This appeal from the grant of summary judgment to the defendant, a lawyer named Norman Wexler, presents questions regarding both the meaning of the Fair Debt Collection Practices Act, 15 U.S.C. sec.sec. 1692 et seq., and the minimum showing required to defeat a motion for summary judgment.

The Act forbids a debt collector, which Wexler is conceded to be, to "use any false, deceptive, or misleading representation or means in connection with the collection of any debt," 15 U.S.C. sec. 1692e, including "the false representation or implication that any individual is an attorney or that any communication is from an attorney." 15 U.S.C. sec. 1692e(3). A lawyer who merely rents his letterhead to a collection agency violates the Act, 15 U.S.C. sec. 1692j(a); Taylor v. Perrin, Landry, deLaunay & Durand, 103 F.3d 1232, 1235-38 (5th Cir. 1997); cf. White v. Goodman, 200 F.3d 1016, 1018 (7th Cir. 2000), for in such a case the lawyer is allowing the collection agency to impersonate him. The significance of such impersonation is that a debtor who receives a dunning letter signed by a lawyer will think that a lawyer reviewed the claim and determined that it had at least colorable merit; so if no lawyer did review the claim, the debtor will have been deceived and the purpose of the Act therefore thwarted. Similarly, a lawyer who, like Wexler, is a debt collector violates section 1692e(3) (and also section 1692e(10), which forbids "the use of any false representation or deceptive means to collect or attempt to collect any debt") if he sends a dunning letter that he has not reviewed, since his lawyer's letterhead then falsely implies that he has reviewed the creditor's claim. Avila v. Rubin, 84 F.3d 222, 228-29 (7th Cir. 1996); United States v. National Financial Services, Inc., 98 F.3d 131, 135-37 (4th Cir. 1996); Clomon v. Jackson, 988 F.2d 1314, 1320-21 (2d Cir. 1993). "A debt collection letter on an attorney's letterhead conveys authority and credibility." Crossley v. Lieberman, 868 F.2d 566, 570 (3d Cir. 1989).

The plaintiffs received dunning letters signed "Wexler & Wexler," the name under which the defendant practices law. The suit charges that no lawyer in fact reviewed the claims made in the letters before they were mailed. The defendant moved for summary judgment and supported the motion with his affidavit. The affidavit states that a lawyer reviews every individual file before the initial collection letter (the letters to the plaintiffs were initial, not follow-up, letters) is sent, that he himself reviewed the plaintiffs' files before approving the sending of collection letters to them, and that in every collection case handled by his office a lawyer "reviews each and every document in the client's file to insure the correctness of the data and the claim, paying strict attention to the various statutes of limitation which may apply . . . [and] to make sure that we comply with state requirements, such as the need for actual presence in the state, a collection agency license or a license to practice law in the state." If the lawyer approves the claim, a form letter is prepared "for review and approval [by him] before [it is] allowed to go into the mail." If "after reviewing the case file . . . the attorney is unable to verify that the client is making a valid claim, the case is pulled out and discussed with our client."

So far, so good. But pretrial discovery revealed that in a recent eight-and-a- half-month period Wexler's firm sent out 439,606 pieces of mail, which according to the deposition testimony of one of the firm's lawyers consisted overwhelmingly of collection letters. That is an average of 51,718 a month. So Wexler's firm must be large--but no, it turns out to have only three lawyers, although it has 45 other employees. Of the three lawyers, two, including the defendant, appear to be engaged in managing the firm and all three engage in normal litigation activities (Wexler's affidavit states that his firm "routinely engages in litigation with regards [sic] to collection matters"), leaving little time for them to be reviewing the routine collection activity that generates the vast bulk of the mailings. In addition, Wexler is constantly changing the form letter. We are amused (though we doubt the recipients were) by one of the variants, which Wexler dropped after being sued on account of it: "YOU ARE ABOUT TO BE TREATED IN A MANNER THAT WILL CAUSE YOU TO THINK TWICE BEFORE YOU WRITE ANOTHER WORTHLESS CHECK. OUR CLIENT HAS INSTRUCTED THAT WE NOTIFY YOU THAT YOU ARE GOING TO BE SUED UNLESS REPAYMENT IS FORTHCOMING AT ONCE." Keele v. Wexler, 149 F.3d 589, 591 (7th Cir. 1998).

Suppose that each of the three lawyers in Wexler's firm devotes four hours a day to reviewing collection files and authorizing the mailing of dunning letters and that the process of review and authorization takes an average of 15 minutes per file. That comes to 16 collection letters a day per lawyer, which is fewer than 50 collection letters a day for all three lawyers or, assuming a five-day work week, fewer than 1,000 letters a month for the entire firm--if it is really true that each lawyer follows the review and authorization procedure set forth in Wexler's affidavit. To say the least, it is difficult to see how the firm could send out fifty times that number of letters yet still have a lawyer review the file and the letter in every one.

And it gets worse. The volume of mail sent by the firm varies from week to week, and in one particularly busy week in June 2000 the firm sent out 23,342 pieces of mail-- 93 times the maximum number consistent with 15-minute review. What is more, Wexler stated at his deposition that he personally is responsible for reviewing most of thecollection letters; if he reviewed all of them himself, the estimate of 93 times the maximum consistent with 15-minute review would soar to 279 times. Our 15- minute estimate may be too high, but it could be cut in three, or for that matter in 15, without explaining the volume of collection letters.

The district judge, in explaining why despite this evidence he granted summary judgment for Wexler, characterized Wexler's affidavit as "uncontradicted and unrefuted." This characterization was possible only because the judge gave no weight at all to the volume of mail, stating that "mathematical inferences about the volume of letters sent are not probative of [Wexler's] state of mind." State of mind is not the issue. The issue is credibility, the truthfulness of the affidavit; Wexler might think he reviewed the two plaintiffs' files carefully yet be mistaken. Circumstantial evidence can create an issue of credibility. United States v. Arias, 238 F.3d 1, 3-5 (1st Cir. 2001); United States v. Drones, 218 F.3d 496, 504 (5th Cir. 2000); United States v. Premises Known as 717 South Woodward Street, 2 F.3d 529, 534 (3d Cir. 1993); United States v. Bouquett, 820 F.2d 165, 167 (6th Cir. 1987). Suppose a witness testifies that he saw a person draw a gun, but there is evidence that a truck was blocking the witness's line of sight. This would be circumstantial evidence that created an issue of the witness's credibility (though not necessarily of his state of mind--he might have thought he saw a gun, but have been mistaken). And so it is here. Wexler's testimony was made incredible, or at least highly implausible, by the evidence of the volume of mail, especially when that evidence is viewed against the background of his testimony that he personally reviewed the plaintiffs' files. The letters to the two plaintiffs were mailed in February and March 2000, respectively. They are very short form letters seeking unexceptional amounts of money; there is nothing memorable about them. Wexler gave his deposition in September, almost six months later--by which time, if his affidavit can be believed, close to 100,000 additional collection matters had passed through his hands. The volume of mail not only undermines Wexler's affidavit, but is evidence from which a reasonable jury could infer that the letters to the plaintiffs were not in fact reviewed by Wexler or any other lawyer.

Wexler argues that the plaintiffs did not conduct enough discovery, that they should have broken down the 439,606 figure into initial collection letters, follow-up collection letters, and noncollection mailings (such as receipts), and should have further broken down the figure for initial collection letters between bounced-check matters, requiring minimal lawyer review, and other matters, requiring more review. We do not agree that the plaintiffs had to do this in order to create a genuine issue of material fact regarding the defendant's compliance with the Fair Debt Collection Practices Act. Remember that there is testimony by one of the firm's lawyers that the overwhelming bulk of the mailings consist of collection letters. Granted that some of them are follow-up collection letters, which require less review than initial collection letters, and bounced-check collection letters, which also may require less review than other collection letters, we do not see how these facts could be thought to put a big dent in our 15-minute estimate, which is the sort of estimate that a reasonable jury might make and from which damaging inferences concerning the veracity of Wexler's affidavit and the review of the dunning letters sent the plaintiffs by his firm arise.

Before a follow-up letter can be responsibly approved by a lawyer, he must inspect the file to determine that there has been no partial payment or other response from the debtor, and, if interest on the debt is sought by the creditor, to determine how much interest is owing. See Miller v. McCalla, Raymer, Padrick, Cobb, Nichols, and Clark, L.L.C., 214 F.3d 872, 875-76 (7th Cir. 2000). This can be a complicated calculation. Duffy v. Landberg, 215 F.3d 871, 875 (8th Cir. 2000). In the case of a bounced-check follow-up letter, if statutory penalties are sought the lawyer must determine whether the grace period for making good on such a check has expired. See 810 ILCS 5/3-806. No doubt the initial review of a bounced-check matter takes little time, but it is hard to believe that bounced checks are so common that they could account for the bulk of these enormous mailings.

An illuminating omission from the defendant's evidence is any indication of how often, if ever, his review of a collection file leads him to refuse to approve a dunning letter and to take the matter up with the client. If this happens often, it would bolster his claim to be engaged in a process of meaningful review of clients' referrals--but it would also dramatize how little time he has to review collection files, since refusal of a referral and ensuing discussion with the client about the refusal must take more than 15 minutes. One of the other lawyers in the firm acknowledged at his deposition that "we get cases all the time where individuals claim that their checks were stolen and that there were forgeries," and he said that in each of those cases the firm would follow up, for example by sending the debtor a forgery affidavit to fill out. A lawyer would have to review the affidavit to determine whether to send a follow-up letter to the affiant.

A further difficulty with the defendant's argument that an unknown fraction of the letters are follow-up rather than initial collection letters is that in a deposition given in a similar case just a few months before his affidavit in this case, Wexler stated that if the debtor doesn't pay up after receiving the initial collection letter, the firm forthwith institutes legal proceedings unless the debtor has either died or declared bankruptcy. If this is true (it can't be--it is unbelievable that Wexler's firm would be authorized by the client to file suit no matter how trivial the amount of the unpaid debt-- but he can hardly gain a forensic benefit from making an incredible admission), there are few follow-up letters and an enormous number of legal proceedings. Those proceedings would leave the lawyers with no time to review what must be a colossal number of files.

The most practical, intuitive, and readily applied criterion for granting summary judgment is whether, if the evidence gathered in discovery were the evidence presented at trial, the party moving for summary judgment (Wexler) would be entitled to judgment as a matter of law because no reasonable jury could render a verdict for the opposing party (the two plaintiffs). E.g., Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250-52 (1986); Wallace v. SMC Pneumatics, Inc., 103 F.3d 1394, 1399 (7th Cir. 1997); Russell v. Acme-Evans Co., 51 F.3d 64, 70 (7th Cir. 1995); Huang v. BP Amoco Corp., 271 F.3d 560, 564 (3d Cir. 2001). A reasonable jury could infer from the evidence as we have summarized it that the defendant violated the Fair Debt Collection Practices Act by rubber stamping his clients' demands for payment, thus misrepresenting to the recipients of his dunning letters that a lawyer had made a minimally responsible determination that there was probable cause to believe that the recipient actually owed the amount claimed by the creditor. A reasonable jury informed of the size of the firm and the duties of the three lawyers, which leave them little time for review of collection files, informed also of the circumstances from which an estimate can be made of the time that it takes a lawyer to review such a file with sufficient care to be able to make a responsible professional judgment that a legally collectible debt is owing, and informed finally of the enormous mass of mailings, most apparently of collection letters, in relation to the number and available time of Wexler's tiny legal staff, could rationally conclude that Wexler's claim to have reviewed these plaintiffs' files was false, that he had not reviewed them (nor had any other lawyer), and therefore that he had violated the Act. These inferences are not certain, because no matter how perfunctory the firm's normal review of creditor files and dunning letters, it is always possible that Wexler for obscure reasons did actually review these two plaintiffs' files. But certainty of winning at trial is obviously not a precondition to getting a trial.

We need not determine in this case the minimum amount of lawyer review required to avoid misleading the debtor into thinking that a lawyer has made a responsible professional judgment about the existence of a legally enforceable debt. No reported decisions address that issue and it may not have to be resolved in this case. For if the jury believes the plaintiffs, Wexler gave their dunning letters no review at all, while if it believes Wexler, he gave it more than enough review. We say "more than enough" because the Act can be complied with by delegation of part of the review process to a paralegal or even to a computer program, see Avila v. Rubin, supra, 84 F.3d at 229-30; ABA Model Rule of Professional Conduct 5.3 (materially identical to Illinois Rule of Professional Conduct 5.3), provided the ultimate professional judgment concerning the existence of a valid debt is reserved to the lawyer. Avila v. Rubin, supra, 84 F.3d at 225, 229; Clomon v. Jackson, supra, 988 F.2d at 1317, 1321. "An attorney's signature implies the attorney has formed a professional judgment about the debtor's case." Avila v. Rubin, supra, 84 F.3d at 229; see ABA Model Rule of Professional Conduct Rule 5.5(b) and comment 1 (materially identical to Illinois Rule of Professional Conduct 5.5(b)). In an age of specialization, professionals are not to be criticized for identifying subroutines that paraprofessionals can adequately perform under a professional's supervision. But Wexler does not argue that he has delegated some of the review tasks (for example, ascertaining whether the amount of the claim submitted by the client is the same as the amount that appears on the form letter prepared by the nonlawyer collection agents whom Wexler employs) to nonlawyers; he argues that he performs all these tasks himself, implying that if he is not telling the truth, no one performs them. We can leave for a future case the determination of the point at which delegation is so extensive or review so perfunctory that the lawyer's supposed authorship of the dunning letter becomes a deception, not because there was no review but because it was too meager to be meaningful. We are not suggesting a 15-minute minimum; we used that assumption merely to illustrate that a rational jury might find, given the volume of mail, that it was unlikely that Wexler had actually reviewed the plaintiffs' files before authorizing dunning letters to be sent to them.

Reversed.