tetlock is one of the best introductions to this topic.

Misbehaving by Richard Thaler (M review + notes).  Gaddis discusses behavioral economics as an example of consilience and “good” social science… Misbehaving is unquestionably the best book out there on behavioral economics, by a very wide margin.

Reread Value: 4/5 (High)

More Detailed Notes + Analysis (SPOILERS BELOW):

IMPORTANT: the below commentary DOES NOT SUBSTITUTE for READING THE BOOK.  Full stop. This commentary is NOT a comprehensive summary of the lessons of the book, or intended to be comprehensive.  It was primarily created for my own personal reference.

Much of the below will be utterly incomprehensible if you have not read the book, or if you do not have the book on hand to reference.  Even if it was comprehensive, you would be depriving yourself of the vast majority of the learning opportunity by only reading the “Cliff Notes.”  Do so at your own peril.

I provide these notes and analysis for five use cases.  First, they may help you decide which books you should put on your shelf, based on a quick review of some of the ideas discussed.  

Second, as I discuss in the memory mental model, time-delayed re-encoding strengthens memory, and notes can also serve as a “cue” to enhance recall.  However, taking notes is a time consuming process that many busy students and professionals opt out of, so hopefully these notes can serve as a starting point to which you can append your own thoughts, marginalia, insights, etc.

Third, perhaps most importantly of all, I contextualize authors’ points with points from other books that either serve to strengthen, or weaken, the arguments made.  I also point out how specific examples tie in to specific mental models, which you are encouraged to read, thereby enriching your understanding and accelerating your learning.  Combining two and three, I recommend that you read these notes while the book’s still fresh in your mind – after a few days, perhaps.

Fourth, they will hopefully serve as a “discovery mechanism” for further related reading.

Fifth and finally, they will hopefully serve as an index for you to return to at a future point in time, to identify sections of the book worth rereading to help you better address current challenges and opportunities in your life – or to reinterpret and reimagine elements of the book in a light you didn’t see previously because you weren’t familiar with all the other models or books discussed in the third use case.


Pages x – xi: Gaddis views his approach here as the intellectual heir to The Historian’s Craft by Marc Bloch and “What Is History? by E. H. Carr.  

Both Bloch and Carr disliked the “methodological modesty” of leaving the process of historians behind closed curtains; Gaddis notes that this:

 “too often confuses our students – even, at times, ourselves – as to just what it is we do.”  

Gaddis aims to clear this up.   Disaggregation.

Page 1: The image on the cover of the book is an 1818 painting by Caspar David Friedrich: The Wanderer above a Sea of Fog.  Gaddis uses it as a metaphor; it’s a:

“fog-shrouded landscape in which the fantastic shapes of more distant promontories are only partly visible […]

it’s impossible to know whether the prospect confronting the young man is exhilarating, or terrifying, or both.”

It reminds me a bit of this 1900 painting of Askeladden (the character in Norse Mythology) by Theodor Kittelsen.  My friend Todd was nice enough to get me a print when I launched ACM:

Th. Kittelsen: Askeladdens eventyr. Image source: Wikimedia Commons (no copyright)

Pages 2, 3: Some of Gaddis’s key themes are the idea of metaphor and representation.  He notes, later, that we can’t literally represent something in its entirety, as to do so would be a useless 1-to-1 map that would become the thing itself.  

While Gaddis sees value in making history more scientific, he also sees value in storytelling and narrative, questioning “physics envy” (as we’ll see later on).

Page 5: One of the benefits of studying history is that it “let[s] us experience vicariously what we can’t experience directly: a wider view.”  

Reminds me a bit of Munger on “the eminent dead.”  See also Shawn Achor on why Yale medical students go to art museums in “ Before Happiness (BH review + notes).

Pages 6x, 7: One of the few strange / unintelligible things Gaddis does is define maturity as “the arrival of identity by way of insignificance.”  Huh?

He basically implies that as we grow older, we (should) lose our sense of uniqueness or specialness. I agree in some senses but disagree in others.  

He does note that as a historian, you “feel large” because you’re in charge of selecting the details that other people will see.  (Gaddis later notes that history is always influenced by the schema of the historian.)

Page 9: As with many others – Feynman in “The Pleasure of Finding Things Out” (PFTO review + notes) or Gonzales in Deep Survival (DpSv review + notes) – Gaddis notes, via Carr, that the effectiveness of human thinking:

“has been multiplied many times by learning and incorporating… the experience of the intervening generations.”  

In other words, culture (the acceleration thereof a product of exponential growth and network effects) is our strength.

Page 11: Gaddis here brings up the context-dependency of trait adaptivity, noting that “generalizations do not always hold in particular circumstances,” so the point of studying history isn’t to mimic exactly what happened, but rather to build up a database of experience to hopefully increase wisdom.

Page 15: Gaddis notes the inherent tradeoff in detail: the more of it there is, the less time you can cover.  Striking the right balance is hard (and most history books usually tend towards the wrong side of the balance, in my view.)

Pages 19 – 20, 23: Gaddis continues this train of thought (with a metaphor, of course) and notes that historians have license to “give greater attention to some things and others.”  He notes, however, that “these procedures are so basic that historians tend to take them for granted” – they deserve more thought.  Historians select the significant details.

What are the consequences of not thinking about these things?  Gaddis points out that some historians – Macaulay and Adams – display notable confirmation bias; Gaddis notes wryly that the Adams family can’t help but view the world through the schema of the Adams family.

Pages 24 – 25: Gaddis notes that historians have (and should utilize) their ability to stand apart from events, and thereby compare them to others, even if those others are far-removed geographically or chronologically.  Historians also get to view things on both a micro and macro scale.

Pages 26 – 28: Gaddis notes the interesting quandary that you can’t literally look at the “entire” set of details, and indeed there is no “entire” set of details because you can always go to a finer resolution.  

He notes the length-of-coastlines issue discussed in some more depth by Geoffrey West in “Scale” (SCALE review + notes).

Page 29: Gaddis discusses time here briefly, noting that the difference between the future and the past is that the future is unknown and the past is fixed; the present is the intersection of the two.

Pages 31 – 33: Gaddis here brings up the idea of mental models: he notes that we can’t literally represent reality; instead we need to distill the critical elements.  He also notes, insightfully:

“There’s no such thing as a single correct map.  The form of the map reflects its purpose.”  

Nice bit on rationality  utility.  He provides the examples of a highway map not needing to note vegetation; he also notes that a globe wouldn’t work for a road trip, but would for a plane flight

He calls history a sort of mapmaking.

Page 35: One difference between historical and geographical maps is that we can’t actually visit history the way we can places.  He also notes that structures and processes provide the “mechanism” for history.

Page 37: Is history a science, and should it be?  Gaddis notes that one of the impressive elements of science – that is not usually seen in other fields – is the near-universal agreement on important principles.  He notes that airplanes have the same aerodynamic experience no matter where they’re made…

He briefly mentions haruspicy, which made me think of Jordan Ellenberg’s amusing “How Not To Be Wrong” (HNW review + notes)).

Page 38: Gaddis notes that agency means that human behavior can’t be predicted with the precision of atomic behavior; Gaddis cites political scientist Stanley Hoffman, who observed humans are not “gases or pistons.”  

This is also sort of an emergence phenomenon, as we’ll get to.

Pages 39 – 41: Gaddis makes a really interesting point here, separating the sciences into those where effects can be directly observed (think basic acid-base chemistry), and those where they can only be simulated and recognized because the time or space scales are too long – think geology or astronomy.  

Or even biology; Gaddis notes that evolution and natural selection work on far too long a timeframe for us to observe it happening directly. That doesn’t make them any less true. (This is one of the more profound and insightful parts of the book.)

What to do, then?  Both historians and scientists in these fields must start with surviving structuresand “deduce the processes that produced them.”  Complicating matters is that we don’t have a perfect record of the past; evidence is sometimes contradictory, and a lot gets lost.   Probabilistic thinkingand  disaggregation / multicausality.

Page 42: Gaddis cites Jared Diamond’s “ Guns, Germs, and Steel ( GGS review) as an example of this sort of historical work.  I think GGS has a lot of innovative points but ultimately suffers from storytelling and confirmation bias and is generally vastly overrated.

Diamond conveniently ignores pretty much everything post the Industrial Revolution, which is the substantially more interesting part of history because it’s much more directly comparable to the present than previous eras.  Stephen Greenblatt’s “Swerve” does a better job with that…

Pages 45 – 46: Gaddis goes back to maps again: he notes that maps are evaluated not necessarily relative to accuracy, but rather to utility.

Page 48: Gaddis doesn’t espouse pure deduction, but rather an iterative loop of feedback wherein curiosity leads to reading which leads to redefining the problem which leads to more reading.  Is this scientific? Yes, apparently this is how physics is done.

Pages 53 – 55: Again here, see West’s “ Scale ( SCALE review + notes).  Gaddis discusses his skepticism of the idea of independent variables; his view is that the world is inherently multicausal.

He starts to get into complexity/chaos theory here; he contrasts the “reductionistic” one-to-one linear view of the world with the nonlinear “ecological” view of the world… he also notes the “sand pile effect” (in other words) – i.e., that systems can behave in emergent ways, and the variables that are important at one level are not necessarily important at another level.

Pages 57 – 58: Gaddis here discusses the tendency of the social sciences to look for idealized platonic aspects of human behavior that can be used for prediction/forecasting; Gaddis seems to make the point that much of this is context-dependent.

I think he doesn’t really do a good enough job of acknowledging that there are, in fact, some durable characteristics; psychology has proven out a lot of them (all the cognitive biases, for example).  Nonetheless, he makes a good point.

Pages 58 – 60: Gaddis cites ideology and (sort of) man-with-a-hammer syndrome: in a lot of situations, for ease of calculation, social scientists make simplifying assumptions like rational actors (cough humans vs. econs cough).  This makes their models neat and clean, but not very useful.

Pages 61 – 62: Gaddis here and elsewhere uses what he terms “consilience,” which is a fancy word for being multidisciplinary – which, in his words, involves “remaining open to what insights from one field can tell you about another.”  He cites behavioral economics as an example of this.  See, of course, Richard Thaler’s phenomenal “ Misbehaving” ( M review + notes) for more.

Page 63: Gaddis again here notes the nature  of trait adaptivity and context-dependency; he cites a scholar named “Collingwood” who alleges that anyone studying “the permanent and unchanging laws of human nature” will mistake “the transient conditions of a certain historical age for the permanent conditions of human life.”

Well, yes and no; again, I don’t think Gaddis fully explains here.  He is half-right – for example, the effects of the Great Depression on a certain generation, and World War II on the Greatest Generation, cannot be overstated.  

On the other hand, psychological research has demonstrated a lot of enduring aspects of human nature – not just cognitive biases like salience and stuff like hedonic adaptation ( contrast bias), but also, for example, the need for social connection.

Pages 64 – 65: Gaddis goes on to make some very good points here on multicausality and counterfactuals / probabilistic thinking.  He glorifies “it depends” (Tetlock would be proud – see Superforecasting (SF review + notes) for more).

He does note, however, the idea of 80/20 / marginal utility and proximate causation: while you could go all the way back to the start of agriculture to explain Stalin’s Soviet Union; that doesn’t make much sense, similarly, more emphasis is placed on Truman’s decision to drop the bomb than on the men who carried it out.

(The story of those men is interesting, though.  See Richard Rhodes’ wonderful The Making of the Atomic Bomb (TMAB review + notes) for more.)

Pages 67 – 68: Another example of schema / confirmation bias and perspectives: Gaddis notes that some Cold War historians focused exclusively on military, rather than economic, power.  Gaddis provides a great example of a triceratops that (in my view) ate too much bacon and died of clogged arteries. (Read the book and you’ll get it…)

Page 69: Gaddis cites how physicians should practice as an example of not focusing purely on independent variables.  

See Groopman’s “ How Doctors Think ( HDT review + notes) for more on how this actually works… and, sometimes, doesn’t.  Groopman notes, for example:

Most people assume that medical decision-making is an objective and rational process, free from the intrusion of emotion […] yet the opposite is true.”

The book extensively discusses why doctors often miss  multicausality, and how other cognitive errors ( recency bias for example) can cause them to miss diagnoses.

Pages 72 – 76: Gaddis here provides an excellent overview (via the metaphor of traffic) of complexity/chaos theory and nonlinearity driven by feedback between linear systems.

Pages 77 – 78: Gaddis here discusses the idea of discontinuities ( critical thresholds) and how complexity can emerge without having been put there.  Again, West…

Pages 79 – 81: Gaddis explicitly mentions the sand pile metaphor here (see Gonzales), and notes that complex doesn’t mean complicated.  He brings up the idea of path dependency.

Gonzales’s Deep Survival (DpSv review + notes) provides an accessible introduction to some concepts from complexity / emergence, in the context of survival situations.  On the sand pile experiment, he notes:

“nothing in the physics of silicon dioxide that could predict the behavior of the sand pile.”  

This is translated to powerful nonlinear forces like avalanches, undertow currents, etc that can easily kill unsuspecting city slickers like you and me.

Pages 81 – 82: Fractals!  See Geoffrey West’s “ Scale ( Scale review + notes).

Pages 84 – 86: Gaddis brings up the idea of power laws… again, see West.

Page 87: On resilience ( margin of safety): Gaddis notes that there’s a tradeoff between predictability and unpredictability of the environment.  No unpredictability, and you’ll develop traits that are too narrowly adaptive; too much unpredictability, and you can’t learn anything (there’s no consistent/clear feedback).

Gaddis cites Waldrop’s “Complexity,” which I finally broke down and ordered.

Page 88: Here is that classic quote about a physicist astonished by economists’ mathematical prowess and, of course, complete divorce from reality… math for the sake of math.

Page 89: Gaddis does not have physics envy.

Pages 94 – 99: Can multicausal situations still be analyzed deductively?  Yes – while the Swiss Cheese model may be true, some causes are more important than others.  Gaddis cites Bloch focusing on the proximate cause as being the “last… the most exceptional.”

Gaddis goes back to the idea of “how far to go back” – not that different from what he said earlier about Stalin, just more developed.  He concludes here with the idea of “necessary” vs. “sufficient”causation… he talks about the idea of critical thresholds and points of no return.

Of course, the proximate cause might not be the easiest one to fix… see Don Norman’s amazing “The Design of Everyday Things” (DOET review + notes).   

Norman explores how ‘human error’ is often viewed as the proximate causes of failures, but it’s immature and impractical to wave your hands and try to magically eliminate human error.  Rather, the system should be analyzed and redesigned:

It’s not possible to eliminate human error if it is thought of as a personal failure: if the system lets you make the error, it’s badly designed. If the system induces you to make the error, it’s really badly designed.

Pages 101 – 102: Gaddis brings up the idea of counterfactuals and controls / a/b testing.  He notes that some counterfactuals have more utility than others.

Pages 105 – 106: How do you pick the important details?  Napoleon’s underwear is frequently cited as an irrelevant detail (Walter Isaacson apparently did not read this book).  Gaddis also notes, again, the idea of context-dependency of certain lessons.

Pages 108 – 109: Gaddis sort of advocates the “ mental models” approach – instead of being a “Hedgehog” trying to fit the world through one explanatory paradigm, he notes that different paradigms work well in different areas, and it works better to have most of them.

Pages 111 – 112: Again, Gaddis cites agency and self-reflection as a key differentiator between humans and animals: animal behavior is predictable in aggregate.  Human behavior is not.

He notes the non-homogeneity of utility (which, among other things, blows up the efficient market hypothesis).  Again, see “ Misbehaving” ( M review + notes) for more.

Page 113: Gaddis notes an interesting dichotomy between “objects of inquiry” and “methods of inquiry.”

Page 114: Gaddis compares the challenges of historians to that of biographers, noting that people’s minds, even if they’re alive, are in some senses as inaccessible as historical landscapes.

Pages 115 – 117: Gaddis notes the challenge of using generalizations to draw conclusions about individual situations; they are (I would point out) no more than a base rate.  He tries to quantify the idea of “character,” comparing it to the self-similarity of fractals.

Page 119: An interesting aspect of history is that people who weren’t important in their time can become important to us by the simple accident of having surviving memories…

Pages 122 – 125: Gaddis asks whether we should try to escape thinking about history in moral terms.  I think Browning does a great job of answering this in “Ordinary Men (OrdM review + notes).

He differentiates between empathy – trying to understand the world from someone else’s perspective – and sympathy, i.e., agreeing with their perspective.  He also talks about the idea of “fitting” (here and elsewhere) and notes that as cultural sensibilities change, so will the fit.

Page 128: Gaddis believes empathy requires metaphors, because we’re humans, not econs.  See  salience and  storytelling.

Page 131: An interesting/amusing anecdote about a highway in North Dakota that makes a sharp 90 degree turn to account for changing longitude.  Gaddis notes the desire of states everywhere to make their territories legible/controllable; from this we get surnames.

Page 136: Gaddis notes the challenges of human memory here: our representations, accurate or not, become our reality over time.

Pages 140 – 143: Gaddis notes our tendency to evaluate historical events through the lens of the present – in other words, a schema bottleneck.  He also cites the idea of n-order impacts by presenting the ecological example of disrupting an ecosystem.

Pages 145 – 146: Gaddis brings up the idea of history as helping us contextualize that we’re not alone, although he notes the risks of ideology in doing so.


First Read: spring 2018

Last Read: spring 2018

Number of Times Read: 1

Planning to Read Again?: maybe


Review Date: spring 2018

Notes Date: spring 2018