On Dec. 12, 2015, the marathon all-nighters were finally over and the Paris climate summit adopted a global climate agreement. The general response has been mostly enthusiastic, with reporters calling the talkshistoric” and even the “end of the fossil fuel era.”

A political science analysis, however, does not quite support such a rosy view. The Paris agreement is a good step forward, but mostly because the negotiators have finally recognized the severe limitations of multilateral diplomacy and adopted a more realistic approach.

How does the Paris agreement work?

The most important feature of the Paris agreement is that it does not impose any binding commitments or targets on the countries. In a departure from conventional multilateral agreements, negotiators in Paris agreed that countries can choose their own targets by submitting “nationally determined contributions” (NDCs) that will be reviewed in a “facilitative dialog” in 2018. These contributions are to be updated regularly, with reviews every five years beginning in 2023.

At first blush, this approach may seem odd. If countries can choose their own targets, how can the treaty enable any real cooperation beyond what countries would do otherwise?

What negotiators have learned in the past 25 years of climate talks is that forcing targets on countries simply does not work. Because there is no world government to enforce rules, commitments must be based on what countries are willing to do.

Climate negotiators are moving toward the majority view among international relations scholars today: International agreements can contribute to cooperation, but they do so within the constraints of an “anarchic” international system — a world without a world government.

The Paris agreement puts pressure on countries to submit and implement ambitious NDCs primary through its review mechanism. Countries are not punished for submitting underwhelming plans or failing to meet their stated goals, but public and regularly reviewed submissions increase the reputation cost of lacking ambition or implementation failure.

This is good news. It allows the agreement to finally do what climate treaties should have been doing all along: support domestic efforts to mitigate climate change.

One can get a sense of the ambition of these country pledges by looking at the preliminary “intended nationally determined contributions” (INDCs) that countries submitted before the climate summit. Climate Action Tracker has done the analysis.

If these contributions are executed in full, the expected rate of climate change decreases from 3.3 to 3.9 degrees Celsius (that’s if everyone continues current policies) to 2.4 to 2.7 degrees Celsius (that’s if everyone fully implements their INDC plans).

In other words, the current pledges would not get us to the stated goal of the negotiations: limiting the global temperature increase to to 2 degrees Celsius.

The Paris agreement does not enforce the implementation of the NDCs. Given how severe the enforcement problem is in climate cooperation, the “if” above is supersized.

Countries can deviate from their pledges whenever doing so is convenient to them.

Hot air in an agreement to end climate change

The agreement has a few features that initially seem exciting but are probably hot air.

The first is that the agreement explicitly mentions “efforts to limit the temperature increase” to 1.5 degrees Celsius. This non-binding and by now entirely unrealistic target has been welcomed by low-lying islands and environmental groups.

But it is hard to see how it would have any effect on actual mitigation policies on the ground. At worst, it could even be harmful — further widening the gap between what countries can achieve and what they have promised to do so in a moment of idealism in Paris.

In a first, the agreement also notes that countries strive to peak global emissions and start reducing them “as soon as possible.” However, no target date is set. The text emphasizes that countries’ emissions will peak at different times in the future. The idea of an aspirational target date for global emissions appears to be stillborn.

I would not lose sleep over this. Predicting the development of global emissions is generally very difficult and any target date would be a crap shot at best.

Slowing climate change is now every nation’s responsibility, developed or not …

A crucial innovation of the Paris agreement is that the sharp distinction between “developed” and “developing” countries has finally been relaxed. In the Kyoto Protocol, only developed countries had any emissions reductions targets. Today, with China the world’s largest emitter, such an approach is no longer sustainable.

A major virtue of the Paris agreement is that reducing greenhouse gas emissions is now seen as a collective responsibility — while at the same time it recognizes that doing so is harder for the less developed countries.

This deviation from the “common and differentiated responsibilities” approach is a major victory for the United States, which has consistently campaigned for removing the arbitrary distinction between developed and developing countries. The poor countries of South Asia, Southeast Asia and Africa have the greatest potential for increasing future emissions growth. Breaking free of the arbitrary developed-developing distinction is welcome news.

And so is figuring out how to get nations out of poverty without frying the planet

The agreement also emphasizes the need to guarantee at least $100 billion dollars in climate finance, beginning in 2025. Pressures to codify climate finance targets have increased over time. This goal is probably necessary to secure the support of the developing countries for the agreement. Implementation will be hard and the politics acrimonious, but it seems that climate finance is here to stay.

The real challenge, however, lies with the future political deals needed to allow billions of poor people to break out of energy poverty without frying the planet. Virtually all industrialization so far has relied on fossil fuels. It would be both politically impossible and immoral to expect the least developed countries not to pursue economic growth to alleviate poverty.

Managing this trade-off is one of the most important challenges of the coming decades for the environment and development communities.

So is this agreement half-empty or half-full?

As has been true of almost all climate negotiations so far, the question is whether the glass is half empty or half full. The Paris negotiators adopted a more strategic and sophisticated approach to climate change than any other Conference of Parties has so far. They deserve to be congratulated.

The future of the planet now formally hinges on voluntary pledges by countries, but this is a necessary evil. If countries had tried to negotiate binding commitments, there would be no global climate deal, and maybe no climate deal at all.

As George Monbiot put it for the Guardian in his immediate reaction to the deal, “The talks in Paris are the best there have ever been. And that is a terrible indictment.”

In the end, the future of climate mitigation remains in the hands of national governments, political parties, interest groups, sub-national jurisdictions, environmental organizations, and billions of individuals looking for a better life.

The Paris climate agreement has more potential to support decentralized climate mitigation than previous agreements have had. But international agreements will not be decisive for solving the world’s greatest collective action problem. Negotiators are catching up with this reality and recognizing the limitations of their approach. This is good news for the planet.

Johannes Urpelainen is associate professor of political science at Columbia University, where he focuses on environment politics, international cooperation and institutions and the problem of energy poverty in developing countries.