In Congressional Debate, we are taught as novices that before we start building our cases and writing our speeches, we need to do some background research: get your bearings, find some context, etc.  “Some” background research, however, often becomes none. After all, why should I spend time reading about congressional stock trading when I can get straight to argumentation? Why should I read an article on the efficacy of carbon taxes before GPTing a Neg case? Why do I need to understand the situation in Darfur when I can focus my speech on a loophole in the aid delivery mechanism?

For one reason or another, modern Congressional Debate lets debaters focus purely on arguments and more or less ignore background information. The NSDA tells us that background information is “broad knowledge on a range of diverse subjects,” like history, economics, and political science. What's especially useful about background information is that since it isn’t hyper-specific, it’s useful and applicable for years.

Take, for instance, a debate on abolishing cash bail; background knowledge would be an understanding of how cash bail works and the system in place today rather than a few arguments on why cash bail is bad. Contrary to the bill’s mechanism, cash bail will likely be around for a while, and learning about it will be helpful if you and/or your friends get into legal trouble. However, since modern Congressional Debate no longer seems to require debaters to form a genuine understanding of topics, you might be in a bit of a pickle if you are arrested down the road.

So what exactly is the problem? Debaters don’t need background information anymore. Sure, it is helpful, but no longer strictly necessary, mostly because of AI’s ability to generate arguments. Part of the problem is the ease of finding cards and constructing cases without picking up knowledge along the way. Google's new AI Overview hands debaters statistics and cards on a platter without requiring them to navigate and read through a maze of context first. Debaters no longer have to develop a solid understanding of their topics before they construct arguments–AI Overview hands them arguments before they question their topic knowledge. 

The problem with this is when background research is neglected, the educational value of debate suffers. Take away the trophies, the outrounds, and bids, and debate is an educational experience. A lion's share of that educational value comes from developing an understanding of the world around us. Where else but Student Congress have you seen high schoolers discussing something as bland as term limits with so much enthusiasm? 

Even beyond education, this is damaging to the debates as a whole. Weak background research makes for a weak round. You may not have any problems refuting arguments, crystalling speeches, or even giving a sponsor. But simultaneously, you’re only aware of the arguments you can search for, and you end up with less nuanced arguments. Today you can just search for arguments for carbon taxes, and an AI overview will give you a pretty good rundown: reducing emissions, forcing accountability, transparency, etc. However, that's all AI gives you; if you start trying to find cards on the three arguments AI hands you, you miss out on all the other potential arguments you would have found by opening up links, not to mention all the background information too. When background research is limited, the variety and nuance of argumentation is also limited. When background research doesn’t happen at all, debaters end up researching based on preconceived arguments that seem logical to them, which, oftentimes, is not the best way to go. They end up missing powerful arguments that take a bit of digging to uncover. 

As a debater, I have been incredibly guilty of this. I rush through my prep and spend nearly all of my time hunting for cards that support things I want to say, rather than building my arguments off of background research. So naturally, I’ve also been thinking of possible solutions for this background information crisis. 

This problem can seem pretty insurmountable; there’s no way to force debaters to research “properly,” and you can’t exactly quiz debaters on their background knowledge. Currently, the only time congress kids are really tested on their background knowledge is cross-ex, where other debaters can ask a wide variety of questions about the topic. What organizations such as the NSDA and other tournament hosts can do is start judging cross-ex as part of the speech score. Currently, speeches are judged out of 6, but cross-ex isn’t necessarily given a score, with some judges more or less ignoring it, especially in local circuits. By implementing a system where cross-ex is also judged but given less weight, competitors are directly incentivized to develop background knowledge to better respond to these questions and get better scores.

But there is only so much the system can do; as long as AI exists, we will always be tempted to take shortcuts. It shouldn't have to be this way. We can take 10 or 20 minutes to understand a bill before we find arguments to protect it or tear it down. We can–at bare minimum–ask AI Overview to give us an overview of the bill's topic. And afterwards, we can gain knowledge from legitimate sources. This way, we’ll develop more nuanced arguments and more engaging rounds and learn more about the topics along the way, which is what debate is all about. We tell our novices to research all the time, and it’s time for us to practice what we preach.