PAPERS, COMMENTARY, BOOKS, ESSAYS

Research & Writing

On the project pages, I am inside the problem, and what I can claim is limited to what I saw over a project cycle or a short visit. Here, I test the same questions against evidence I did not gather myself and against people who know more than I do.

Publication is not influence. I have no way of knowing whether anyone acted on any of this work, and I would not claim they did. What each piece can be held to is its argument and its evidence. The paper leads because it is the only item here that underwent peer review, and the clearest piece a stranger can assess without taking my word for it.

Three questions this work keeps returning to

The collections are organised by form, but the thinking is not. Each question below is approached from several directions: a published paper, commentary, an essay, a book and a room I sat in. That is why this page is built as an argument rather than a reading list.

1 How does a claim cross a border, and who is left responsible for it?

A false claim moves faster than any jurisdiction that could act on it. My paper traces that problem through the Russian-Ukrainian war and ends up somewhere I did not expect: not with better enforcement, but with the question of how a system can be built so that accuracy is the incentive rather than the exception.

2 Who does a rule actually reach once it is written down?

A scheme can be published, funded and available, and still not arrive. Somewhere between the drafting and the counter, eligibility turns into a reading test that nobody designed and nobody administers. I meet this question weekly at a constituency office, and then go and read about it.

Commentary

The Language of Exclusion

March 2026. Written directly out of the casework, about half of which is conducted in Mandarin.

Book

Words Unspoken

Who gets to speak, and who gets heard, is not only a matter of confidence or cultural habit; systems shape it too.

3 What makes any of it last once the people who started it leave?

My own paper answers a misinformation question by arriving at an institutional one: reactive regulation always lags, so the only durable answer is a system where accuracy is structurally incentivised rather than occasionally enforced. I encounter the same institutional problem at Project Flare, where a curriculum has to be documented well enough to be inherited.

Essay

Man vs Machine

Harvard 2024. Ends on who funds a reweighted curriculum, taking SkillsFuture as the working example.

The complete collections

1

Peer-reviewed research

The published paper, with its method, findings, limits and how to verify it.

Read the research

21

Policy commentary

Monthly since December 2024, in four categories, filterable by theme.

Read the commentary

5

Selected essays

Philosophy, legislation and economics. Longer arguments, 2023 to 2025.

Read the essays

2

Books

An anthology on public speech and a children’s picture book, and what happened after them.

Read about the books

14

Forums and debate

Dialogues, forums, committees and rounds. Where the argument had to be made out loud.

See the rooms

Where this connects

The four initiatives behind the writing, and the question each of them runs into.

See the initiatives

The one reply I have

I first met the Prime Minister at the SGLaw 200 Youth Forum. That exchange stayed with me, and later I sent his office some of my recent writing, including Words Unspoken and A Weapon of Mass Distraction. A reply came several weeks later.

Jerome Ong speaking with Prime Minister Lawrence Wong among students at the SGLaw 200 Youth Forum.

In conversation with Prime Minister Lawrence Wong at the SGLaw 200 Youth Forum, May 2026.

“I am glad to see you writing so actively, and taking such a keen interest on important issues shaping our society and our future.”

Mr Lawrence Wong, Prime Minister of Singapore

Where the writing stops

My own paper is the clearest case. It works through a war, two regulatory responses and a hybrid model, and then lands on a claim about structural incentives that I cannot yet make concrete. I can say what a durable information environment would have to do. I cannot say who would build it, who would pay for it or what would hold a government to it once the war that prompted it had ended.

The commentary can show that a scheme is written in the wrong register and cannot say who would have to be made responsible for rewriting it. The books reach readers who came looking, which is the easiest kind of reader to reach. The read-aloud reached sixty children in one morning because a school opened a door, and it opened because I had been a pupil there.

What I want to learn is the part after the argument: how a rule is drafted so that it still works on the platform it was written for; how a service is funded so that it outlasts the enthusiasm that created it; and how anyone finds out afterwards whether either of those worked.