How is the Free Speech Gap measured?

Two independent axes. Protection scores what the statute book permits — constitutional guarantees, criminal insult, defamation, blasphemy and hate-speech offences, penalties and defences — with every score cited to the legal text. Enforcement scores what actually happens: convictions and police-contact events per capita from official statistics where published. The Gap is Protection + Enforcement − 100: arithmetic, not a judgement.

Why do opaque countries not score as free?

Where a state publishes no enforcement statistics at all (confidence grade D), enforcement is imputed from what the statute book permits rather than assumed to be zero. Absence of data is never treated as evidence of restraint — otherwise states that hide their numbers would outrank states that publish them.

Data layer · Interactive · 66 countries

The Free Speech Gap

Q: Can you be prosecuted for insulting a politician in Germany?

Yes. StGB §185 (insult) and §188 (insult of persons in political life) are criminal offences; Germany recorded roughly 27,900 insult convictions in 2021 (Destatis), and a Bundestag document (Drs. 20/12694) itemises 800+ criminal complaints filed by sitting ministers since 2021.

Q: Can you be arrested for a tweet in the UK?

Yes. FOI-reconstructed figures show roughly 12,183 arrests for online communications offences in 2023 under Communications Act 2003 s.127 and the Malicious Communications Act 1988, with about 1,119 sentencings that year. No central government publication tracks these arrests — the count itself comes from freedom-of-information requests.

Q: Is this a political ranking?

No. The dataset codes what each law permits and what enforcement records show — never whether a restriction is justified. The same eight components are coded for every country, every scored cell links to a statute or official statistic, and the Gap is a subtraction of the two published axes.

What countries promise vs what they prosecute. Backed by data, not government claims — every score links to a statute or an official statistic.

See the Gap on the Map →

The picture

Every country, two axes: what the law says × what the state does

Where can you say it? — 66 countries · statute-coded · cited

Country

What do you want to say?

Check the full map for this →

Ranking

Freedom of speech by country, ranked

verified enforcement data only (grades A–B)

#	Country	Tier	Protection	Enforcement	Gap	Data

Protection: statute-coded, 0–100, higher = more protective. Enforcement: official per-capita statistics where published, 0–100; italic = imputed (no statistics published). Gap = Protection + Enforcement − 100. Not legal advice.

Head to head

Compare two countries

Country A

Country B

Worked example

Germany vs the United States: the gap in one table

Metric	Germany	United States
Criminal insult	Yes — StGB §185, up to 1–2 yrs; §188 aggravated for political figures	No federal offence; barred for protected speech
Insult convictions	≈27,900 (2021, Destatis §§185–187, 189)	0 for protected speech (First Amendment)
Police-recorded insult cases	237,784 (2023, BKA PKS)	—
Complaints by ministers	800+ since 2021 (Bundestag Drs. 20/12694)	—
Constitutional guarantee	Art. 5 GG, limited by "general laws"	First Amendment, Brandenburg standard

Neither row is an opinion. Germany's insult laws are real, actively enforced criminal statutes; the US First Amendment makes equivalent prosecutions unconstitutional. Both countries describe themselves as free-speech democracies — the enforcement record is where they diverge, and that distance is what the Gap number measures.

FAQ

Frequently asked questions

How is the Gap measured?

Two independent axes: Protection (what the statute book permits, 8 cited components) and Enforcement (convictions and police contact per capita from official statistics). Gap = Protection + Enforcement − 100. It's arithmetic on two published numbers, not a judgement.

Can you be prosecuted for insulting a politician in Germany?

Yes — StGB §185/§188 are criminal offences. ≈27,900 insult convictions in 2021; a Bundestag document itemises 800+ criminal complaints filed by sitting ministers since 2021.

Can you be arrested for a tweet in the UK?

Yes — ≈12,183 arrests for online communications offences in 2023 (FOI-reconstructed; no central statistic is published), with ≈1,119 sentencings under s.127/MCA that year.

Why don't opaque states rank as free?

If a state publishes no enforcement data, enforcement is imputed from what its statute book permits — absence of statistics is never scored as restraint. Those countries render hatched on the map and hollow on the chart.

Is this a political ranking?

No. The same 8 components are coded for every country from primary law; every cell cites its statute or statistic; the dataset never scores whether a restriction is justified. Dispute an entry via the methodology page — corrections are published in a changelog.

Speech is one layer. See all ten.

Pin your shortlist — sunshine, cost, tax, visas, safety, and the Free Speech Gap as deltas on one map.

Open the Map → Where can you mock a politician? →

About the data: the Free Speech Gap dataset is coded from primary law (official legal portals, official translations) and enforcement statistics (national justice ministries, police statistics, FOI reconstructions — each figure carries its source and year). Where a state publishes nothing, that absence is displayed, never papered over. The full dataset is published under CC BY 4.0. See the methodology for the component weights, formulas, and known gaps. Not legal advice.

Sources: national statutes · Destatis · BKA PKS · UK MoJ / House of Lords Library · Ministère de la Justice (Infostat) · EU DSA Transparency Database · Access Now #KeepItOn · Academic Freedom Index (CC BY 4.0) · documented case law. Full per-country citations in the open dataset.