Magdalene M. SMOOT and Ryan M. Smoot, Plaintiffs-Appellants, v. MAZDA MOTORS OF AMERICA, INC. and Tokio Marine and Fire Insurance Company, Ltd., Defendants-Appellees, 469 F.3d 675

POSNER, Circuit Judge. The district judge, after barring the plaintiffs’ expert from testifying, dismissed this diversity personal-injury suit (the substantive issues in which are governed by Wisconsin law) on the ground that without expert testimony the plaintiffs could not prove their case. Before reviewing that ruling, we remark the confusion in the parties’ briefs concerning the elements of the diversity jurisdiction.

The jurisdictional statement in the appellants’ brief states that the federal district court’s jurisdiction was based on diversity of citizenship “and the jurisdictional amount of $75,000.” In fact diversity jurisdiction depends on the jurisdictional amount’s exceeding $75,000, exclusive of interest and costs. 28 U.S.C. § 1332(a).

The jurisdictional statement goes on to recite that the plaintiffs are citizens of Wisconsin (a proper jurisdictional allegation since the plaintiffs are natural persons) and that defendant Mazda “is a foreign corporation incorporated under the laws of the State of California.” A corporation, however, has two places of citizenship: where it is incorporated, and where it has its principal place of business. 28 U.S.C. § 1332(c)(1). If Mazda’s principal place of business is in Wisconsin, diversity is destroyed.

To ensure that litigants in diversity cases attend carefully to the dual citizenship of corporations, our Circuit Rule 28(a)(1) requires the jurisdictional statement in a diversity case to specify both the state (or other jurisdiction) in which a corporate party is incorporated and the state in which its principal place of business is located. The appellants’ jurisdictional statement violates our rule but more remarkably it does not so much as mention the second defendant, the Tokio Marine & Fire Insurance Company.

The appellees’ jurisdictional statement begins promisingly by stating that the appellants’ jurisdictional statement “is neither complete [n]or correct.” But neither, it turns out, is the appellees’. It does not mention the amount in controversy, erroneously alleged in the appellants’ statement; and concerning citizenship it violates Rule 28(a)(1) by stating that the appellees are “citizens of a different state” from the appellants, without indicating what state they are citizens of.

It turns out that the insurance company is actually a citizen of a foreign country, so that the relevant provision of the diversity statute, unmentioned in either jurisdictional statement, is 28 U.S.C. § 1332(a)(2). We asked the parties to submit supplemental jurisdictional statements.

The appellants’ supplemental statement corrects the omission of Mazda’s principal place of business (also California), but blunders with respect to the insurance company by stating that it “a corporation organized under the laws of Japan with a United States branch domiciled in the State of New York with its principal place of business located at 230 Park Ave, New York, N.Y. 10169” (emphasis added). The location of a branch office is irrelevant to diversity jurisdiction.

But reference to “domicile” and “principal place of business” naturally raises the question, unaddressed in the statement, whether this branch might be a corporation having its principal place of business in New York but incorporated elsewhere, such as Wisconsin. We might have expected the blunder to be corrected by the major Chicago law firm representing the appellees. No such luck. Its supplemental jurisdictional statement repeats that the insurance company “is a foreign corporation organized under the laws of Japan with a U.S. Branch. The principal place of business of the U.S.

Branch is New York, New York.” The fact that “Branch” is capitalized and its principal place of business alleged suggests that it might be a corporation, but at argument the appellees’ lawyer said no, it’s just a branch. When asked by one of the judges why then it was mentioned in the jurisdictional statement, the lawyer replied inconsequently that “with a U.S. Branch” is Japanese corporate lingo. The appellees’ supplemental jurisdictional statement contains two further errors. It says that the amount in controversy “allegedly” exceeds $75,000.

Actually, as we know, the amount in controversy in the appellants’ jurisdictional statement is $75,000, not $75,000 plus. In addition, the use of the words “alleged” or “allegedly” in this connection is erroneous. The amount in controversy in a diversity case is the stakes that the plaintiff or defendant alleges, and provided the allegation is not false to a “legal certainty” the amount is taken as true for purposes of jurisdiction. E.g., Mt. Healthy City School District Board of Education v. Doyle, 429 U.S. 274, 276-77, 97 S.Ct. 568, 50 L.Ed.2d 471 (1977).

In other words, “When the complaint includes a number, it controls unless [the plaintiffs] recovering that amount [in the litigation] would be legally impossible.” Rising-Moore v. Red Roof Inns, Inc., 435 F.3d 813, 815-16 (7th Cir.2006). The appellees’ use of “allegedly” suggests an inclination to question whether the amount in controversy exceeds the jurisdictional minimum, but they do not pursue the point. We are satisfied that the parties’ errors in regard to the amount in controversy are harmless, given the severity of the injuries alleged. Besides unpleasant medical treatments for Mrs.

Smoot that included her having to wear an orthopedic repositioning appliance on her jaw, she claims to have sustained permanent injuries consisting of TMJ pain, clicking, popping, and inability to open her mouth fully and eat and chew all varieties of foods, permanent scarring, and humiliation at work because she is a customer service representative and her injuries prevent her from speaking normally. (Her husband’s claim is for loss of consortium.) We conclude that the district court had jurisdiction.

But the lawyers have wasted our time as well as their own and (depending on the fee arrangements) their clients’ money. We have been plagued by the carelessness of a number of the lawyers practicing before the courts of this circuit with regard to the required contents of jurisdictional statements in diversity cases. See, e.g., BondPro Corp. v. Siemens Power Generation, Inc., 466 F.3d 562 (7th Cir.2006) (per curiam), and cases cited there; Hicklin Engineering, L.C. v. Bartell, 439 F.3d 346, 348 (7th Cir.2006); Wild v. Subscription Plus, Inc., 292 F.3d 526, 528 (7th Cir.2002).

It is time, as we noted in BondPro, that this malpractice stopped. We direct the parties to show cause within 10 days why counsel should not be sanctioned for violating Rule 28(a)(1) and mistaking the requirements of diversity jurisdiction. We ask them to consider specifically the appropriateness, as a sanction, of their being compelled to attend a continuing legal education class in federal jurisdiction. E.g., In re Maurice, 69 F.3d 830, 832, 834 (7th Cir.1995); DiPaolo v. Moran, 407 F.3d 140, 144, 146 (3d Cir.2005); In re Dragoo, 186 F.3d 614, 615-16 (5th Cir.1999).

Are we being fusspots and nitpickers in trying (so far with limited success) to enforce rules designed to ensure that federal courts do not exceed the limits that the Constitution and federal statutes impose on their jurisdiction? Does it really matter if federal courts decide on the merits cases that they are not actually authorized to decide? The sky will not fall if federal courts occasionally stray outside the proper bounds.

But the fact that limits on subject-matter jurisdiction are not waiva-ble or forfeitable — that federal courts are required to police their jurisdiction — imposes a duty of care that we are not at liberty to shirk. And since we are not investigative bodies, we need and must assure compliance with procedures designed to compel parties to federal litigation to assist us in keeping within bounds. Hence Rule 28 and hence the responsibility of lawyers who practice in the federal courts, even if only occasionally, to familiarize themselves with the principles of federal jurisdiction.

It would be delightful, but irresponsible in the extreme, for us to ignore the limits on our jurisdiction, forget the rules intended to prevent us from ignoring those limits, direct the Clerk of the court to tear out the parties’ jurisdictional statements before distributing the briefs to us, and jump directly to the merits of any case that the parties would like to litigate in federal court. To the merits, at long last. Mrs.

Smoot was driving her one-year-old Mazda at 35 to 40 m.p.h. when she struck either a chunk of asphalt that had been dislodged from the pavement (her current version) or, more likely, a large pothole (the defendants’ version — but also what Mrs. Smoot told the police officer who investigated the accident). Deployment of the airbags was triggered by the collision, causing the injuries of which she complains. The day before the accident she had received a notice from Mazda that there was “an increased risk of airbag deployment in a low speed crash or minor impact to the undercarriage” in the model that Mrs.

Smoot was driving, and that the owner should contact a Mazda dealer to have the airbag control unit reprogrammed. Her husband had made an appointment with the dealer for a few days later — too late. The windshield and front left wheel and tire of the car were damaged in the accident, apparently from the impact with whatever it collided with — asphalt or pothole. Photographs were taken, which of course did not show the airbag mechanism. The car was repaired and sold before the lawsuit and cannot be traced.

The Smoots’ lawyer wanted to base their case on the venerable common law doctrine of res ipsa loquitur (the thing speaks for itself). A plaintiff who establishes that the accident in which he was injured was of a kind that could not reasonably have been expected to occur unless the injurer had been negligent has made out a prima facie case of tort liability, which is to say has presented enough evidence to withstand a directed verdict or equivalent — enough in other words to get his case to a jury. E.g., Lambrecht v. Estate of Kaczmarczyk, 241 Wis.2d 804, 623 N.W.2d 751, 761 (2001); Peplinski v.

Fobe’s Roofing, Inc., 193 Wis.2d 6, 531 N.W.2d 597, 600 (1995). Canonical statements of the doctrine, in Wisconsin as elsewhere, require that the defendant have had exclusive control of whatever it was that caused the accident. If taken literally, this would bar applying the doctrine to a products liability case, as this case is, since “unlike an ordinary accident case the defendant in a products case has parted with possession and control of the harmful object before the accident occurs.” Welge v. Planters Lifesavers Co., 17 F.3d 209, 211 (7th Cir.1994).

But as we went on to explain, “the doctrine [of res ipsa loquitur] instantiates the broader principle, which is as applicable to a products case as to any other tort case, that an accident can itself be evidence of liability.” Id. Welge was a case governed by Illinois law, not as here by Wisconsin law. But Wisconsin too allows the doctrine to be applied in a products case as long as the product defect that is claimed to have caused the accident existed before the defendant shipped the product rather than being created by tampering or use after he parted with it. E.g., Hoven v.

Kelble, 79 Wis.2d 444, 256 N.W.2d 379, 383-84 (1977); Gierach v. Snap-On Tools Corp., 79 Wis.2d 47, 255 N.W.2d 465, 467 (1977). The defendant can contest the prima facie case with evidence that this particular accident could and did occur without negligence on his part, but unless the defendant’s evidence is conclusive the jury will have to weigh it against the general probability that established the prima facie case. Turk v. H.C. Prange Co., 18 Wis.2d 547, 119 N.W.2d 365, 370 (1963).

An older view was that if the defendant presented evidence, the presumption of liability created by evidence of res ipsa loqui-tur evaporated. Bollenbach v. Bloomenthal, 341 Ill. 539, 173 N.E. 670, 672-73 (1930), overruled by Metz v. Central Illinois Electric & Gas Co., 32 Ill.2d 446, 207 N.E.2d 305 (1965); see also Spaulding v. Chicago & Northwestern Ry., 33 Wis. 582, 591 (1873). That view was unsound, e.g., Metz v. Central Illinois Electric & Gas Co., supra, 207 N.E.2d at 307, and has long been rejected by the Wisconsin courts. Lipsky v. C.

Reiss Coal Co., 136 Wis. 307, 117 N.W. 803, 804 (1908); see also Weggeman v. Seven-Up Bottling Co., 5 Wis.2d 503, 93 N.W.2d 467, 472 (1958); Wood v. Indemnity Ins. Co. of North America, 273 Wis. 93, 76 N.W.2d 610, 613-14 (1956). The presumption created by res ipsa loquitur is not a device for forcing the defendant to present evidence, if he has any; it is, rather, the acknowledgment of a probability (what statistical theorists call a “prior probability”) that the accident was due to the defendant’s negligence.

That probability is weakened, but not necessarily to the point of extinction, by contrary evidence presented by the defendant, though the probability is not so great that it entitles the plaintiff to a directed verdict if the defendant presents no evidence. So really the term “presumption” is a misnomer as applied to res ipsa loquitur and should be replaced by “permissible inference of negligence” — as the Wisconsin cases have done. E.g., Peplinski v. Fobe’s Roofing, Inc., supra, 531 N.W.2d at 599; American Family Mutual Ins. Co. v. Dobrzynski, 88 Wis.2d 617, 277 N.W.2d 749, 754 (1979).

Turning to the specific issue presented by the appeal, we agree with the plaintiffs that in a proper case of res ipsa loquitur the plaintiff does not, at least initially, have to present expert testimony; it may be obvious to judges and jurors that the accident that befell him is the kind that rarely occurs without negligence on the part of the injurer. E.g., Lambrecht v. Estate of Kaczmarczyk, supra, 623 N.W.2d at 760 n. 14; Peplinski v. Fobe’s Roofing, Inc., supra, 531 N.W.2d at 601. A typical example is where the plaintiff is discovered after his appendectomy to have a surgeon’s sponge where his appendix was.

Francois v. Mokrohisky, 67 Wis.2d 196, 226 N.W.2d 470, 473 (1975); Utica Mutual Ins. Co. v. Ripon Cooperative, 50 Wis.2d 431, 184 N.W.2d 65, 67 (1971). This would be a similar case had the airbags deployed when Mrs. Smoot parked her car and turned off the ignition, or when while driving steadily she had blown the car’s horn. Lawson v. Mitsubishi Motor Sales of America, Inc., No. 05-CC-0257, 2006 WL 2548769 (La. Sept.6, 2006); cf. Edwards v. Ford Motor Co., 934 So.2d 221, 223-24 (La.App.2006).

Expert testimony on behalf of the plaintiff in a case based on res ipsa loquitur might seem mandatory (and this regardless of the character or strength of the defendant’s evidence) if the inference of negligence from the accident itself was obvious only to an expert. Suppose there is no sponge but when the patient wakes up he discovers that his right leg is paralyzed. A medical expert might testify that it was obvious to him (the expert) that the surgeon had sliced a nerve in the patient’s abdomen rather than that the nerve had snapped spontaneously.

But at this point the doctrine of res ipsa loquitur would drop out of the case because the expert’s evidence would have provided a complete explanation of the accident, superseding any inference that might have been drawn from the accident itself. E.g., Peplinski v. Fobe’s Roofing, Inc., supra, 531 N.W.2d at 603; Utica Mutual Ins. Co. v. Ripon Cooperative, supra, 184 N.W.2d at 69; Fehr-man v. Smirl, 25 Wis.2d 645, 131 N.W.2d 314, 318 (1964). To instruct the jury on res ipsa loquitur in such a case would merely confuse.

It would be different if all the expert had done had been to rebut evidence given by the defendant that indeed the accident might well have occurred without negligence on the defendant’s part. Refuting that evidence would just repel a challenge to the inference created by the accident itself. Despite this point, the courts are divided over whether it is ever appropriate to permit expert testimony to be given to bolster the plaintiffs invocation of res ipsa loquitur, Connors v. University Associates in Obstetrics & Gynecology, Inc., 4 F.3d 123, 127 (2d Cir.1993).

But Wisconsin, consistent with its view that the function of res ipsa loquitur is just to identify a ground for an inference of negligence, allows the plaintiff to present expert testimony to show that such an injury would indeed, despite what the defendant may have tried to show, not ordinarily occur in the absence of negligence. E.g., Lambrecht v. Estate of Kaczmarczyk, supra, 623 N.W.2d at 760 n. 14; Peplinski v. Fobe’s Roofing, Inc., supra, 531 N.W.2d at 601; Kelly v. Hartford Casualty Ins. Co., 86 Wis.2d 129, 271 N.W.2d 676, 678 (1978); Utica Mutual Ins. Co. v. Ripon Cooperative, supra, 184 N.W.2d at 67-68.

Although we have been speaking so far of “negligence” because it is primarily in negligence cases that res ipsa loquitur is invoked, this is a products liability case and the issue is not whether the defendant was negligent but whether its product, namely the car in which Mrs. Smoot was injured, was defective. However, there need be no practical difference between a claim that a product was negligently manufactured and a claim that it has a defect rendering it unreasonably dangerous, see Mesman v.

Crane Pro Services, 409 F.3d 846, 849-50 (7th Cir.2005), and so it is no surprise that, as we have seen, res ipsa loquitur is applied in products cases. It would make no difference, so far as application of the doctrine was concerned, if a car accelerated when the brake was depressed because the brake had been manufactured negligently or designed improperly- The district judge was correct, however, to reject the plaintiffs’ attempt to invoke the doctrine in this case, or, to state the point more practically, was correct to rule that the plaintiff could not prove a product defect without expert testimony.

What triggers an airbag is not the speed at which the car is traveling, but the rate of deceleration. Adnan Shaout & Charles A. Mallon, “Automotive Airbag Technology Past, Present and Future,” 13 Int’l J. Computer Applications in Technology 159, 160 (2000); Stefan Duma, Rodney Rudd & Jeff Crandall, “The Automotive Airbag System,” Professional Safety, Oct. 1998, pp. 24, 25.

By our rough calculation, if you hit a wall head on while driving at 35 miles per hour and decelerate to zero miles per hour in a tenth of a second, you’ll want your airbag to deploy because you’ll have hit the wall with the same force as if you had fallen from a window 40 feet above the ground. We don’t know the rate of deceleration of Mrs. Smoot’s car when it hit the obstacle that triggered the airbag.

Even if it was a chunk of asphalt rather than a pothole, the fact that the front of the car was damaged suggests rapid deceleration, and one could not allow a jury to speculate that it was not rapid enough to trigger a properly controlled airbag. The plaintiffs concede that a “sudden slowing” in the speed of the car by only 8 m.p.h. would have triggered a properly controlled airbag, and we cannot say as a matter of common sense or common experience that hitting a pothole or a chunk of asphalt could not cause a “sudden slowing” of the car from 35 to 27 m.p.h.

The investigator’s report depicts a pothole approximately two feet in diameter, though its depth is not indicated. By the time the judge ruled against the plaintiffs on res ipsa loquitur, discovery was closed and the plaintiffs had not retained an expert. But the judge gave them time to find one, and they did. His qualifications to testify about airbags were poor, but passing that, his study of the accident was so perfunctory that he quite rightly was barred from testifying. He flunked all three requirements of Fed.

R.Evid. 702 — that the expert’s testimony be “based upon sufficient facts or data,” that it be “the product of reliable principles and methods,” and that the expert have “applied the principles and methods reliably to the facts of the case.” Zenith Electronics Corp. v. WH-TV Broadcasting Corp., 395 F.3d 416, 418 (7th Cir.2005). The plaintiffs’ expert cannot be faulted for not having inspected the car’s airbag control unit, though his clients can be.

The car should not have been sold or repaired (it was repaired before it was sold, and the repairs included replacing the airbag control unit) before the unit was inspected. And whether or not a “spoliation of evidence” instruction would have been proper had this case gotten to a jury, see, e.g., Jagmin v. Simonds Abrasive Co., 61 Wis.2d 60, 211 N.W.2d 810, 821 (1973); Estate of Neumann v. Neumann, 242 Wis.2d 205, 626 N.W.2d 821, 841-42 (2001), the plaintiffs cannot escape the responsibility for having placed their expert in a difficult position.

Even so, he could have inquired into the circumstances behind the recall notice, into the results of the recalls (were the airbag control units found to be defective in all of the recalled vehicles? Some? None?), and into the experience of premature deployment of the airbags in Mrs. Smoot’s Mazda model. He could have tried to infer deceleration from the car’s weight and the damage to it. He did none of these things. He also did not examine another car of the same model; interview Mrs.

Smoot, the investigating police officer, or any of the mechanics who repaired the vehicle; review crash testing data for the model involved; or review technical specifications or other literature regarding the manufacture, design, or functioning of airbag systems in Mazdas. He offered the naked unsubstantiated opinion that an airbag should not deploy when the car is traveling at a speed of only 35 to 40 m.p.h. and hits something unlikely to have brought the car to a complete and sudden stop — yet Mrs. Smoot had told the police investigator that the car had been “severely jolted” by the collision.

Without expert testimony, the plaintiffs were left essentially with the recall notice plus a certain implausibility in the notion that a properly controlled airbag would deploy when a car traveling at a relatively low speed hit a chunk of asphalt (though probably it really hit a pothole). The plaintiffs have not shared with us the details of the recall.

But according to the documents available at the National Highway Traffic Safety Administration’s website, http://www-odi.nhtsa.dot.gov, the percentage of the recalled Mazdas that turned out to have a defect that would trigger airbag deployments prematurely is unknown. The recall covered approximately 214,270 vehicles. NHTSA’s investigation preceding the recall discovered 88 incidents, causing a total of 56 injuries. In 2002, Mazda stated that the eomplaint rate for improper airbag deployment for the recalled vehicles had been 14.6 per 100,000 vehicles per year.

These numbers would preclude inferring liability from the recall alone, which anyway the plaintiffs do not ask us to do. A case based on so little evidence gives rise to an inference that the plaintiffs searched no further because they were pessimistic that their case had any real merit. The judge was right to keep the case from reaching a jury. AFFIRMED.