Did a feedback mechanism between propositional and prescriptive knowledge create modern growth?
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Economics. The day in economics: theory, mechanisms and the evidence beneath them.
Ruth: We're looking at the paper: "Did a feedback mechanism between propositional and prescriptive knowledge create modern growth?".
Warren: Here's the gist of the paper. Joel Mokyr's thesis posits that self-sustained modern economic growth arose from a positive feedback loop between propositional (theoretical) and prescriptive (applied) knowledge, which intensified during the eighteenth-century "Industrial Enlightenment." This paper provides the first quantitative test of this hypothesis using English publications from 1600 to 1800.
Ruth: The research introduces novel text-based measures for innovation and knowledge spillovers, utilizing a semantically deep and historically fine-tuned BERT model to derive these measures in the absence of traditional citation data or in their presence.
Warren: The core empirical findings demonstrate that the feedback loop between propositional and prescriptive knowledge became positive around 1740–1759, directly supporting Mokyr's prediction that growth in one type of knowledge increased the marginal product of the other.
Ruth: Furthermore, this relationship was linked to real economic outcomes; spillovers from propositional knowledge were positively associated with patent citations in 1700–1800, indicating that changes in the knowledge economy correlated with changes in the real economy.
Paper summary: Ruth: So we're looking at this paper titled "Did a feedback mechanism between propositional and prescriptive knowledge create modern growth?" by Julius Koschnick. This study is trying to put a quantitative test on Joel Mokyr's idea that the Industrial Enlightenment involved a positive feedback loop between theoretical and applied knowledge.
Warren: Exactly. The core of the paper claims that this loop became positive around one thousand seven hundred forty to one thousand seven hundred fifty-nine which supports Mokyr's prediction that growth in one type of knowledge boosts the other <ref:2512.16587#pg1>. It also links this to real economic outcomes, showing that spillovers from theoretical knowledge were associated with patent citations between one thousand seven hundred and one thousand eight hundred <ref:2512.16587#pg1>.
Iris: What I find interesting is how they tackled the evidence. They used novel text-based measures for innovation and knowledge spillovers, utilizing a semantically deep and historically fine-tuned BERT model to do this without relying on traditional citation data or even when it was present.
Ruth: That’s a big methodological step, Iris. It means they’re trying to measure these things using the actual historical text itself. They found that over the next one hundred forty years, negative knowledge spillovers phased out until around one thousand seven hundred forty when the relationship became positive for the first time <ref:2512.16587#pg3,spillovers phased out until around 1740 when the relationship became positive for>.
Warren: And linking that back to economic data, they found a positive association between spillovers from theoretical knowledge and patent citations during that same period, specifically spanning from one thousand seven hundred to one thousand eight hundred <ref:2512.16587#pg1>.
Iris: The paper also looked at how this feedback loop actually got going by investigating several mechanisms. They found that upper-tail human capital, specifically among Royal Society fellows and authors with university education, helped facilitate these knowledge spillovers.
Ruth: That points toward who was driving the innovation in that era. They also examined method revolutions as complements to this process; they found precise measurement acted as a key catalyst for incorporating theoretical knowledge into applied techniques.
Warren: And Newtonian mechanics and the scientific method complemented that by integrating practical insights back into theoretical understanding. It’s interesting how they broke down those mechanisms beyond Mokyr's initial theory.
Iris: They provided direct causal evidence by using an exogenous shock, specifically the release of John Harris’s lexicon technicum in one thousand seven hundred four as a shock to propositional knowledge <ref:2512.16587#pg1>. Using a difference-in-differences framework on prescriptive topics not covered by that lexicon, they found that knowledge spillovers from this source led to increased innovation in prescriptive knowledge over the next two decades.
Ruth: That causal evidence using the lexicon is pretty strong for showing positive spillovers from propositional to prescriptive knowledge via codified encyclopedias. It really solidifies the idea of a direct path between theory and practice driving change.
Warren: What this means for us on air is that we can now quantitatively estimate structural change in the knowledge economy using historical text data, which offers a new way to look at growth drivers beyond just traditional citation counts.
Iris: It also addresses debates on science's influence by showing how innovation-enhancing spillovers from science happen through various indirect channels involving codified knowledge and methodological shifts. They also introduced novel NLP techniques to quantify the knowledge economy using historical text data, offering robust measures that circumvent biases inherent in citation-based indicators.
Ruth: Overall, this paper offers a new quantitative approach to estimate structural change in the knowledge economy and provides strong evidence for Mokyr's theory of a feedback loop driving the Industrial Revolution. It also investigates mechanisms beyond Mokyr’s initial theory, highlighting the role of method revolutions and upper-tail human capital in facilitating spillovers.
Warren: The paper reconciles debates on science's influence by showing how innovation-enhancing spillovers from science occur through various indirect channels involving codified knowledge and methodological shifts. But I do want to mention what they did not do, which is that they focused heavily on the English publications from one thousand six hundred to one thousand eight hundred so we don't have a full picture across all time periods <ref:2512.16587#pg1,English publications from 1600 to 1800>.
Iris: That’s a fair limitation. They also found evidence of neutral or negative knowledge spillovers at the beginning of the seventeenth century before that positive feedback loop kicked in around one thousand seven hundred forty which is important context for understanding the timeline <ref:2512.16587#pg1>.
Ruth: So, to wrap up this study on "Did a feedback mechanism between propositional and prescriptive knowledge create modern growth?", it really shows how we can use historical text data to test major economic theories about innovation. It proves that the connection between what we theorize and what we apply wasn't always positive, but it turned positive during the Enlightenment.
Warren: And the implication is that when you look at economic history, you can’t just look at isolated inventions; you have to see how ideas from one sphere feed into another to create sustained growth. The feedback mechanism is the engine here.
Iris: The paper brings together different fields—economics, computer science with the BERT model, and historical analysis—to give us a new lens on the knowledge economy. It shows that measuring knowledge spillovers doesn't always have to rely on traditional citation metrics alone if you use these advanced text analysis tools.
Ruth: It changes how we think about what drives economic growth at this level of detail, moving beyond just counting patents to understanding the actual flow of ideas between theory and practice. That’s what this paper does for the world.
Conclusion: Ruth: So we’re wrapping up this look at that paper by Koschnick about whether a feedback loop between theory and practice drove modern growth.
Warren: Yeah, so the title itself is really direct, isn't it? Did a feedback mechanism between propositional and prescriptive knowledge create modern growth. It sets up this big question right off the bat.
Iris: I think what’s important is that they did this by using English publications from one thousand six hundred to one thousand eight hundred as their data set. That’s a huge chunk of history, but it means they are testing a very specific idea on a very specific time frame.
Ruth: Exactly. They aren't trying to map the whole Industrial Revolution; they're focusing on this feedback loop becoming positive around one thousand seven hundred forty to one thousand seven hundred fifty-nine. That’s where the timing really matters for the theory, right?
Warren: And that timing is key because it connects directly with their economic findings. They found that spillovers from theoretical knowledge were positively associated with patent citations between one thousand seven hundred and one thousand eight hundred. It shows a correlation in real economic activity.
Iris: But the methodology they used to find that correlation is what’s interesting for us, because they used this semantically deep BERT model to measure innovation and knowledge spillovers without relying on traditional citation counts at all. That’s a novel way to get those numbers.
Ruth: It means they're trying to see the flow of ideas in the historical text itself rather than just counting how many times someone cited another person, which is a much different kind of measurement for us.
Warren: And they didn't just stop at correlation, though. They looked at causal evidence too, using John Harris’s lexicon technicum as an external shock to see if that really caused more innovation in applied knowledge later on. That’s a strong part of the argument.
Iris: What it means for us is they found ways to study structural change in the knowledge economy using historical text data, and they showed how science's influence flows through indirect channels like codified knowledge and shifts in methods.
Ruth: So the big picture here is that this paper offers a new way to see how economic growth wasn't just about isolated inventions; it was about these ideas feeding into each other in a specific, positive loop.
Warren: It really forces us to think about how we measure innovation when we look back at history and whether those measures capture the real engine of change.
Iris: It also highlights that upper-tail human capital, like Royal Society fellows or university authors, played a role in facilitating these spillovers, which adds a layer to who was actually doing the heavy lifting in that era.
Ruth: So they’re not just proving Mokyr right; they’re showing us *how* the mechanism works and what factors—like those specific types of people and those methodological shifts—help it turn positive.
Warren: That leads us nicely into how these kinds of feedback loops apply to modern economic growth, which is where we can take this idea next.
University of Southern Denmark
econ.GN, q-fin.EC
Submitted: 2025-12-18
Updated: 2026-11-05
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 93/100
The gist: Joel Mokyr's thesis posits that self-sustained modern economic growth arose from a positive feedback loop between propositional (theoretical) and prescriptive (applied) knowledge…
Key concepts
- Propositional Knowledge
- This refers to theoretical or intellectual knowledge—the 'what' of science and ideas. In this context, it includes scientific theories, mathematical principles, and abstract concepts developed by thinkers of the era. The study measures how changes in this type of knowledge affect applied practices.
- Prescriptive Knowledge
- This is the practical or applied knowledge—the 'how' of doing things. It involves techniques, methods, and technologies used in industry and science. The research tracks how insights from theoretical knowledge influence the development and improvement of these practical techniques.
- Feedback Loop
- The core hypothesis suggests a cycle where advancements in one area (e.g., new theory) boost the other (e.g., better methods), which in turn enhances the original area. The study empirically confirms this loop became positive around 1740–1759, showing mutual reinforcement between ideas and practices.
- Method Revolutions
- These are fundamental shifts in how knowledge is acquired or applied, such as adopting Newtonian mechanics or precise measurement. The paper shows that these revolutions acted as catalysts, helping to bridge the gap by enabling theoretical ideas to be incorporated into practical techniques.
Terminology
Summary
Joel Mokyr's thesis posits that self-sustained modern economic growth arose from a positive feedback loop between propositional (theoretical) and prescriptive (applied) knowledge, which intensified during the eighteenth-century The measures are based on comparisons between the past and future of different fields while operationalizing
a semantically deep and historically fine-tuned BERT model. With these two measures, it becomes
possible to derive measures of innovation and knowledge spillovers i) in the absence of citation data
and ii) in the presence of historical and context-sensitive text.Industrial Enlightenment.
This paper provides the first quantitative test of this hypothesis using English publications from 1600 to 1800. The research introduces novel text-based measures for innovation and knowledge spillovers, utilizing a semantically deep and historically fine-tuned BERT model to derive these measures in the absence of traditional citation data or in their presence. p. 1Where the paper says it
The core empirical findings demonstrate that the feedback loop between propositional and prescriptive knowledge became positive around 1740–1759, directly supporting Mokyr's prediction that growth in one type of knowledge increased the marginal product of the other. Furthermore, this relationship was linked to real economic outcomes; spillovers from propositional knowledge were positively associated with patent citations in 1700–1800, indicating that changes in the knowledge economy correlated with changes in the real economy. p. 2 The paper further links these results to the real
economy by estimating the association between spillovers from propositional knowledge and patent
citations in 1700–1800. As before, we find evidence of a positive relationship for the second half of
the eighteenth century.Where the paper says it
The study investigates several mechanisms facilitating this feedback loop. p. 4 natural resources (Wrigley, 2010) or factor prices (Allen, 2009), although these
accounts can clearly be seen as complementary. 3
Second, the paper moves beyond Mokyr (2002)’s theory and investigates the mechanisms behind
the feedback loop mechanism. To further understand the mechanism underlying the positive knowledge spillovers emerging
at the end of the eighteenth century, the paper tests whether upper-tail human capital facilitated
knowledge spillovers, as predicted by Mokyr (2002, 2016). Extracting information on authors’
careers from the ESTC, we find that fellows of the Royal Society, the Society of Arts (Howes,
2020), and authors with a university education (Koschnick, 2025) were more likely to incorporate
spillovers from other fields. The paper finds that for the adoption of propositional knowledge in new
prescriptive knowledge, precise measurement acted as a key catalyst. Vice versa, for the adoption
of prescriptive knowledge in new propositional knowledge, Newtonian mechanics and the scientific
method were key catalysts.Where the paper says it
Where the paper says it
Where the paper says it
The paper provides direct causal evidence by exploiting an exogenous shock: the release of John Harris’s Lexicon technicum in 1704, which served as a shock to propositional knowledge. Using a difference-in-differences framework on prescriptive topics not covered by the Lexicon, the study finds that knowledge spillovers from this propositional source led to increased innovation in prescriptive knowledge over the subsequent two decades, offering causal support for positive spillovers from propositional to prescriptive knowledge via codified encyclopedias. p. 3 Exploiting variation from the strength of spillovers between sub-topics, the paper adopts a difference-indifferences framework for prescriptive topics that were not covered in the Lexicon technicum. We
find that knowledge spillovers from propositional knowledge from the Lexicon technicum led to
increased innovation over the next 20 years.Where the paper says it
Overall, the paper contributes significantly across four dimensions: first, it offers a new quantitative approach to estimate structural change in the knowledge economy and provides strong evidence for Mokyr's theory of a feedback loop driving the Industrial Revolution. Second, it investigates mechanisms beyond Mokyr’s initial theory, highlighting the role of method revolutions and upper-tail human capital in facilitating spillovers. p. 4 3
Second, the paper moves beyond Mokyr (2002)’s theory and investigates the mechanisms behind
the feedback loop mechanism. The paper places new emphasis on the role of revolutions in methods that occurred during the seventeenth and eighteenth century, such as the scientific method, the
Newtonian Revolution, and precise measurement. We are therefore confident
that these measures will be useful to other researchers who are in need of text-based innovation and
spillover measures in the absence of citation data. Moreover, because these measures avoid some
of the biases inherent in citation data, such as peer and network effects, they also offer a useful
substitute for traditional citation-based indicators.Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI systems can quantitatively test Joel Mokyr’s feedback loop hypothesis by introducing new text-based measures for innovation and knowledge spillovers using a semantically deep and historically fine-tuned BERT model.
-
The improved system can derive measures of innovation and knowledge spillovers in the absence of citation data, as well as in the presence of historical and context-sensitive text, based on comparisons between the past and future of different fields.
-
The system can directly test Mokyr’s feedback loop thesis by estimating whether
over time, spillovers from one type of knowledge increased the innovativeness of the other.
-
The system can identify structural breaks in Britain’s knowledge economy during the seventeenth and eighteenth century by showing that
growth (spillovers) in one type of knowledge indeed seems to have increased the marginal product in the other type of knowledge.
-
The improved system can investigate mechanisms behind positive knowledge spillovers by testing whether
upper-tail human capital facilitated knowledge spillovers,
using proxies such as authors’ careers and memberships in societies like the Royal Society. -
The system can test if
revolutions in methods that occurred during the seventeenth and eighteenth century could have acted as complements to feedback loop processes
by calculating embedding space similarities to descriptive terms of methods like Newtonian mechanics or precise measurement. -
The system can provide direct causal evidence of positive spillovers from propositional knowledge to prescriptive knowledge by exploiting an exogenous shock, such as the release of John Harris’s Lexicon technicum, using a difference-in-differences framework.
-
The improved system can reconcile debates on science's influence by providing quantitative evidence of innovation-enhancing spillovers from science in a setting that allows for indirect effects, such as the adoption of the scientific method or precise measurement.
-
The system can identify key drivers of knowledge spillovers by showing that
all three methods, Newtonian mechanics, precise measurement, and the scientific method were important for integrating insights from lambda into the stock of knowledge in omega.
-
The improved system can quantify how upper-tail human capital and enlightenment networks facilitate knowledge spillovers by testing models where authors’ affiliations, occupations, and education are explanatory variables for received spillovers.
Abstract
Joel Mokyr has argued that self-sustained modern growth originated from a feedback loop between propositional (theoretical) and prescriptive (applied) knowledge that turned positive during the eighteenth century "Industrial Enlightenment". This paper presents the first quantitative test of whether such a feedback loop existed. For this, it introduces new text-based measures for 1) the innovativeness of publications and 2) knowledge spillovers. It then applies them to ca. 300,000 English publication titles and short patent descriptions, 1600--1800. The paper finds that a feedback loop between propositional and prescriptive knowledge became positive by the mid-eighteenth century, supporting Mokyr's original hypothesis.
Sources
- SimCSE: Simple Contrastive Learning of Sentence Embeddings
- BERTopic: Neural topic modeling with a class-based TF-IDF procedure
Related papers
- Huge If True: Learning from the Size of the Surprise
- The Rebundling of Freelance Work in the AI Age
- Dynamic investment and coordination in teams: Theory and experimental evidence
- Does Frontier AI Strengthen Cyber Defense? Evidence from Project Glasswing
- Generative AI at Work
- What is Work For? Labor, Politics, and Meaning in the Age of AI