The Rhetoric of Economics

The author’s trilogy on the rhetoric of economics: If You’re So Smart, The Rhetoric of Economics, and The Vices of Economists all discuss literary criticism of economics. The first book is not recommended at all—its fragmented literary criticism is tedious, and its flowery language cannot hide the lack of structure. The second book, however, is much better. This article summarizes the second book.

Rhetoric, in essence, is persuasion with a purpose. Ancient Greeks used arguments, structure, and style to build the framework of rhetoric, while contemporary economic rhetoric has richer content—facts, stories, logic, and metaphors. Unlike abstract natural language, which is created solely for description, social science writings are actually written with intent, giving economics a humanistic flavor through rhetoric.

In this blog, “the author” refers to Deirdre N. McCloskey, the author of The Rhetoric of Economics.

Rhetorical Analysis Structure Table:

Four rhetorical perspectives: facts, stories, logic, metaphors.

What is Rhetoric, and How to Do It?

The Anonymous Spokesperson of Authority.

Economics is not the voice of God but the language humans use to persuade others to believe a certain truth. Although it is human language, it pretends to be the spokesperson of historical laws, metaphorically anonymizing itself as science itself.

Do not applaud me, audience, for today it is the voice of history that speaks.

Thus, economics always uses the eternal present continuous tense to describe factory production, profit formulas, even though these organizational forms should be analyzed in historical contexts.

Creator Techniques.

First, poets and metaphors.

Extending game theory to nations, markets become competitive arenas, and economists tell their countries, “We are going to lose the game because we have xx problem.” Metaphors add persuasive power to the necessity of models. Using network metaphors, small companies as families, durable goods as children…

Second, novelists.

Economics must learn to combine stories.

xx was once poor, but then embraced capitalism, and xx became rich. Story combination is about linking related events through state descriptions, actions, and internal connections (causation, temporal relationships) to form a narrative (positive and negative).

The end of each story signifies a new state. In economics, perfect and imperfect endings are equilibrium and disequilibrium. A perfect story needs a curtain call, a prophecy of the ending state rather than just a summary of the past, which can be extended.

Texts lacking actions and states do not qualify as stories.

Economists emphasize starting with exogenous variables because, in the story system, this serves as the beginning. Starting with endogenous variables makes the story feel like it begins in the middle.

This is also for speculative reasons. Blaming historical bad debts on obvious roles makes for a very uninspiring ending.

Performance.

Fiction does not imitate reality; fiction creates reality. Influential economics papers often contain convoluted and obscure sentences, such as Keynes’s The General Theory of Employment, Interest and Money. When writing papers, try to create “the author’s audience” (knowing the story is fictional) and “the narrative audience” (unaware the story is fictional)¹.

Because economists themselves need to realize that they are not necessarily stating facts but creating an audience.

Every text has an implied author, an implied reader, a history, and a form.

The Literary Characteristics of Economics: Science, Literature, Art

The author wrote this book to explore the boundaries between economics, science, and text. In the author’s view, science is merely a way of speaking, not a domain of truth.

• Science requires texts to express itself.
• Scientific experiments are limited, and often people judge the value of a theory based on its aesthetic appeal.
• The methods of persuasion in texts are similar to literature and speeches.

Thus, science, literature, and art intersect rather than oppose each other in the traditional sense.

Personally, I reserve my opinion on this.

Economic arguments employ the following argumentative methods:

The first three points are from British scientific argumentation. From a literary perspective, the author expands these to more points.

Statistical tests, specific market analysis, experiments on theories, introspection (What would I do if xx happened?), thought experiments (Based on experience, what would others do in xx situations?), case studies, folk wisdom (merchants), traditional academic knowledge, symmetry of laws (e.g., defining glorious periods and garbage periods), pure definitions, analogies…

For example, economic papers do not describe the motion of objects but depict human behavior, which aligns with literary creation. Even though economic papers describe a category of people, readers imagine specific individuals rather than abstract object movements, a role depiction shared with literature.

Naming is part of rhetoric and image creation. For example, the term “preference” carries a strong literary sense of desire, and most economic terms have this quality, such as Hayek’s spontaneous order, Adam Smith’s animal spirits…

Rhetorical Techniques in Economics

Take Samuelson’s use of mathematics as an example—when using statistics and mathematical models, he uses we, but when discussing economic propositions, he uses I.

Because Samuelson states in the preface that economic propositions are debatable and refutable, using I shows a more conversational stance, while we conveys authority (implying consensus).

If we conduct an experiment, we find that most people do not care about mathematical proofs or statistical details, but the mere presence of mathematics enhances persuasiveness.

John Stuart Mill’s paper on expectations emphasizes the role of economics professors—criticizing people’s slow market expectations from the perspective of economic researchers.

The author’s critique of economic forecasting is interesting—economic forecasts are always conditional, and conditional forecasts are inherently consistent with experience, so how can they truly exceed expectations?

I have heard similar views from others—econometrics analyzes causality through exogeneity, so how can we expect it to predict exogenous events?

Fogel used the method of sacrificing the small for the big, as he was refuting the mainstream conclusion that “railroad construction promoted U.S. economic growth.” Regarding x, y, coefficients—perhaps y is small; x is large; we overestimated the coefficient, leading to today’s conclusion. His paper uses hints of uncertainty to cast doubt on theories.

Fogel

Fogel was the first to compare railroads to river transport in terms of contribution, using historical quantification and counterfactual thinking, inspiring counterfactual causal inference. He argued that past views overestimated the contribution of railroads.

Since his report, economics has also used quantitative history to revolutionize the paradigm of economic history.

Dave was inspired by this, measuring market accessibility based on railroad expansion, establishing a general equilibrium of railroads and market accessibility, allowing comparative advantage to be empirically measured, and winning the 2017 Clark Medal.

Coase’s² papers are written in a lawyer’s tone, proposing various theoretical possibilities and then refuting them one by one. Coase’s papers resemble the structure of ancient Greek oratory—introduction, narrative (facts), detailed classification (controversial and non-controversial claims), evidence, refutation, conclusion. Thus, his classic paper The Nature of the Firm is filled with adversarial language.

Other rhetorical techniques include—appeals to authority, verbal hints (metaphors for competition), metaphors³ (zero-sum games, children as durable goods), parables (metaphors + stories)…

Literary understanding is so broad…

Does Quantification Need Rhetoric?

Two questions—How big is big? So what?

On “How Big is Big”

Some believe that by mathematizing problems, pure mathematical thinking can separate what is within human reach. But is this really the case?

How big is a number? What counts as big or small is still a process of trying to persuade others, which still requires human comparison to reach a conclusion. Thus, the process still involves rhetoric.

Of course, if you study advanced econometrics⁴, you will find that hypothesis testing constructs small probability events—“Type I errors” and “Type II errors” to test significance. Although this is considered standard practice, it is still a relative standard.

I agree with Shi Zhentao’s view on econometrics teaching—“How big is credible” is not just a statistical question but also a philosophical one. Advanced econometrics merely makes people start thinking about this question.

For example, Bayesians and frequentists have long debated, and Bayesians themselves are divided into subjective and objective schools. While it all looks like mathematical derivation, the basic construction of credibility measurement always involves worldview. Everything can indeed be philosophical.

For example, the classic case—statistical significance and economic significance should be distinguished.

On “So What”

Economic papers need to address specific audiences, and you are also narrating a fictional society. Different groups see the world differently, and to gain their recognition, you need to resonate with their “consensus”—that is what captures their attention.

Things to Note in Econometrics

Distinguish between statistical significance and economic significance.

In short, paper arguments cannot rely solely on statistical significance.

You need to clarify sample distribution assumptions, the real-world meaning and form of coefficients (e.g., whether they are elasticities), incorporate real-world consensus analysis, strictly define significance, ensure a natural transition from statistical significance to real-world impact, and recognize that statistical, policy, scientific, and economic significance are different. (This section is taken from AER’s records of misapplications of statistical significance in the 1980s.)

The author’s attacks on modernism, represented by mathematics and statistics, are numerous, but the principle is simple:

Many view falsification as science, but falsification is an arbitrary issue. Too much real-world evidence does not guarantee full coverage. When a theory is tested, in economics, it often manifests as widespread interest in it. Thus, the author ultimately calls for skepticism toward any rule-based methodology⁵. We should oppose anti-rhetoric and anti-modernism.

Will Rhetoric Be a Future Discipline?

Two trends are evident in economics today.

1. The fundamentals of mathematics and statistics are gaining popularity. In advanced econometrics courses, I see history students enrolling, while economics seniors are moving toward mathematics. The deep intersection of mathematics and economics is becoming the norm. Mathematical economists are joining, and standards are being raised.

2. On the other hand, as mathematics and statistics are widely applied, their rationality is increasingly questioned. The instrumental nature of mathematics is indeed giving way to rhetoric. The themes, stories, and inspiration of papers are becoming more interesting to people.

This trade-off, like “correctly reaching a conclusion” versus “reaching the correct conclusion,” leaves people unsure of how to choose and which direction will dominate.

If we examine science solely based on “persuasion,” then all methodological rules become irrelevant. With such a liberal definition, worldviews might turn into function-oriented utilitarian thinking.

But rhetoric, standing against modernism, undoubtedly offers a fresh perspective to examine science. Innovation requires boldness, and this freedom provides a springboard. Therefore, I believe The Rhetoric of Economics is absolutely worth reading.