The House of Representatives voted by a wide margin on Wednesday to reauthorize the FISA Amendments Act (FAA) for another five year term, despite some vigorous bipartisan opposition. But the supporters of sweeping electronic surveillance authority often seemed confused about what, exactly, the law authorizes—and about the causes of the intelligence failures that allowed the terror attacks of 9/11 to succeed.

The most common refrain from FAA supporters was that the law only concerned surveillance targeting “foreigners in foreign lands”—meaning it could not possibly affect the rights of Americans. Rep. Trey Gowdy (R‑SC), in an impressive display of lung power, delivered a five minute floor shout to this effect. “This bill has nothing to do with Americans on American soil,” Gowdy thundered, “This bill doesn’t implicate the Bill of Rights, any more than it implicates any other part of our Constitution, unless you think that foreign nationals who are on foreign land fall within the protections of the United States Constitution.” But Gowdy has to know that this is false, because the secretive Foreign Intelligence Surveillance Court has already ruled on at least one occasion that surveillance authorized by the FAA did violate the Fourth Amendment’s prohibition on “unreasonable searches and seizures.”

It did not rule this way, of course, because foreigners on foreign soil have Fourth Amendment rights, but because the FAA authorizes large scale surveillance of Americans’ communications. Supporters of the act suggested again and again that this can’t be true, because the law requires NSA surveillance programs to have a foreign “target.” But this is based on a misunderstanding of what “target” means in FISA. As former Deputy Attorney General David Kris explains at length in his book on the law, the “target” of a surveillance program under FAA is typically just the foreign group—such as Al Qaeda or Wikileaks—that the government is seeking information about. The FISA court approves general procedures for surveillance, but it’s NSA agents who decide which particular phone lines and e‑mail accounts will be wiretapped, and there is no explicit requirement that these particular phones and e‑mail addresses be foreign—only the program’s overall target. And of course, there is something historically very strange about imagining that surveillance can only violate the rights of named targets: The Founders abhorred “general warrants,” which they passed the Fourth Amendment to abolish, precisely because these warrants authorized searches of people and homes who were not specifically named targets, exactly as the FAA does.

Rep. Lamar Smith (R‑TX) cited an additional protection: The law, he claimed, prohibits any interception of purely domestic communications. (Think about it for a second: If the “targeting” requirement really limited surveillance to foreigners, why would you even need this additional prohibition?) But Smith conveniently omitted a crucial clause: “Known at the time of acquisition.” In other words, the NSA is generally permitted to acquire communications under FAA programs unless it knows in advance that both sides are domestic. With phone calls, that will often be obvious—but with e‑mail, it rarely will. So imagine NSA wants to download the entire contents of a GMail account that has traded messages with one of their foreign targets. Well, anyone can open a GMail account from anywhere—and even if they somehow know that account belongs to an American, any of the people he’s communicating with might be overseas—so typically those communications won’t be “known” to be entirely domestic until after they’re acquired.

Intelligence Committee Chairman Mike Rogers (R‑MI) was slightly more equivocal, seemingly acknowledging that the law might permit surveillance of Americans, but that this would happen only very rarely. The mystery here is how he could possibly know that. Sen. Ron Wyden (D‑OR) has repeatedly asked the NSA for a rough ballpark estimate of how many Americans—100? 1,000? 100,000?—have had their communications caught up in the agency’s FAA dragnets. If Rep. Rogers were correct, you’d expect the answer to be “almost none”—but instead the agency has repeatedly insisted that it is unable to provide even an approximate figure. Unless Rep. Rogers somehow knows things about the NSA’s databases that the NSA does not know, he can’t have any real basis for this claim.

Finally, Rep. Dan Lungren (R‑CA) suggested that the necessity of the FAA was demonstrated by the failures of intelligence leading up to 9/11. After all, the 9/11 Commission had again and again emphasized the central failure to “connect the dots” that would have revealed an imminent attack before it occurred—and to “connect the dots,” Lungren asserted, intelligence agencies would need still more expansive power to first “collect the dots.” This is, in a way, the most breathtakingly erroneous statements heard during Wednesday’s floor debates, because turns the 9/11 Commission’s findings completely on their head. Their report conspicuously did not identify a lack of legal authority to conduct surveillance as a serious problem: If anything, the trouble was that agencies were drowning in information they lacked the capacity to analyze and put to use. Perversely, Lungren trades on a familiar phrase—“connect the dots”—to utterly invert the Commission’s diagnosis of the causes of 9/11.

In short, the supporters of the FISA Amendments Act appear to misunderstand both what the law they just voted for actually does, and what broader defects in our intelligence system present the most serious risks to national security. If you’re wondering why we don’t have better intelligence policy, it’s not that hard to connect the dots.