Corey Rayburn Yung

The complete model includes the Framers, pretext, subtext, text, Judges, precedent, historical setting, and modern setting. It also includes a coding and decoding process for pretext, text, and subtext.

New originalism only includes Framers, text, judges, and the historical setting. There is still an encoding and decoding process for the text.

From my 2016 article, this was my complete model of how judges should interpret the Constitution and the simplified model that original-meaning-originalists implicitly use. Those familiar with the Shannon-Weaver model can see its influence.

Oh no!

Frontiers | Quantifying Legal Entropy Many scholars have employed the term “entropy” in the context of law and legal systems to roughly refer to the amount of “uncertainty” present in a given law...

More recently, the first article in my Systems of Law series talks about information theory and Shannon's concept of entropy. Related to that, Ted Sichelman has a fascinating article (not for the faint of heart) about how to calculate Shannon entropy in law:

Originalists, in particular, typically have a naive conception of communication embedded in their theories (although this problem isn't unique to originalists) that, IMHO, undermines their claim of legitimacy based on original public meaning. A little Claude Shannon engagement would help.

FWIW, this was my article where I adapted and updated the Shannon-Weaver communication model to constitutional interpretation, including visualizations of the dominant theories of interpretation at the time (2016). I think constitutional scholars should do far more of this.

I suppose it is both. But that was my assumption as well.

The Information: A History, A Theory, A Flood A History, A Theory, A Flood

Indeed.

I highly recommend @petergleick.bsky.social's The Information on the fascinating topic of information theory and Shannon's incredibly important contributions:

bookshop.org/p/books/the-...

Claude Shannon - Wikipedia

Shannon's writing is dense and not readily accessible (he was a non-academic engineer, after all). But anyone writing about legal interpretation or applying quant methods to communication should be familiar with his scholarship, IMHO. Anyway, here is his wiki page: en.wikipedia.org/wiki/Claude_...

Shannon's "The Mathematical Theory of Communication" is an incredibly important academic work that was generated outside of academia. After writing the article, he joined forces with Warren Weaver to refine his ideas and created the Shannon-Weaver model of communication.

Despite being one of the few law professors to write about Claude Shannon (first in 2016), I didn't know, until I read this, that Claude.ai was named after him. Shannon was a truly amazing figure. As a Bell Labs engineer, he created information theory and the first modern communication theory model.

I need more time to walk through the math, but I LOLed at this: "Here is a simplification of that Python program, converted into C form." Only a computer science prof would choose to convert perfectly readable Python code to C in the text of an article.

What do you call your act?
The Textualists!

I spoke to @janus.bsky.social about age verification, IDs, and what happened in Kansas.

today is the day

Calling Cormac McCarthy’s authorial style a “mistake” that AI no longer makes is so funny

It's the year 2026. How does Bluesky still not have polls?

Once upon a time, right-wingers blamed video games and violent media for kids becoming mass shooters. Now, they produce and post such content on the official White House account.

wishing a full and speedy recovery to an absolute legend

Honestly, I think ChatGPT's abstract (the 3rd one) is better than mine (the 1st). Gemini's is pretty bad. Claude has some good points, but it uses some awkward phrases and concepts. I'd be curious to hear the impressions of others.

FWIW, I used the free versions of ChatGPT and Claude. But, I'm currently on a free trial for Gemini Pro. Not sure if that matters, though.

My experiment doesn't address whether scholars should use LLMs for abstract writing. I just wanted to test the efficacy of doing so. I haven't looked at the results carefully yet, but the abstracts do seem to confirm that LLMs really love em-dashes.

The abstract I created: "Analyzing a system requires different theories and methods than studying its components. That fundamental insight has spurred methodological and ontological revolutions across the social sciences. In contrast, despite the frequent references to legal “systems,” scholars of law remain wedded to a reductionist and atomistic paradigm. Researchers seek to isolate the objects of study like scientists examining single cells through a microscope, ignoring interactions, contextual effects, and behavioral adaptation. Or, in the case of quantitative empirical research, scholars examine aggregate data, assuming observed correlations are linear, causation is mechanistic, and the internal aspects of law that distinguish it from other social systems are irrelevant. While these approaches are adequate for examining discrete doctrines and simple legal problems, they obviate an understanding of the systemic and dynamic dimensions of legal regimes. The limitations and shortcomings of the traditional modes of analysis are particularly exposed during this period of legal crisis. To address these shortcomings, this Article proposes a legal systems framework that adapts elements of systems and complexity theories to better explain and understand the dynamics of modern law."

From Gemini: "Despite the dynamic and interconnected nature of modern legal regimes, contemporary legal scholarship remains largely confined to reductionist and atomistic paradigms that assume linear, mechanistic relationships between legal inputs and social outcomes. To address these analytical shortcomings, this Article introduces "jurisystems," a novel theoretical and methodological framework tailored to the study of American law. Jurisystems synthesizes foundational elements of autopoietic systems theory—particularly Niklas Luhmann’s insights on law as a cognitively open communication network—with core concepts of modern complexity theory, such as emergence, nonlinearity, adaptation, and path dependence. By advocating for the integration of advanced quantitative methodologies, including agent-based modeling, network analysis, and information theory, this framework bridges the historical divide between internal doctrinal analysis and external empirical observation. Ultimately, adopting this complex systems perspective equips scholars and practitioners with more sophisticated, empirically grounded tools to understand, model, and reform legal institutions in an era of unprecedented social complexity."

From ChatGPT: "Legal scholars frequently invoke the concept of a “legal system,” yet the dominant methods of legal analysis remain largely reductionist, focusing on discrete doctrinal elements or linear empirical relationships rather than the dynamics of law as a complex system. This Article argues that modern law cannot be adequately understood through these atomistic approaches. Drawing on intellectual traditions from systems theory and complexity science, it develops a theoretical and methodological framework for analyzing legal systems as adaptive, nonlinear networks of interacting institutions, actors, and communications. After tracing the historical development of systems thought—from Enlightenment reductionism through structuralism and contemporary complexity theory—the Article examines how American legal scholarship has addressed (and largely neglected) systemic analysis through both internal doctrinal methods and external empirical approaches. It then introduces a new framework, termed jurisystems, which integrates insights from systems theory, particularly Niklas Luhmann’s account of law as a communicative system, with modern complexity concepts such as emergence, adaptation, and path dependence. Finally, the Article outlines methodological tools—including agent-based modeling, network analysis, and information theory—that enable scholars to study legal systems empirically while preserving the distinctive internal features of law. By conceptualizing law as a complex adaptive system rather than a collection of discrete rules or outcomes, the jurisystems framework provides a foundation for more realistic analysis of legal change, institutional behavior, and the systemic consequences of legal reform."

From Claude: "Legal scholarship has long relied on reductionist and atomistic methods — analyzing individual cases, statutes, and doctrines in isolation — while largely ignoring the systemic dynamics that shape legal outcomes at scale. This Article argues that these dominant methodologies, whether doctrinal formalism or quantitative empirical analysis, are fundamentally ill-equipped to explain the nonlinear, adaptive, and emergent behavior of modern legal institutions. Drawing on intellectual traditions from classical and autopoietic systems theory, complexity science, and American legal thought, the Article proposes "jurisystems" — a discipline-specific framework for studying American law as a complex adaptive system. The framework synthesizes Niklas Luhmann's conception of law as an operationally closed but cognitively open communication network with complexity theory's insights about emergence, path dependence, feedback loops, and nonlinear dynamics. To operationalize this framework, the Article outlines three methodological pillars — agent-based modeling, network theory, and information theory — that together offer tools capable of capturing law's multidimensional complexity in ways that regression analysis and doctrinal synthesis cannot. Rather than displacing existing methods, jurisystems integrates them into a broader analytical architecture, enabling legal scholars to ask — and begin to answer — foundational questions about how law evolves, adapts, and sometimes fails in the face of social change."

In relation to the discussion about using LLMs for abstract writing between @hoffprof.bsky.social, @jtlg.bsky.social, @evanbernick.bsky.social, and others, I prompted Gemini, ChatGPT, and Claude to write new 1-paragraph abstracts for my Systems of Law: Foundations draft.

Which is fair. But, I'm curious. I struggle so much writing effective abstracts because they run contrary to my overall approach to writing. Despite @jtlg.bsky.social's belief that more practice will yield better abstracts, I fear I've reached my ceiling on that front.

I'm still not inclined to use LLMs for any writing. But this use case is the first time I'm tempted. As I posted in another skeet, I think abstracts are advertising and not scholarship (they don't get quoted, after all). If an LLM creates a "better" abstract, maybe some scholars should use them. 🤷

But is abstract writing an important skill? It's not the actual text. It doesn't get quoted by other authors. It seems more akin to advertisement than scholarship. And maybe we shouldn't be afraid to outsource advertising for our scholarship. But I'm just thinking outloud here.

I've never used AI to write an abstract, but now I'm curious about it. My writing style focuses on specifics and precision. Abstracts are literally abstract. I struggle writing at a high level of generality. I'm not saying I will use an LLM for my abstracts, but it's an interesting use case, IMHO.

I spend a ton of time writing my first few introductory paragraphs and I think I'm pretty good at it. But, I hate abstracts with the passion of a thousand suns and struggle to write good ones. I can't explain the disjunction.

As an IR professor, it's depressing that this is outperforming all others as the best theory of US foreign policy

Thanks!

Key Details of 13-Year-Old Trump Accuser’s Accounts Are Verified The newly corroborated details suggest the woman was truthful about numerous aspects of her life.

There is apparently credible evidence that the president of the United States is a pedophile. The comforting phrase “accused him without evidence” no longer fits this situation, and journalists must resort to the more perfunctory “hasn’t been formally charged.” www.thedailybeast.com/key-details-...