So Sue Me (with data)

This year’s Climate and Society class is out in the field (or lab or office) completing a summer internship or thesis. They’ll be documenting their experiences one blog post at a time. Read on to see what they’re up to.

Dominick Dusseau, C+S ’18

The law is generally thought of as unwavering, qualified truths. We often view data the same way. Data is supposed to show us what is really happening when analyzing something we don’t fully understand. I have spent the past summer at the New York State Office of the Attorney General’s Environmental Protection Bureau where they combine data and the law a lot. One would think that the law and data would go hand in hand, but sometimes in this realm of black and white, a lawyer must work within shades of gray. The reason is uncertainty in our data. When we don’t have all the data—as is often the case—we don’t know if we are missing something important. We may only be seeing a part of the whole picture.

Take flood maps for example. Federal Emergency Management Agency (FEMA) flood maps are used to determine flood risk and what areas will flood from storms. But these maps are based on only a few decades of data so more likely than not, extreme and rare storms haven’t occurred in those decades. This is a sign that there is a lot of uncertainty associated with these maps because we are missing information. That missing information vacant clear in New York City, for example, when FEMA flood maps vastly underestimated the amount of flooding that occurred during Hurricane Sandy.

Flooding in the Hugh L. Carey Tunnel from Hurricane Sandy storm surge.
(Source: Wikipedia Commons Hugh L. Carey Tunnel during Hurricane Sandy by Metropolitan Transportation Authority of the State of New York licensed under CC BY 2.0)

Things get complicated very quickly when trying to minimize uncertainty in a court case. That’s because legal arguments must be grounded in logic and facts. In order for data to be defensible in a court of law, the uncertainty must low. One way to boost your argument is to use reliable sources of data, such as the U.S. government. But the uncertainty in your data can still be huge.

Say you are trying to get your neighbor to upgrade their septic system so that when it rains, the soil doesn’t fill up with water and waste. If there is more liquid than the soil can hold, your backyard gets flooded with all the fun stuff your neighbor flushed down the toilet, creating a health hazard.

You could try to sue your neighbor under a private nuisance law claim; after all, your neighbor is not allowing you to enjoy your backyard because they won’t upgrade their septic system. But your backyard hasn’t flooded yet so how can you definitively prove that a storm will drop enough rainfall for your property to get inundated with sewage?

We often estimate the probability of a certain amount of rainfall by analyzing past rain events. But extreme events are naturally rare so there is not enough data on those events to calculate how often they will actually happen. Ultimately, the problem is that you want to show something bad will happen before it actually happens yet you don’t have enough information to predict how often that catastrophic event could occur. But to get data on that extreme event, it first has to happen so that we can measure it and understand it better. A classic chicken or the egg scenario. This creates quite a conundrum for the lawyer working on your case.

The best one can do is estimate. But an estimate may not hold up to the standard of evidence necessary in civil or criminal cases. A criminal case requires proof beyond a reasonable doubt and a civil case requires a preponderance of evidence. Throw in data uncertainty and there is no knowing what court judgment will get handed down. Scientists make estimates all the time, but the real question is if that estimate is good enough for the judge or jury. It appears there is uncertainty about uncertainty.

Data uncertainty inhibits a lawyer’s ability to bring claims against people that are endangering others. That is, until we have more data. But collecting data means spending years of taking measurements and in that time, a catastrophic storm could hit. Laws are meant to maintain the health and safety of society, but the laws won’t matter if you can’t enforce them.