While reading prefaces and afterwords, I found myself wanting to analyze what people think when they lose a war.
By the time I had read every preface and afterword, the reading notes came to 892 entries. After sorting and merging them and adding tags, the result is the published *Prefaces and Afterwords*. Using that as the base, I analyzed "what people think when they lose a war."
But as noted above, these results pass through the filter of my own subjectivity; they are not objective data. By analyzing that subjectivity objectively, I can see what the data I made actually is. On that basis I decided to think about what people think when they lose a war.
Counts of Emotion Words
The number of notes containing at least one word from that category. A single note can fall into several categories, so the total exceeds 883.
"Expectation and hope" at 122 is 3.3 times "self-reflection and self-blame" at 37. The forward-looking words together come to 3.4 times the introspective ones.
This survey began as a way of checking whether anyone, right after the defeat, was reflecting on Japan's responsibility as an aggressor. In other words a bias toward finding self-reflection must have been at work in me without my noticing, and yet the forward-looking words still dominate — which must mean something. The result was nothing like what I had foreseen.
Source: a recount of the 883 reading notes (2026-08-05). The word list uses only the "sample keywords" left in the body text, so it differs from the table of 2026-05 (base 874; expectation 142 / self-reflection 39). The two cannot be compared directly.
The Ratio of War Words to Reconstruction Words
A note containing both is counted on both sides. "Total" = the total number of times the words appear; "entries" = the number of notes containing that word group. The two are different scales, so they are shown as separate charts.
Total occurrences
Total ratio about 1:1.13. Reconstruction words are slightly ahead.
Entries in which it appears (out of 883)
For both word groups, 60% of the notes touch on neither.
But think about it: a writer who used a personal connection to evacuate to the countryside could not eat or drink well, yet some of them felt little hardship beyond boredom. Naturally there are degrees in the experience of war and defeat. Even in a defeated country, not everyone is thinking about the war.
Another thing: depending on the life a person has come through and on their makeup, some cannot interpret their own experience or accept it. A shocking, cruel experience does not necessarily change a person.
Breakdown (totals; both columns on the same scale)
War and defeat words
Reconstruction and building words
"War" at 309 stands out on its own. But on the war side the top three words (war, defeat, the end of the war) take almost everything, while on the reconstruction side democracy, construction, peace and freedom stand side by side. The first are names of events; the second are names of policies.
Source: as above (recounted 2026-08-05). The word list is the full version given in this page's body text.
Counts by Tag
Compound tags are split and each counted as one (up to 4 tags per entry; 344 entries have two or more). The tags were applied by the reader; they are not a contemporary classification.
What struck me while tagging was how much hope was placed in children. I also felt a strong will to build an environment for learning. A third was the view that every citizen ought to think scientifically, and these show up in the results. And there was a firm resolve that from now on it would be democracy.
These three are connected. Establishing democracy and building a nation on science are both indispensable to reconstruction. Spreading democracy and science requires education. The ones who learn are the children, and so it leads to hopes for the children — that is, to believing that children will grow.
"We Were Right All Along" is where people who had been suppressed during the war wrote things like, "I was writing this before and during the war. I was right. From now on it is my era." I remember reading a fair number of these, but the count is smaller than I expected. There were a few written purely for self-protection, which I dislike, and that is probably why they stuck in my mind.
The numbers the text contrasts
Indifference was a surprise. War and defeat are enormous events. My assumption was that nobody could avoid touching on them. But indifference — people who never mention the war — was far more common than I expected. Some simply felt it had nothing to do with them; others feared the risk of raising the subject. The fact that "Discontent with the Occupation" appears only once carries the same kind of information.
Source: a recount of 883 reading notes across 35 tags (2026-08-05). The colors follow the tags the body text uses for its contrast; the tags themselves carry no forward-looking / self-reflective attribute.
Frequent Two-Character Compounds
Morphological analysis with fugashi + unidic-lite, extracting two-character nouns (numerals, pronouns and proper nouns excluded). Counts are total occurrences.
The faint rows (this book, the author, oneself, today) come from the form of a preface as a genre; there is no point in reading content into them.
Against 127 occurrences of "freedom," "equality" appears 4 times. In prewar Japan it was equality that was prized and freedom that was neglected. This ratio is what that distortion looks like after the defeat turned it over.
My own instinct says equality should win by a mile, so I am not quite convinced. But this is objective data, so I have to accept it.
That science and daily life rank above democracy is a little unexpected; what astonished me was how much freedom appears. Why so many people wanted to write about freedom rather than equality in their prefaces and afterwords, I do not know. If I had to hypothesize: "equality" was too self-evident to be worth mentioning, and on meeting the defeat people accepted "freedom" gladly, if warily. Something like that.
Source: the morphological analysis as of 2026-06. Its method differs from a simple string count, and against the current reading notes a few words are off by ±1 (war 309, democracy 136, science 373).
Was There Anything That Caused the Reconstruction?
Reading the prefaces and afterwords, I wondered whether the causes of the recovery were there in the feelings written into them.
The standard answer to "how did Japan recover?" runs as follows.
External and structural tailwinds
The special procurement of the Korean War (1950–53), and the foreign currency and demand generated by U.S. military purchases of goods and services, lifted the economy at a stroke. Bank of Japan governor Ichimada Hisato is said to have called it a "divine wind."
Behind that lay the Cold War. The United States positioned Japan as a bulwark against communism and supplied security, market access and technology transfer, which let Japan hold defense spending down. (The Yoshida Doctrine.)
Add to that the fixed exchange rate of 360 yen to the dollar, favorable to exports, and an international environment of cheap crude oil and expanding world trade.
Institutional and policy causes
Another standard explanation: the Occupation reforms — land reform, the dissolution of the zaibatsu, labor reform — broke up landlordism and zaibatsu control, widened the domestic market and stimulated competition.
The Dodge Line pushed inflation down and forced companies to rationalize and accumulate capital, while government industrial policy nurtured heavy and chemical industry.
The mechanism by which high household savings flowed through banks (the main bank system) into capital investment — "investment calling forth investment" — is also told over and over.
Legacies from before the war
The air raids destroyed the plant, but the accumulated technology, the organizational capability and the relatively high level of education remained. Military technology was converted to civilian goods. The abundant young workforce (the demographic dividend) belongs here too.
Cultural explanations
Diligence, lifetime employment, loyalty to the company — "Japanese-style management" and *Nihonjinron* factors are also popularly wheeled out. But these have come in for criticism in recent years, and the prevailing view is that they are a story told after the fact, reasoning backward from the success to name virtues.
Do the standard account of the recovery and what the prefaces say connect?
The standard answers to "how did Japan recover?" bundled into four, and checked against the 883 notes for matching statements. Figures are total occurrences; entry counts in parentheses.
| Standard factor | Connection | Strength | Evidence in the corpus |
|---|---|---|---|
| External and structural Korean War procurement, Cold War, 360 yen, trade expansion |
✕none | Korean War 0 / special procurement 0 / Cold War 0 / exports 2 / trade 8(3). A future invisible to those living then | |
| Institutional and policy Occupation reforms, land reform, zaibatsu dissolution, industrial policy |
△partial | Democracy 136 / freedom 127 are dense, but land 7(1) / zaibatsu 8(8) / labor 28(9): references to economic institutional reform itself are thin | |
| Legacies from before the war accumulated technology, education levels, human capital |
◯strong | Science 373(111) / research 232(105) / education 160(68) / technology 88(43). Tags "Building a Nation on Science 124," "Hopes for the Children 160" | |
| Cultural explanations diligence, loyalty, Japanese-style management |
⚠caution | "Diligence" appears only 3(2) times in total. But effort 104(86) and expectation/hope 122 show that the forward-looking bent itself is real |
The "strength" bars use the degree of connection (✕△◯⚠) as a rough length; they are not measurements. The grounds for the judgment are the figures in the right-hand column.
What can be read from it
1. The standard account lies beyond the prefaces and afterwords. The effects of special procurement, the Dodge Line and land reform all come from 1947 onward. What the prefaces and afterwords hold is the state of mind before the recovery began. They are no more than "primary sources reflecting what people heading toward recovery thought," and there were no prophets. Still, there were buds: people thinking about land reform, people set on trade being the future, people scheming to make industry more efficient.
2. The recovery came out of "expectation," not "self-reflection." I hypothesized that "for a recovery there must have been self-reflection." But against 37 notes of self-reflection and self-blame, the forward-looking words were overwhelming (Building a Nation on Science 124, Hopes for the Children 160, The Mood of Reconstruction 161). What the material shows is people skipping the reflection and facing forward.
3. A limited counter, from primary sources, to the "cultural explanation = after-the-fact myth" criticism. "Diligence" appears only 3 times in the data. What comes through strongly instead is a forward-looking bent — "rebuild through science / entrust it to the next generation" — heading, if anything, toward overcoming what was "Japanese" before and during the war.
The Objective Analysis, Summed Up
The standard account can be called a macro view worked back from the outcome. The preface-and-afterword data is a micro view, and can also be read as the motive for publishing a book in that period. What connected most strongly with the standard account was building a nation on science, education (human capital), and forward-looking hope.
To be honest, I think Japan managed its recovery, such as it was, because it was lucky. Lucky people are basically cheerful. That the data comes out cheerful seems only natural to me, but that sort of thing cannot be turned into an objective number. And this data reflects my own subjectivity strongly in the first place. In the end, this data cannot be handled without subjectivity.
Source: all word counts are from the 883 reading notes (recounted 2026-08-05). The structure of the table and the judgments of connection come from the body text.