I've long disliked the "roadmap paragraphs" that many law reviews require you to write, usually as the last paragraph in the Introduction. Here's the one from my student article:
This Comment, in Part I, explains what speech harassment law restricts, and how it restricts it. Part II confronts the arguments, made by some courts and some commentators, that harassment law can already be justified under some of the existing First Amendment doctrines -- for example, as a time, place, or manner restriction, or a legitimate attempt to protect a captive audience -- but finds that none of the arguments has merit. Finally, Part III introduces the directed speech/undirected speech distinction, and argues that it is the most practical place to draw the line between harassing workplace speech that must be protected and harassing workplace speech that may be restricted.
Seems forced, boring, and hard to read -- yet unfortunately many roadmaps are framed pretty much the same way. (Just search for "Part I" in the same paragraph as "Part II" and "Part III" and you'll see.)
Yet I agree that having some roadmap of which part of the article will do what is useful. The key, I think, is to try to make the roadmap an integral part of the discussion, and the best way I've found is to make the Introduction itself a roadmap. Here's an example from my latest published article:
[Four hypotheticals that aim to illustrate my point omitted. -EV]
My claim is that all four cases involve the exercise of a person's presumptive right to self-defense--lethal self-defense in Katherine's case, and what I call "medical self-defense" in the others.
Part I [each part corresponds to one of the hypotheticals -EV] argues that the right to medical self-defense is supported by the long-recognized right to lethal self-defense: the right to protect your life against attack even if it means killing the attacker. The lethal self-defense right has constitutional foundations in substantive due process, in state constitutional rights to defend life and to bear arms, and perhaps in the Second Amendment. But even setting aside those constitutional roots, the right has long been recognized by statute and common law. Even if the Supreme Court stops recognizing unenumerated constitutional rights, legislatures should presumptively protect people's medical self-defense rights just as they protect people's lethal self-defense rights. While a legislature need not fund people's self-defense, it generally ought not substantially burden people's right to defend themselves.
Part II discusses one context in which medical self-defense has already been recognized: Roe v. Wade and Planned Parenthood of Southeastern Pennsylvania v. Casey secure not just a previability right to abortion as reproductive choice, but also a separate postviability right to abortion as medical self-defense when pregnancy threatens a woman's life. And it can't be that a woman has a constitutional right to protect her life using medical procedures, but only when those procedures kill a viable fetus; given that Alice has a right to defend herself even when doing so means aborting a viable fetus, Ellen and Olivia should have the same right to defend themselves through other medical procedures. Alice is free to have surgery in which a doctor inserts devices into her body to excise a fetus that, tragically, threatens her life. Ellen should likewise be free to have a procedure in which a doctor inserts chemicals into her body to destroy a tumor that threatens her life. And the government should not place substantial obstacles in the way of Olivia's having a procedure in which a doctor inserts an organ into her body to replace a failing organ that threatens her life.
Parts III and IV apply the abortion-as-self-defense and lethal self-defense analogies in more detail to experimental drugs and to compensation for organs. Part III argues that the right of medical self-defense offers extra support for the Abigail Alliance panel's controversial holding. Part IV contends that the right makes the organ sales ban presumptively improper and unconstitutional as applied to organs that are needed to protect people's lives; some concerns about organ transactions may justify regulation of organ markets, but not outright prohibition of such markets.
Part IV also argues that, while this presumption of impropriety and unconstitutionality is rebuttable, it should take much to rebut it. Recognizing medical self-defense as a constitutional or moral right means the government should need a very good reason to substantially burden that right, and any restrictions that do burden it should be as narrow as possible.
In particular, while the exercise of the right to medical self-defense may be regulated in some ways--for instance, to prevent organ robbery--such regulations can and should be far less burdensome than the current total ban on organ sales is. We respect and value self-defense rights enough that we allow lethal self-defense, despite the risk that a false claim of self-defense will be used as a cloak for murder. Rather than prophylactically banning all use of lethal force, we outlaw certain uses and rely on case-by-case decisionmaking to discover and deter these improper uses. A similar approach should apply to payments for organ transplants.
Finally, the Conclusion argues that a right to medical self-defense is not only logically supportable, but also viable both in political debate and in the judicial process. Both liberal and conservative judges and voters may be open to it, and I hope that the analogies in this Essay can be used to help persuade them.
The Introduction summarizes the argument in my article, and as it summarizes what is said in each Part, it refers to that Part. My sense is that this better lays out where everything goes, and does it without a clunky roadmap paragraph that no-one enjoys reading. One can also just include the same summary but include the part number references as footnotes.
What do others think? Does the Introduction-as-roadmap work better than the roadmap paragraph? Are there still better ways of solving this problem? Or is it a false problem -- should we just abandon roadmaps altogether?