For several nights this week, thousands of Georgians gathered outside parliament in Tbilisi carrying Georgian and European Union flags. Police responded with tear gas, water cannon and arrests as protesters attempted to prevent parliament from advancing legislation governing organisations that receive funding from abroad.
By Thursday, the government retreated.
The governing Georgian Dream party announced that it would withdraw the proposed legislation after the protests expanded and international criticism intensified. Parliament subsequently voted down the bill.
At first, the dispute appears relatively straightforward. Protesters describe the legislation as a Russian-style attempt to suppress independent civil society, while the government argues that organisations receiving substantial foreign funding should operate transparently.
The difficult part is that both principles contain something worth protecting.
Foreign governments should not possess unlimited ability to influence another country’s politics secretly. Democracies regulate campaign donations, lobbying and political financing precisely because money can become political power. If an organisation is effectively operating on behalf of a foreign government, citizens have legitimate reasons to know.
Moldova demonstrated the problem last month. A small democracy exposed to pressure from a much larger neighbour needs institutions capable of identifying covert political financing and foreign interference without treating ordinary political disagreement as treason.
Georgia faces an even more complicated version because its modern political history is inseparable from Russia.
Georgia was part of the Soviet Union until 1991. Relations with Moscow deteriorated after independence, particularly as Georgia moved towards closer relationships with NATO and the European Union. In 2008, Russia and Georgia fought a brief war, after which Moscow recognised the separatist territories of Abkhazia and South Ossetia as independent states. Russian troops remain stationed in both territories, which most of the international community continues to recognise as parts of Georgia.
For many Georgians, therefore, Russian political influence is not an abstract concern.
It is connected to territorial loss, military conflict and an unresolved question about the country’s geopolitical future.
That is what makes the accusation that the government is adopting a “Russian law” so politically powerful.
Russia’s own foreign-agent legislation began with requirements concerning organisations receiving foreign funding and later expanded substantially. The label acquired a strongly stigmatising meaning and has been applied to media organisations, civil-society groups, journalists and individuals.
The Georgian bill is not identical to every stage of Russian legislation.
Historical analogy should not replace legal analysis.
But institutional precedents matter because governments rarely receive enormous discretionary powers in their final form.
Powers expand through steps.
A registration requirement.
A reporting obligation.
A new legal category.
A broader definition.
Stronger penalties.
Each step can be defended individually.
The cumulative system may eventually look very different from the first law.
This is why citizens often oppose apparently limited institutional changes based partly on the trajectory they fear rather than only the immediate text.
Governments respond that opponents are exaggerating hypothetical future abuse.
Sometimes they are.
The problem is that constitutional safeguards exist precisely because citizens cannot know who will use a power later.
A law should therefore be evaluated not only by asking whether the current government promises to use it responsibly, but whether the power would remain acceptable in the hands of a future government we distrust.
The veil-of-ignorance test returns again.
Would Georgian Dream support the same foreign-agent framework if an opposition government controlled the institutions responsible for designating organisations?
Would the protesters support strong foreign-funding disclosure requirements if the principal foreign-funded organisations supported political causes they disliked?
A durable rule should survive both questions.
This suggests an important distinction between transparency and stigma.
Transparency says: tell the public where the money comes from.
Stigma says: because the money comes from abroad, the recipient should be institutionally classified as serving a foreign interest.
Those are not equivalent.
A Georgian environmental organisation may receive funding from a German foundation because environmental projects require resources unavailable domestically.
A human-rights organisation may receive European grants.
A media organisation may rely on international donors because Georgia’s advertising market is small.
Foreign funding can create incentives and dependencies worth disclosing.
It does not automatically establish foreign control.
This is a basic problem of causal inference.
Funding is evidence of a relationship.
It is not sufficient evidence of agency.
If the European Union funds an organisation advocating judicial reform that the EU also supports, there are at least two possible explanations. The organisation promotes judicial reform because Brussels pays it to do so, or Brussels funds the organisation because both independently support judicial reform.
The observed correlation is identical.
The political interpretation is not.
This is why the word “agent” is so consequential.
Agency implies direction.
It suggests the organisation acts for someone else rather than possessing its own domestic political preferences.
That can delegitimise civil society without proving actual foreign control.
Governments facing criticism naturally benefit from that interpretation.
An NGO says the government is corrupt.
The government can respond to the substance of the allegation.
Or it can ask who funds the NGO.
The second response changes the subject from “is the criticism true?” to “whose interests does the critic represent?”
Sometimes that question is legitimate.
Used systematically, it becomes a method for avoiding the first question.
This is why civil society occupies such an awkward position in democratic politics.
NGOs are not elected.
Journalists are not elected.
Think tanks are not elected.
Universities are not elected.
Yet they influence public policy.
Governments can therefore reasonably ask what makes their influence legitimate.
The answer cannot simply be that civil society represents “the people”, because it does not necessarily. An NGO with five employees and substantial foreign funding may possess far less democratic representativeness than a government elected by millions.
Its legitimacy comes from somewhere else.
Civil society contributes information, organisation and scrutiny rather than electoral authority.
A journalist investigates corruption.
A legal organisation challenges government action in court.
A policy institute produces research.
A citizens’ group organises around disability rights.
None possesses the right to govern.
Together they increase the amount of information available to citizens who do.
This is why civil society can constrain governments without itself becoming government.
The distinction is essential.
An NGO criticising legislation does not override parliament.
It provides information and mobilisation that may affect how parliament behaves.
The protests this week demonstrate that mechanism directly.
Parliament initially advanced the bill.
Citizens mobilised.
The political cost changed.
The governing party withdrew it.
Some people will describe that as government surrendering to street pressure.
But democratic representation does not end on election day.
Citizens retain the right to organise between elections, and politicians retain the right to update policy when public reaction reveals information they did not previously possess.
The important democratic boundary is that protesters do not acquire permanent authority simply because they mobilised successfully.
They stopped one bill.
They did not become Georgia’s legislature.
This is the same distinction between electoral breadth and protest intensity that appeared in Sri Lanka, Iran and Israel.
Elections determine who governs.
Protests reveal how intensely some citizens oppose what government is doing.
A competent democracy processes both signals.
The government’s withdrawal may therefore be interpreted not only as political weakness but as institutional responsiveness.
A policy produced unexpectedly intense opposition.
The government changed course.
That is how feedback can work.
The harder question is whether the reversal is durable or merely tactical.
Georgian Dream said it would explain the purpose of the legislation more clearly to the public. That suggests the underlying concern has not disappeared.
Nor should it.
Foreign influence remains a genuine issue.
The task now is designing a narrower system that targets actual foreign political control without creating an instrument capable of stigmatising independent domestic organisations.
That can be done.
Require organisations above certain thresholds to disclose funding sources.
Publish the information in accessible databases.
Apply equivalent rules to organisations regardless of whether they support or oppose government.
Distinguish private international foundations from foreign states.
Require evidence of direction or coordination before describing an organisation as an agent.
Provide judicial review.
Make penalties proportionate.
Transparency can be robust without becoming political labelling.
The fact that this is administratively harder than simply creating a broad legal category is exactly why good institutional design is difficult.
Simple rules often create large collateral effects.
Precise rules require bureaucratic competence.
There is also a geopolitical layer because protesters have repeatedly carried European Union flags. Georgia formally applied for EU membership in 2022, but unlike Ukraine and Moldova it did not initially receive candidate status. The European Commission identified reforms Georgia would need to undertake before progressing further.
The protests therefore concern more than one law.
For many demonstrators, the bill represents a choice about Georgia’s political direction: towards European integration or towards a political model associated increasingly with Russia.
That framing is powerful.
It may also oversimplify domestic politics.
Georgian Dream officially supports European integration and has governed during periods of closer institutional ties with Europe. Its supporters would reject the claim that passing one foreign-funding law means choosing Moscow.
Political systems rarely divide cleanly into “pro-European” and “pro-Russian” camps.
A government can support EU membership while adopting policies Brussels criticises.
A citizen can favour closer economic relations with Russia while opposing Russian military policy.
Geopolitical orientation exists on several dimensions.
Reducing every domestic institutional dispute to a proxy contest between Russia and Europe risks weakening Georgia’s own political agency.
Georgians should evaluate the law because of what it does to Georgian institutions, not merely because another country has something that looks similar.
This is particularly important because European influence is foreign influence too.
If the principle is that foreign money threatens sovereignty, European and American funding cannot simply be exempted because many Georgians prefer the geopolitical objectives of the donors.
Consistency requires acknowledging that Western governments and foundations also attempt to shape political environments abroad.
Democracy promotion is influence.
Human-rights funding is influence.
Development assistance is influence.
The relevant distinction should not be Russian influence bad, Western influence good.
It should be between transparent cooperation and covert control, between supporting independent institutions and secretly directing political actors.
That standard can be applied regardless of the country providing the money.
This is where sovereignty becomes more complicated in small states.
Georgia cannot realistically isolate itself from external influence.
Its economy, security and political development are deeply connected to larger powers and international institutions.
The objective therefore cannot be eliminating foreign influence.
It must be preserving enough domestic agency that Georgians still determine political outcomes.
That requires transparency.
It also requires pluralism.
If citizens can see where organisations receive money and several competing sources of information remain available, no single external actor easily monopolises political discourse.
Closing civil society may actually increase vulnerability because it concentrates information and political organisation closer to the state.
Pluralism can therefore be a security asset.
This is the opposite of the authoritarian intuition that political uniformity creates strength.
Uniformity can create apparent cohesion while removing independent sources capable of identifying mistakes or external manipulation.
Again, open systems look messier.
Their redundancy can make them more resilient.
The police response this week also deserves scrutiny. Georgian authorities used tear gas and water cannon as demonstrations intensified, while protesters threw objects and attempted to breach barriers around parliament.
The state has a legitimate obligation to protect the legislature and maintain public order.
Protesters do not gain a right to violence because they oppose a law.
The government does not gain unlimited coercive authority because some protesters behave violently.
The proportionality principle from Peru applies again.
Democratic policing distinguishes individuals according to behaviour rather than treating the crowd as one political enemy.
This is particularly important when protests concern the government’s commitment to democracy itself.
An unnecessarily repressive response can validate the protesters’ argument.
Governments need to understand the informational content of their own actions.
Police behaviour becomes political evidence.
The eventual withdrawal of the bill probably reduced the risk of further escalation.
That is another reason policy reversal should not automatically be treated as weakness.
Leaders frequently fear that conceding after protests creates incentives for future mobilisation.
It can.
But refusing to change policy simply to demonstrate strength creates another incentive problem: governments become trapped defending increasingly unpopular decisions because reversal would damage reputation.
That is sunk-cost politics.
A government should ask whether the policy remains worth implementing after new information arrives.
The protests provided new information.
The international reaction provided new information.
The government updated.
Perhaps the original proposal should never have advanced.
Changing course can still be preferable to doubling down.
Georgia’s experience therefore contrasts interestingly with China’s white-paper protests. Beijing eventually began rapidly dismantling zero-COVID restrictions after years of strict controls and growing public dissatisfaction, although the exact causal role of protests remains debated.
Authoritarian governments can update too.
The democratic difference lies partly in whether citizens can organise openly enough for political pressure itself to be considered legitimate rather than merely a threat to regime survival.
Georgia’s protesters acted through a competitive political system where opposition parties, media and civil-society organisations already exist.
That institutional context makes mobilisation easier to translate into policy feedback.
The irony is that the proposed law concerned some of the organisations helping make that feedback possible.
This is why civil society should not be understood as something existing outside the political system.
It is part of the infrastructure through which the political system learns.
Governments receive information from ministries.
They also need information from institutions that have incentives to tell them the ministry is wrong.
Independent media.
Opposition parties.
Universities.
NGOs.
Courts.
Trade unions.
Business organisations.
Each possesses biases.
Their disagreement creates informational competition.
That is useful.
A government does not need to believe every NGO.
It needs to operate in an environment where NGOs can produce evidence the government would prefer not to produce itself.
This is the same information argument that appeared in China, but within a competitive democracy.
Political pluralism creates noise.
It also makes informational monopoly harder.
Georgia’s government has now withdrawn a law critics believed threatened part of that pluralism.
The next challenge is not pretending foreign influence does not exist.
It clearly does.
Russia has interests in Georgia.
The European Union has interests in Georgia.
The United States has interests in Georgia.
International NGOs have interests.
Domestic parties have interests.
Everyone influencing politics has interests.
Transparency helps citizens evaluate them.
The danger begins when government decides that some interests are not merely interests but evidence that the people holding them are foreign agents.
That moves from disclosure towards delegitimation.
A confident democracy should be able to tell citizens who funds an argument without telling them whether they are permitted to believe it.
Perhaps that is the line Georgia needs to find.
The government has every right to protect the country’s political independence.
Civil society has every right to demand that protection does not become control.
Those objectives are not incompatible.
In fact, Georgia probably needs both for the same reason.
A country located beside a much stronger power, containing unresolved territorial conflicts and attempting to integrate with Europe needs unusually resilient institutions.
Resilience requires protection against covert foreign influence.
It also requires citizens capable of challenging their own government without being classified as instruments of somebody else’s.
The bill attempted to solve the first problem.
The protests demonstrated how easily the solution could create the second.
For now, the government has retreated.
That is not the end of the argument.
It may be evidence that Georgia’s democratic feedback mechanisms still work.
