Many jurisdictions have mechanisms to block self-represented filings by "vexatious litigants"—people who file many unfounded lawsuits. California, for instance, provides that if someone has lost five self-represented cases in the preceding seven years, he or she can be put on a vexatious litigant list. If they are on this list, then courts must screen their filings to see if there is a "reasonable probability that the plaintiff will prevail in the litigation."
Once they're on the list, any new filings of theirs may be screened by a court to see if there is a "reasonable probability that the plaintiff will prevail in the litigation." Depending on the circumstances, a court that finds no such reasonable probability may dismiss the lawsuit, or may require a plaintiff to put up a bond that would compensate the defendant if the plaintiff indeed eventually loses. (I oversimplify things slightly here.) This scheme is an attempt to balance the interests of plaintiffs in making their claims, including without a lawyer, with the interests of defendants in not having to face frivolous lawsuits and the interest of the public in not having courts clogged with frivolous lawsuits. For more on this, including why such rules are constitutional despite the Petition Clause, see, e.g., Wolfgram v. Wells Fargo Bank (Cal. App. 1997).
Like so much in our civil justice system, the primary responsibility for spotting vexatious litigants falls on their adversaries. A defendant who is sued may check whether the plaintiff is on the vexatious litigant list—but even if the plaintiff isn't on the list yet, the defendant may search to see whether plaintiff has filed and lost the requisite number of self-represented cases in the requisite number of years. The defendant may also more broadly see what the plaintiff has done in past cases, to see if the plaintiff might be identifiable as a vexatious litigant on other grounds, for instance a pattern of frivolous filings.
