Black's Law Dictionary is the premier source for LD definitions. This is uncontroversial. Because of it, all LDers considering the March / April vigilantism resolution should be very, very familiar with two key definitions from that source.
Black's (8th ed.) defines it as "The act of a citizen who takes the law into his or her own hands by apprehending and punishing suspected criminals."
This is important for at least two reasons. First, it makes vigilantism primarily an individual act, rather than the domain of "vigilance committees," along with the concomitant advantages (it's therefore not necessarily widespread, threatening anarchy) and disadvantages (it may lack popular support or legitimacy). Second, the definition requires that the criminals in question are only suspected, raising an epistemic hurdle that the Aff must be able to clear. Negs are going to argue that individuals lack the resources, time, motivation*, and legal knowledge to fully pursue justice, and that vigilantism is thus inherently corrupt.
*In fact, their primary motivation, it can be argued, is to speed up the process of apprehension and punishment, reducing or eliminating the rights of suspects--if there even is a process!
Black's again: "To give force or effect to (a law, etc.); to compel obedience to." This prompts an interesting and critical question: does enforcement include punishment? If not, then the Aff can argue that we're not talking about a situation where the government has failed to convict known criminals. Rather, we're talking about a government that has failed to deter crime. This is good for the affirmative, since it means that the rule of law has, if not utterly toppled, at least reached a tipping point.
I've only begun to consider the implications of these definitions. Your thoughts and suggestions, as always, are welcomed in the comments.