In Indiana "[a] person who intentionally acquires, receives, sells, or transfers fetal tissue commits unlawful transfer of fetal tissue, a Level 5 felony." Ind. Code § 35-46-5-1 .5(d). A federal district court held that several terms in this statute are unconstitutionally vague and that it must be treated as if it read: "A person who intentionally sells fetal tissue commits unlawful transfer of fetal tissue, a Level 5 felony." 289 F.Supp.3d 905 , 934-35 (S.D. Ind. 2018). The district court also held that a definitional clause is invalid. As enacted, § 35-46-5-1.5(b) reads: "As used in this section, 'fetal tissue' includes tissue, organs, or any other part of an aborted fetus." This must be treated as if it read: "As used in this section, 'fetal tissue' includes tissue or organs of an aborted fetus." The district court thus held that the words "acquires", "receives", and "transfers", and the phrase "any other part", are too uncertain to have legal force. If that is right, then big chunks of the legal system are invalid, because those words are ubiquitous in statutes, regulations, and judicial opinions.

This case began when Indiana University and three of its faculty members filed this suit, under 42 U.S.C. § 1983 , against the state's prosecuting attorneys in two counties. They asked the district court to enjoin the prosecutors from attempting to enforce any part of § 35-46-5-1.5. According to the plaintiffs, the statute not only is excessively vague but also violates the First Amendment by blocking one kind of medical research, takes the University's property without just compensation, violates the Equal Protection Clause by distinguishing fetal tissue produced by abortions from that produced by miscarriages, and violates the dormant Commerce Clause by regulating the interstate market in fetal tissue. Plaintiffs do not contend that the statute imposes an undue burden on any woman who seeks to have an abortion, nor would they have standing to make such an argument. Instead the three faculty-member plaintiffs contend that the statute interferes with medical scholarship.

As we have recounted, the district court found four words or phrases to be unconstitutionally vague. It rejected plaintiffs' theories under the First Amendment and the Equal Protection Clause of the Fourteenth Amendment. And it did not come to a conclusion with respect to the Takings Clause or the Commerce Clause, reasoning that because it could not determine what the statute means, it could not properly analyze it under either of those provisions. It then entered a permanent injunction in plaintiffs' favor and closed the case. Both sides have appealed. Plaintiffs want the whole statute enjoined, while defendants want the injunction vacated and the suit dismissed.

Justiciability is an initial problem. Indiana University, which is part of the State of Indiana, see Haynes v. Indiana University , 902 F.3d 724 , 731 (7th Cir. 2018), has sued two prosecutors who carry out state powers. The two defendants have been sued in their official capacities, which means that they must be treated as the State of Indiana. Will v. Michigan Department of State Police , 491 U.S. 58 , 109 S.Ct. 2304 , 105 L.Ed.2d 45 (1989). There is considerable doubt that federal courts are authorized to adjust grievances among different parts of a state government. See, e.g., Illinois v. Chicago , 137 F.3d 474 (7th Cir. 1998) (a state can't sue part of itself); Arlington Heights v. Regional Transportation Authority , 653 F.2d 1149 , 1150-53 (7th Cir. 1981) (part of a state can't sue the state); Branson School District RE-82 v. Romer , 161 F.3d 619 , 628 (10th Cir. 1998) (collecting cases). Cf. South Bend v. South Bend Common Council , 865 F.3d 889 (7th Cir. 2017) (a city can't sue a part of itself). The suit is saved, however, by the fact that three faculty members are plaintiffs. All three have standing to litigate in their personal capacities, and with one exception (to which we return) they present all of the complaint's legal theories.

We start with vagueness, the Due Process Clause theory on which the district court based its injunction, and then address the other four theories.

The Constitution requires criminal statutes to have a core of understandable meaning. See, e.g., Johnson v. United States , --- U.S. ----, 135 S.Ct. 2551 , 192 L.Ed.2d 569 (2015) ; United States v. Powell , 423 U.S. 87 , 96 S.Ct. 316 , 46 L.Ed.2d 228 (1975) ; Nash v. United States , 229 U.S. 373 , 33 S.Ct. 780 , 57 L.Ed. 1232 (1913). Some uncertainty at the margins does not condemn a statute. It is therefore hard to see what can be wrong with words such as "acquires," which people use and understand in normal life. A person "acquires" a car by buying it or leasing it or receiving it as a gift from a parent or spouse-or by stealing it. Even a protean word such as "reasonable" has enough of a core to allow its use in situations where rights to speak are at issue. See Thomas v. Chicago Park District , 534 U.S. 316 , 324, 122 S.Ct. 775 , 151 L.Ed.2d 783 (2002). Words such as "acquire" are materially (another protean legal word) more definite than "reasonable."

Johnson shows that uncertainty so pervasive that most of a law's potential applications are impossible to evaluate may rule out enforcement. But the district court did not deny that each of the words "acquires", "receives", and "transfers", and the phrase "any other part", has a substantial, understandable core. Instead the judge worried about the periphery.

Take "transfers." The judge thought it hard to know whether a medical researcher "transfers" fetal material by passing a pipette containing fetal tissue to someone else at the same laboratory bench. 289 F.Supp.3d at 920 . Or take the phrase "any other part." Although this ensures that the statute covers every part of a fetus, the judge thought it hard to say how things work at the level of individual cells or strands of DNA. Suppose someone in Washington state (from which much of Indiana University's fetal tissue comes) extracts a few cells from an aborted fetus and uses them to create a line of stem-cell tissue, exemplars of which (dozens of generations later) are transferred to a researcher in Indiana. Is anything derived from fetal tissue included in the phrase "any other part"? The judge did not see a clear answer. Id . at 918-19. These and similar open questions led the judge to deem the words and phrase unconstitutional.

The two prosecutors did not help their defense by professing to see answers to these and similar questions. They assured the judge that of course stem cells derived from fetal tissue are not "any other part" of a fetus. They asserted that moving a pipette across a lab bench is of course not a "transfer" of fetal tissue, because "transfer" means "convey ownership." That's implausible. Federal statutes forbid the transfer of heroin and other contraband, see, e.g., 21 U.S.C. § 841 (a)(1), and judges implementing such laws do not condition their application on a change of ownership; those laws apply to people who act as agents as well as to principals. The prosecutors also asserted that placental or umbilical cord tissue is of course not "any other part" of a fetus. Maybe: The placenta is an independent organ and so may be outside the statutory scope, but the statute does not address the topic. Prosecutors can't offer definitive interpretations of criminal laws, and one prosecutor can't bind a successor in office. More: Indiana has 90 other counties, whose prosecutors may have different ideas about the statute's scope. The district judge was right to say that, if the prosecutors' assurances are all the plaintiffs have to go on, they are at needless risk.

Yet although prosecutorial assurances should not set anyone's mind at rest, the legal system offers a way to work out the uncertainties that lurk at every statute's periphery: the judiciary. Resolving edge questions is a principal role of the courts. If the district court's approach is correct, then every time a court needs to decide a tough question about just how far a statute reaches, it should declare the law unconstitutional. That is fundamentally inconsistent with the Supreme Court's approach, under which a core of meaning is enough to reject a vagueness challenge, leaving to future adjudication the inevitable questions at the statutory margin.

A federal district judge cannot definitively interpret Ind. Code § 35-46-5-1 .5, but the state judiciary can do so. Declaratory judgments are available in Indiana under Ind. Trial Rule 57 and Ind. Code § 34-14-1-1 . Any medical researcher can file a suit with the theme: "I want to do X and fear that I will be prosecuted, so please give me a declaratory judgment that X is lawful." Yet none of the plaintiffs has filed such a suit seeking assurances about the scope of § 35-46-5-1.5. Instead of using a readily available state-law remedy for unwelcome risk, they asked a federal court to blot the law from the books. That's not how uncertainty should be addressed.

We have held this already about uncertainty under Indiana law. Indiana forbids judges, and candidates for judicial office, from making any public commitment "inconsistent with the impartial performance of the adjudicative duties of judicial office". The scope for debate about the meaning of that clause puts to shame any uncertainty about the meaning of Ind. Code § 35-46-5-1 .5, but we held the language valid, even when rights to speak are at stake-not because it is clear, but because the state offers a process to resolve debatable issues. Bauer v. Shepard , 620 F.3d 704 , 715-17 (7th Cir. 2010). We explained ( id . at 716-17 ):

Plaintiffs want us to deem the law vague by identifying situations in which state officials might take an untenably broad reading of the [language], and then predicting that they will do so. It is far preferable, however, and more respectful of our judicial colleagues in Indiana, to assume that they will act sensibly and resolve the open questions in a way that honors candidates' rights under the first amendment.

When a statute is accompanied by [a] system that can flesh out details, the due process clause permits those details to be left to that system. Parts of the Hatch Act are every bit as vague as the [language here], but in [ Civil Service Commission v. Letter Carriers , 413 U.S. 548 [ 93 S.Ct. 2880 , 37 L.Ed.2d 796 ] (1973) ] the Court held that problems of implementation could be tackled by administrative adjudication. 413 U.S. at 580 [ 93 S.Ct. 2880 ]. Similarly, in Parker v. Levy , 417 U.S. 733 [ 94 S.Ct. 2547 , 41 L.Ed.2d 439 ] (1974), the Court held that an article of the Uniform Code of Military Justice making it a court-martial offense to engage in "conduct unbecoming an officer and a gentleman" is not unconstitutionally vague, because military tribunals have elaborated on what is "unbecoming" for an officer and made it more specific than the unadorned words. The National Labor Relations Act is full of vague terms [such as "unfair labor practice"], and the National Labor Relations Board has yet to make all of them concrete, but no one supposes that the whole Act could be chucked out. The Justices have been chary of holding laws unconstitutional "on their face" precisely because they have recognized that vagueness will be reduced through a process of interpretation.

This is equally true of Ind. Code § 35-46-5-1 .5. Declaratory-judgment actions can resolve ambiguities with limited (if any) risk to medical researchers.

One thing the state judiciary might do-in addition to resolving concrete disputes such as the coverage of cells derived from fetal tissue-is read the word "intentionally" in § 35-46-5-1.5(d) to protect researchers who do not subjectively understand that they are violating the law. That is what the Supreme Court did in Screws v. United States , 325 U.S. 91 , 65 S.Ct. 1031 , 89 L.Ed. 1495 (1945), to save what is now codified as 18 U.S.C. § 242 , which one might classify as the ultimately vague criminal statute. (It forbids any act under color of law that deprives anyone of any constitutional right.) This potential for risk reduction through interpretation, plus the fact that all of the contested terms have a substantial core of ascertainable meaning, leads us to reject the district court's vagueness holding.

Our dissenting colleague calls our approach "a novel variation on Pullman abstention" (below at 544), but it is neither novel nor a form of abstention. It reflects the Supreme Court's holdings in Letter Carriers , Parker , and many other decisions, such as Rose v. Locke , 423 U.S. 48 , 96 S.Ct. 243 , 46 L.Ed.2d 185 (1975), that when considering whether statutory terms are too vague a federal court must take into account how they have been interpreted and applied. Johnson is an example. The Justices declared a statute unconstitutionally vague only after this process of interpretation had been used for more than 20 years without curtailing uncertainty. For Indiana's fetal-tissue statute, by contrast, the process has not even begun. Whether the process of interpretation is administrative (as in Letter Carriers and Bauer ) or judicial (as in Parker , Rose , and Johnson ) does not matter. What does matter is whether the process can answer important questions about the statute's scope.

Plaintiffs believe that, if they sue in federal court before the state judiciary has had a chance to interpret state law, they can ensure that the state never gets that chance. Akron v. Akron Center for Reproductive Health, Inc. , 462 U.S. 416 , 451-52, 103 S.Ct. 2481 , 76 L.Ed.2d 687 (1983), used vagueness doctrine in that fashion without citing Letter Carriers , Parker , or Rose . Akron was overruled by Planned Parenthood of Pennsylvania v. Casey , 505 U.S. 833 , 112 S.Ct. 2791 , 120 L.Ed.2d 674 (1992), and we are not aware of any other decision by the Supreme Court that has used vagueness doctrine to prevent the state judiciary from having even a chance to give the law a construction that will produce adequate clarity.

Thus we arrive at the four theories that are the subjects of plaintiffs' cross-appeal.

The parties agree that the statute does not burden a suspect class or affect a fundamental right, so the rational-basis standard applies to the equal-protection theory. The district court held that ethical considerations support a distinction between fetal tissue obtained from abortions and tissue obtained from miscarriages. 289 F.Supp.3d at 931-33 . There is a moral debate about abortion and no equivalent debate about miscarriages. The Supreme Court wrote in Washington v. Glucksberg , 521 U.S. 702 , 731, 117 S.Ct. 2258 , 138 L.Ed.2d 772 (1997), that "[t]he State ... has an interest in protecting the integrity and ethics of the medical profession." Cf. Cavel International, Inc. v. Madigan , 500 F.3d 551 (7th Cir. 2007) (ethical considerations support a ban on slaughtering horses for human consumption). Plaintiffs stress that under Roe v. Wade , 410 U.S. 113 , 158, 93 S.Ct. 705 , 35 L.Ed.2d 147 (1973), and its successors, a fetus is not a "person." See also Planned Parenthood of Indiana and Kentucky, Inc. v. Commissioner of Indiana State Department of Health , 888 F.3d 300 , rehearing en banc denied, 917 F.3d 532 , 2018 WL 3655854 , 2018 U.S. App. LEXIS 17676 (7th Cir. June 25, 2018), petition for certiorari pending, No. 18-483. But that does not eliminate the possibility of serious debate about when, if at all, it is ethical to perform medical experiments on aborted fetal tissue. The rational-basis standard does not require much in the way of justification, see FCC v. Beach Communications, Inc ., 508 U.S. 307 , 314-15, 113 S.Ct. 2096 , 124 L.Ed.2d 211 (1993), and we agree with the district court that the statute survives an equal-protection challenge. Indiana's statute may or may not be sensible-for even persons who find abortion immoral must recognize that neurological research using fetal tissue can save innocent lives-but choosing sides in an ethical debate does not condemn a law.

The First Amendment argument is a non-starter. The statute regulates conduct, not speech. See Clark v. Community for Creative Non-Violence , 468 U.S. 288 , 104 S.Ct. 3065 , 82 L.Ed.2d 221 (1984). True, plaintiffs want to use fetal tissue in research that could lead to speech, in classrooms or research papers. But a desire to obtain an input into speech does not convert regulation of conduct into regulation of speech. Surely plaintiffs do not think that researchers at Indiana University are entitled to blow up a commercial airliner, or administer dangerous pathogens to patients in the University's hospital, just so that they can observe the results and write down their findings. Nor do plaintiffs think that they are entitled to steal pens, paper, and computers from the local Office Depot so that they can write articles at lower cost. Plaintiffs assert that the statute casts "a pall of orthodoxy over the classroom", but it does not. They can say, write, and teach anything they want. They can refer to results generated by work done on aborted fetal tissue in other states and nations. But they cannot treat enforcement of rules about conduct as equivalent to prohibitions of speech. See Rumsfeld v. Forum for Academic & Institutional Rights, Inc ., 547 U.S. 47 , 126 S.Ct. 1297 , 164 L.Ed.2d 156 (2006) (Solomon Amendment does not regulate speech).

Plaintiffs contend that Indiana's law violates the Commerce Clause because much of the tissue they seek to use comes from other states. Yet the law does not discriminate against interstate commerce; it applies equally to fetal tissue from Indiana and fetal tissue from Iowa, Illinois, or Indonesia. Nor does it bear more heavily on interstate commerce as a practical matter, the subject of Pike v. Bruce Church, Inc ., 397 U.S. 137 , 90 S.Ct. 844 , 25 L.Ed.2d 174 (1970). We have held repeatedly that outright bans on particular items do not offend the dormant Commerce Clause. See, e.g., National Paint & Coatings Association v. Chicago , 45 F.3d 1124 (7th Cir. 1995) (spray paint); Park Pet Shop, Inc. v. Chicago , 872 F.3d 495 (7th Cir. 2017) (dogs from puppy mills). Plaintiffs do not contend that Indiana's law concerns a subject on which there is a "compelling need for national uniformity in regulation." General Motors Corp. v. Tracy , 519 U.S. 278 , 298 n.12, 117 S.Ct. 811 , 136 L.Ed.2d 761 (1997).

Finally, the claim under the Takings Clause is confined to the University itself. None of the three individual plaintiffs contends that he has any property interest in particular fetal tissue, so nothing has been taken from any of them. The statute does render valueless any fetal tissue, derived from abortions, owned by Indiana University. But as we observed earlier, the University, as part of Indiana, is not entitled to sue its own state. Indiana's legislature is free to decide what use (including none) to make of Indiana's property. See, e.g., Great Lakes Higher Education Corp. v. Cavazos , 911 F.2d 10 , 14-15 (7th Cir. 1990). Our decision in Illinois Clean Energy Community Foundation v. Filan, 392 F.3d 934 (7th Cir. 2004), does not hold otherwise. The Illinois Clean Energy Community Foundation is a charitable foundation funded by private donations. Confiscation of its property is a taking in a way that a state's decision about what to do with its own property never could be.

Several lesser arguments have been considered but do not require discussion. We conclude that the district court should have entered judgment in defendants' favor. The injunction is reversed, and the case is remanded for that purpose.

We should affirm. The Indiana law making it a felony to acquire, receive, sell, or transfer the "tissue, organs, or any other part of an aborted fetus" is unconstitutionally vague. As explained below, some unusual features of this law and this lawsuit lead me to that conclusion: both the State's lawyers and the authoring legislators have tried to run away from the apparent meaning of the statutory language. As a result, it's clear that the law does not give fair notice of its scope, and it effectively abandons the proper separation of powers by delegating critical policy decisions to prosecutors and judges. I agree with my colleagues, however, that plaintiffs cannot prevail on their First Amendment, Equal Protection, and Commerce Clause challenges.

The unusual features of this law, its enactment, and this lawsuit mean that the district judge's reasoning does not actually threaten to invalidate "big chunks of the legal system." It was not the district judge who found it hard to know what "transfer" means in this statute, nor was the problem that the judge "did not see a clear answer" as to what is meant by "any other part." See ante at 540-41. If the district judge erred, it was in taking the State of Indiana at its word that the statutory text does not mean what it seems to say. The subject of this litigation is not handwringing about the "periphery" of the new law. The majority's hypothesized "core of understandable meaning"-the transfer or acquisition of fetal organs-addresses a factual scenario that is not and was not actually happening. The new law is relevant only to practices as to which its application seems, based on the State's defense, unforeseen and indeterminate.

This case fits comfortably within the majority's description of what Johnson v. United States , --- U.S. ----, 135 S.Ct. 2551 , 192 L.Ed.2d 569 (2015) teaches: "that uncertainty so pervasive that most of a law's potential applications are impossible to evaluate may rule out enforcement." Ante at 540. Instead of affirming, however, the majority proposes a novel variation on Pullman abstention in which the federal courts direct plaintiffs who reasonably fear prosecution under a vague law to file a series of declaratory judgment cases in the state courts.

Plaintiffs should not have to take such extraordinary measures to determine the scope of a criminal law. Legislatures must draft criminal statutes so that ordinary people have fair notice of what acts are criminal. Skilling v. United States , 561 U.S. 358 , 402-03, 130 S.Ct. 2896 , 177 L.Ed.2d 619 (2010) ; United States v. Sylla , 790 F.3d 772 , 774-75 (7th Cir. 2015). That does not mean that the correct application of even a criminal law must be free from doubt in all cases. There will often be, perhaps will always be, a periphery where there is room for disagreement and interpretation. That possibility does not mean that a law is void for vagueness, particularly as applied to the core of matters clearly covered by the statute.

Still, there are limits. Fair notice of what is prohibited is an essential element of due process and the rule of law. Vague laws "can invite the exercise of arbitrary power ... by leaving the people in the dark about what the law demands and allowing prosecutors and courts to make it up." Sessions v. Dimaya , --- U.S. ----, 138 S.Ct. 1204 , 1223-24, 200 L.Ed.2d 549 (2018) (Gorsuch, J., concurring); see also Marinello v. United States , --- U.S. ----, 138 S.Ct. 1101 , 1108, 200 L.Ed.2d 356 (2018) ("to rely upon prosecutorial discretion to narrow the otherwise wide-ranging scope of a criminal statute's highly abstract general statutory language places great power in the hands of the prosecutor").

Vagueness doctrine is also built on the separation of powers. Legislators "may not 'abdicate their responsibilities for setting the standards of the criminal law.' " Dimaya , 138 S.Ct. at 1227 (Gorsuch, J. concurring), quoting Smith v. Goguen, 415 U.S. 566 , 575, 94 S.Ct. 1242 , 39 L.Ed.2d 605 (1974). Vague laws transfer legislative power to courts and police and prosecutors, where such power most emphatically does not belong. Id . at 1228.

The unusual record here shows just such a legislative abdication. As a result, scientists engaged in potentially invaluable medical research risk criminal prosecution based on vague statutory language that leaves the hard choices to prosecutors and judges. "Under the Constitution, the adoption of new laws restricting liberty is supposed to be a hard business, the product of an open and public debate among a large and diverse number of elected representatives. Allowing the legislature to hand off the job of lawmaking risks substituting this design for one where legislation is made easy, with a mere handful of unelected judges and prosecutors free to 'condem[n] all that [they] personally disapprove and for no better reason than [they] disapprove it.' " Dimaya , 138 S.Ct. at 1228 (Gorsuch, J., concurring), quoting Jordan v. De George , 341 U.S. 223 , 242, 71 S.Ct. 703 , 95 L.Ed. 886 (1951) (Jackson, J., dissenting).

To explain these issues and why they matter, I first provide some background on the plaintiffs and their research on Alzheimer's disease. I then explain the vagueness problems in terms of the statutory language, the unusual course of this litigation, and the legislative process that led to the challenged law. I conclude with the university's takings claim.

Plaintiffs and Their Alzheimer's Research : Few diseases are as frightening as Alzheimer's disease, a progressive disease that causes brain cells to waste away and die. The symptoms can progress from severe memory loss to declines in thinking and behavioral and social skills, spiraling downward to dementia and the loss of a person's very selfhood, and ultimately to death. Between five and ten percent of Americans aged 65 already have Alzheimer's disease. Between 40 and 50 percent of Americans who turn 85 years old will have Alzheimer's. Dkt. 77-3 at 6, 16 (Lamb Dep. I, at 18, 61).

Alzheimer's disease is the focus of extensive medical research toward treatment and possible cure. The Indiana University School of Medicine houses the National Cell Repository for Alzheimer's Disease, as well as an Alzheimer's Disease Center, one of 32 research centers funded by the National Institutes of Health on the subject. Indiana University also operates the Stark Neurosciences Research Institute, which conducts research on Alzheimer's disease and other neurological disorders.

One plaintiff, Dr. Debomoy Lahiri, is a leading neurological researcher on the university faculty. He and his colleagues specialize in the study of brain disorders, primarily Alzheimer's disease. They use tissue from aborted fetuses and derivatives from such tissue, including cells cultured in laboratories from fetal cells, as well as DNA, RNA, proteins and other molecules and sub-cellular structures obtained from such cells. The evidence in this record establishes that fetal brain tissue-or at least cells and complex molecules derived from such tissue-is essential to research on possible cures and treatments. There is no adequate substitute.

Since 2011, Dr. Lahiri has received about 25 shipments of fetal tissue from the Birth Defects Research Laboratory at the University of Washington, which receives tissue from abortions performed lawfully in the State of Washington. The record here shows that the nationally recognized laboratory adheres to strict federal and state regulations for receiving donated fetal tissue and providing the tissue for research, including ensuring (1) that all women donating fetal tissue have given informed consent, and (2) that the laboratory does not sell or profit from providing fetal material. Dkt. 77-20 at 1-3 (BDRL affidavit), The shipments do not include any intact organs but only tiny tissue samples. Each shipment includes tissue contained within approximately two teaspoons of liquid medium.

Even if one could say confidently that the Birth Defects Research Laboratory material is "fetal tissue" for purposes of the statute (and I am not so confident), it becomes even less clear once Dr. Lahiri works with the material. One critical step in research with these tiny samples of tissues is to "dissociate" and culture them-i.e., to separate the cells from each other and to place them in a laboratory dish with conditions and nutrients that allow the cells to grow, divide, and redivide, thus multiplying the material available for research.

Also relevant here, medical research is collaborative, both within Indiana University and with the NIH and other institutions. To be valuable, research must be replicable. That requires cooperation and collaboration among researchers. Dr. Lahiri and his colleagues are requested and expected to transfer tissue samples between laboratories in the course of that collaboration. Such collaboration is also, understandably, a condition of their NIH funding.

The Statutory Language, This Lawsuit, and the Legislative Process : The Indiana law poses serious vagueness problems in terms of what counts as "fetal tissue," what researchers may do with fetal tissue, and whether the law even applies to fetal tissue that is not actually sold. Some of the problems are apparent from the statutory text itself. Others have emerged in this lawsuit and examination of the legislative process that led to enactment of the law in 2016.

The statute says: "A person who intentionally acquires, receives, sells, or transfers fetal tissue commits unlawful transfer of fetal tissue, a Level 5 felony." Ind. Code § 35-46-5-1 .5(d). A key definition provides: "As used in this section, 'fetal tissue' includes tissue, organs, or any other part of an aborted fetus." § 1.5(b).

The sharpest dispute between the parties is the scope of the "any other part" phrase in the definition of fetal tissue. Let's assume the statute makes criminal the receipt of shipments of fetal tissue like those plaintiffs received in the past from the birth defects laboratory in Washington.

That raises the question whether the "any part of" definition in the Indiana law extends to make criminal the receipt of lines of cells derived from fetal tissue. The parties agree here that, as a matter of biology, the process of cell division produces a culture that contains some cells remaining from the original tissue. The new culture also contains other, newly-divided cells that were not part of the original tissue, but which may well contain complex molecules (such as DNA, RNA, or proteins) that were present in the original tissue sample. See Prosecutors' Br. at 10; Plaintiffs' Br. at 40. So, to rephrase the key question, does a second-, third-, tenth-, or nth-generation cell with such molecules fall under the statutory definition as "any other part" of an aborted fetus?

That question is not at the periphery of this statute but at its core. The statutory language, "any other part," offers no guidance. It does not signal that the legislature even recognized this critical issue. And as the law has been defended in this lawsuit, the problem has actually become even more severe. In the district court, the State took the position that "any other part" does not apply to:

(a) placental or umbilical cord blood tissue or cells, or biological material obtained from such blood, tissue or cells;

(b) fetal cell lines including but not limited to HEK293 (or biological material obtained from such cell lines) derived from the cells of an aborted fetus, but which do not contain cells which were originally part of an aborted fetus;

(c) fetal cells (or biological material obtained from such cells) which were derived from aborted fetal cells, but are not cells which were originally part of an aborted fetus.

Dkt. 75 at 2. 1

Later in the district court proceedings, the State asserted that once cells have been separated from the fetal tissue and grow, divide into new cells, and are moved from the first dish to the second dish, they can no longer be considered part of the aborted fetus. Dkt. 81 at 48. Or, the State suggests, that separation may occur even earlier in the process of culturing cells, even before the first passage, when cells are dissociated. Id . at 48.

Those limits imply that the Indiana law would not interfere with plaintiffs' research as long as the laboratory in Washington or some other lab outside of Indiana could perform the first round of cell growth and division. If the cultures of cells would not count as "fetal tissue" under the Indiana law, plaintiffs would be free to obtain and use them for research.

The vagueness problem stems from the fact that it's impossible to know what weight to give the State's assurances. The proposed limits might well be sensible as a matter of policy. But as the district judge pointed out, these proposed limits have no apparent basis in the statutory text, and the State has not explained how it derives those supposedly comforting limits from the Indiana law. The State's lawyers asserted in oral argument that their assurances would be binding on the State and its present and future prosecuting attorneys. That remains to be seen. 2

Another focus of the vagueness challenge is the key verbs in the statute. If you have "fetal tissue" within the meaning of the statute, what can you do with it? A person who intentionally "acquires, receives, sells, or transfers fetal tissue" commits a felony. § 1.5(d). Federal law, however, already makes it a federal crime to sell fetal tissue for "valuable consideration," regardless of source, see 42 U.S.C. § 289g-2(a), so "sell" is not at issue here. 3

The focus is on "acquire," "receive," and "transfer." My colleagues correctly point out that these are common, well-understood terms in criminal law. But when plaintiffs challenged the new law, the State took the following position:

The word 'transfer' in Section 1.5(b) of the Statute does not apply to physical movement of cells or biological materials received, generated or stored at Indiana University before the enactment of the Statute provided that:

(a) the physical movement is within or among laboratories, buildings or physical facilities of Indiana University by any faculty member, employee, representative or agent of Indiana University;

(b) the physical movement is the result of a faculty member, employee, representative or agent of Indiana University permanently removing such materials from Indiana University's buildings or physical facilities; or

(c) the physical movement is at the request of the National Institute of Health ('NIH') or pursuant to the terms or conditions of any NIH grant.

Dkt. 75 at 2.

The State further explained that there will be no "transfer" if there is no change in ownership or possession of the tissue, referring apparently to possession by Indiana University as an institution, not possession being transferred from one human being to another inside a university laboratory. Dkt. 90 at 16. The apparent concession that plaintiffs can move fetal tissue around at the request of the NIH or to comply with a grant sounds reasonable, but where does that come from?

As with the proposed limits on "fetal tissue" itself, there is no apparent textual basis for these limits. These proposed limits are also inconsistent with the way the law regulates transfer and possession of so many other forms of contraband, such as illegal drug, firearms, and other weapons. If one person cooks crack cocaine in a kitchen and hands a pot of the drug to a confederate at the kitchen table, federal criminal law would say there was a transfer of possession, regardless of the ownership of the drug. The same problems are inherent in the State's approach to "acquire" and "receive," proposing limits that have no apparent textual or other legal bases.

These attempts to defend the law by limiting it in these ways only add to the mystery of its meaning. One is left to wonder why the State proposes them. Members of this court repeatedly asked the State's lawyers to explain why they felt these implausible limitations were needed. No answer was forthcoming.

As best I can tell, the State's lawyers seem to be trying to prevent the new law from having effects that would flow naturally from the statutory language but that seem not to be sensible from a practical or policy perspective. Implicit in the effort is the concession that, as drafted, the operative portions of § 1.5 will have undesired and unintended effects that will interfere with important and legitimate medical research.

In other words, the State's lawyers seem to be trying to save the legislature from itself. The majority nonetheless defends the potentially extreme outcome of this statute, noting that there is "serious debate about when, if at all, it is ethical to perform medical experiments on aborted fetal tissue." Ante at 543. Thus, regardless of whether the statute is "sensible," the fact that the Indiana legislature "cho[se] sides in an ethical debate does not condemn [the] law." Ante at 543. Unfortunately, the majority describes a legislative process regarding this statutory provision that did not actually happen.

The available evidence from the Indiana legislature indicates that it adopted this criminal provision on fetal tissue research without realizing it would have any impact on medical research using fetal tissue, let alone shut down this research on Alzheimer's disease. The legislature simply did not engage with the hard policy choices and the natural consequences of its language. The result was to distill here both concerns at the center of vagueness doctrine, the fair notice and separation of powers so critical to liberty.

Section 1.5 was adopted as part of a broader bill, House Enrolled Act No. 1337, that imposed a variety of restrictions on abortions and related issues. The primary focal points of this bill have already been considered by this court. See Planned Parenthood of Indiana and Kentucky, Inc. v. Comm'r of Indiana State Dep't of Health , 888 F.3d 300 (7th Cir. 2018) (holding unconstitutional provisions prohibiting women from terminating pregnancies due to fetal genetic abnormalities, race, or gender and requiring burial or cremation of embryonic or fetal tissue), petition for cert. filed (U.S. Oct. 12, 2018) (No. 18-483); Planned Parenthood of Indiana and Kentucky, Inc. v. Comm'r of Indiana State Dep't of Health , 896 F.3d 809 (7th Cir. 2018) (holding unconstitutional provision requiring women to obtain ultrasound at least 18 hours before obtaining an abortion), petition for cert. filed (U.S. Feb. 4, 2019) (No. 18-1019).

In over six hours of archived video footage of the relevant Indiana House and Senate committees' discussions of this bill, dozens of witnesses testified in favor of and against the bill. Not one witness mentioned medical research using aborted fetal tissue. If the legislature was actually choosing sides on such a significant issue, one would expect some mention of it. See Barnhart v. Sigmon Coal Co ., 534 U.S. 438 , 469, 122 S.Ct. 941 , 151 L.Ed.2d 908 (2002) (Stevens, J., dissenting) ("I think judges as well as detectives may take into consideration the fact that a watchdog did not bark in the night."), quoting Harrison v. PPG Industries, Inc ., 446 U.S. 578 , 602, 100 S.Ct. 1889 , 64 L.Ed.2d 525 (1980) (Rehnquist, J., dissenting).

Nor was there any discussion of § 1.5 in the floor debates on the bill. The only substantive discussion of the provision I could find in the legislative history came in seven minutes of a meeting of the Indiana Senate's Health & Provider Services Committee. The discussion shows not a policy choice on an important issue but instead a startling disconnect between the statutory language and the stated intentions of the legislators. One of the bill's sponsors, Senator Holdman, said that the provision "has to do with the selling of fetal tissue. Which makes it illegal in the state of Indiana to sell fetal tissue, which basically addresses the issue we've seen around the country with Planned Parenthood and some of the things that have been going on there." Hearing, Indiana Senate Health and Provider Services Committee at 18:26-18:45 (Feb. 24, 2016), http://iga.in.gov/information/archives/2016/video/commit-tee_health_and_provider_services_3900/. 4

Senator Stoops asked Senator Holdman a series of questions about whether parents of an aborted fetus could consent to an autopsy or other research use of the tissue. Id. at 21:39-22:14. Senator Holdman said that would be possible, prompting this follow-up:

S. Stoops: Right, that is a specific autopsy. But what if the parents want to donate the fetus for research into the condition? I guess that would not be allowed. I mean an autopsy to determine maybe a genetic defect after the abortion is one thing. But I think actually allowing the fetal tissue to be used for research purposes seems to be outlawed in this bill.

S. Holdman: Well, I wouldn't say that it [research with fetal tissue] is outlawed . I think the word autopsy encompasses that research portion of what is done with that piece of tissue that is diagnosed after the fact through an autopsy. So maybe it is just a matter of semantics, here, perhaps Senator Stoop. But autopsy is the word that we landed on.

S. Miller (Committee Chair): And just if I could elaborate, I was a part of those conversations. This is up to a parent to request the autopsy. And I think at the time the parent requests the autopsy, they could list the kind of things they are looking for. And what kind of results they want as part of that investigation. So I think for the purposes that we put in the bill, that this covers it, Senator Stoops. At least I'm comfortable with that.

S. Holdman: That's right. And to your point, Madame Chair, in any case the parent has to consent to this procedure being done. It is not going to be done just because the physician or the hospital may want to have that done. It has to be consented to by the parent itself.

S. Stoops: Right. So on page 15, line 9, where it says: 'A person who intentionally acquires, or sells, etc. or transfers fetal tissue, commits unlawful transfer of fetal tissue and is a Level 5 felony.' So, is that if that person did not get permission from the parent? So otherwise if the parent gave permission, then that does not apply?

S. Holdman: That would be my understanding, yes. I thought you were going to ask me what a Level 5 Felony outcome was. And we have the Criminal Courts Chair here who could answer that question for you, as well.

Id. at 23:17-25:43

That exchange is remarkable, particularly because of the role of the pregnant woman's consent for the research. Federal law already made clear that fetal tissue could not be used for any research purposes without the consent of the pregnant woman. 42 U.S.C. § 289g-1(b). The evidence here shows that the birth defect laboratory in Washington always obtains the woman's consent to use tissue for research. And when called upon to explain § 1.5 to their colleagues, the sponsors said (a) that it would apply only to sale of fetal tissue in Indiana, and (b) that the consent of the pregnant woman for research use would amount to a complete defense to the criminal restrictions.

Now, I am fully aware that legislatures are not required to produce legislative history. They are not required to explain or discuss the language they enact. Courts generally presume that all legislators have read and understood the bills they vote upon. As we try to be faithful agents of the lawmakers, we thus apply the ordinary meaning of the statutory language. After all, as the majority points out, one can have a "serious debate about when, if at all, it is ethical to perform medical experiments on aborted fetal tissue," ultimately "choosing sides." Ante at ----.

In this instance, however, there was no such debate. Instead, the law's defenders have backed away from those ordinary meanings at every opportunity, first in the legislative process and then in the lawsuit. This highly unusual record ought to raise warning flags for us. The statute is fatally vague because its supporters did not pay enough attention to render it intelligible as applied to this medical research, which is the relevant core, not the periphery, of this statute.

Consider this warning: "Under the Constitution, the adoption of new laws restricting liberty is supposed to be a hard business, the product of an open and public debate among a large and diverse number of elected representatives. Allowing the legislature to hand off the job of lawmaking risks substituting this design for one where legislation is made easy, with a mere handful of unelected judges and prosecutors free to 'condem[n] all that [they] personally disapprove and for no better reason than [they] disapprove it.' " Dimaya , 138 S.Ct. at 1228 (Gorsuch, J., concurring), quoting Jordan , 341 U.S. at 242 , 71 S.Ct. 703 (Jackson, J., dissenting).

What happened here was just such a "hand-off" of lawmaking power. It highlights why the majority's solution here, abstaining and leaving the statute- writing task to the state courts, is not an appropriate response. The majority proposes that Indiana state courts should be permitted a decade or two to resolve "edge questions." like the "concrete disputes such as the coverage of cells derived from fetal tissue." Ante at 541-42 & 541.

For example, will state courts decide that "placental or umbilical cord tissue" counts as "any other part of a fetus"? Maybe, the majority says. On one hand, the majority muses, "[t]he placenta is an independent organ and so may be outside the statutory scope, but the statute does not address the topic." Ante at 540. On the other hand, however, Indiana permits the purchase or sale of fetal material for "fetal stem cell" research if "the biological parent has given written consent for the use of the fetal stem cells," Ind. Code § 35-46-5-3 , but later defines "fetal stem cell" to include only the "placenta," "umbilical cord," "amniotic fluid," and "fetal tissue" if "taken from a fetus that was either miscarried or stillborn"-not "any cells that are taken as a result of an abortion," Ind. Code § 16-18-2-128 .5. Although that provision is limited to commercial transactions, which § 1.5 is not, perhaps that definition will support the criminalization of the non-commercial acquisition and transfer of abortion-derived umbilical and placental cell. Plaintiffs apparently will just have to wait and see how the Indiana state courts will write that part of the statute.

The majority bases its preference for a state-court iterative process on Johnson , in which the Supreme Court declared the statute at issue vague only after struggling with it for 20 years, while here "the process has not even begun." Ante at 542. But that was not the real lesson of Johnson , even though the Court acknowledged that "persistent efforts" by courts to establish a meaningful standard for a statute can be one source of "evidence of vagueness." Johnson , 135 S.Ct. 2551 , 2558 (2015), quoting United States v. L. Cohen Grocery Co. , 255 U.S. 81 , 41 S.Ct. 298 , 65 L.Ed. 516 (1921). The Court did not signal that prosecutors and judges would have several decades to define what "edge" behavior is actually criminal just because a court hypothesizes that there must be some knowable core of prohibited conduct. To the contrary, Johnson observed that its " holdings squarely contradict the theory that a vague provision is constitutional merely because there is some conduct that clearly falls within the provision's grasp." Id . at 2561 (describing examples). And note, critically, that Johnson was only about a sentencing enhancement, not whether conduct was criminal in the first place.

In a context more similar to this case, the Court struck down a criminal statute that required "humane and sanitary" disposal of fetal remains. City of Akron v. Akron Center for Reproductive Health, Inc. , 462 U.S. 416 , 451, 103 S.Ct. 2481 , 76 L.Ed.2d 687 (1983), overruled on other grounds by Planned Parenthood of Southeastern Pennsylvania v. Casey , 505 U.S. 833 , 112 S.Ct. 2791 , 120 L.Ed.2d 674 (1992). To defend the provision, Akron argued that it clearly intended only to "preclude the mindless dumping of aborted fetuses onto garbage piles." Id. (internal quotation and citation omitted). The Court did not, however, send the matter to the state courts to work out case by case what would count on the periphery as "humane and sanitary." Rather, "[t]his level of uncertainty is fatal where criminal liability is imposed" because the statute "fails to give a physician 'fair notice that his contemplated conduct is forbidden,' United States v. Harriss , 347 U.S. 612 , 617, 74 S.Ct. 808 , 98 L.Ed. 989 (1954), [and thus] it violates the Due Process Clause." Id. at 451-52, 103 S.Ct. 2481 . Neither did the Court indulge Akron's proposal to sever "humane" from the statute: "The uncertain meaning of the phrase 'humane and sanitary' leaves doubt as to whether the city would have enacted [the statute] with the word 'sanitary' alone." Id . at 452 n.45, 103 S.Ct. 2481 . In other words, the city was free "to enact more carefully drawn regulations," id. , but that was a legislative job-not a prosecutorial or judicial one. 5

The majority also reads too much into Bauer v. Shepard , 620 F.3d 704 , 706-07 (7th Cir. 2010), which considered a challenge by two Indiana judges and an advocacy group to provisions of Indiana's Code of Judicial Conduct. In Bauer , the plaintiff-judges feared that responding to the group's questionnaires might violate the state's judicial ethics code. In that case, the Commission on Judicial Qualifications might issue public admonitions (if the subject of the admonition agreed), or a court of appeals could remove the judge from a particular case, or, as a matter of last resort, "the state's Supreme Court c[ould] remove [the] judge from office or impose substantial discipline." Id. at 707-08 . We acknowledged that "[t]o the extent there is uncertainty" regarding some provisions, "Indiana provides means of clarification," and the Commission on Judicial Qualifications "already has issued several clarifying advisory opinions." Id . at 712. Judges who worried about further applications of the Code should "wait and see," we said, given that beyond the "advisory opinions that reduce uncertainty," when the "Commission brings a proceeding the state judiciary will issue an opinion that makes the rule more concrete." Id. at 716 .

The majority quotes at length from Bauer to explain that it is "more respectful of our judicial colleagues in Indiana, to assume that they will act sensibly and resolve the open questions in a way that honors candidates' rights under the first amendment. When a statute is accompanied by [a] system that can flesh out details, the due process clause permits those details to be left to that system." Ante at 541, quoting Bauer , 620 F.3d at 716 . The bracketed "[a]" in the majority's quotation, however, stands in for Bauer's actual language: "an administrative system"-which in that case was the Commission on Judicial Qualifications and its ability to provide advisory opinions. There is no parallel administrative system here that would justify the majority's novel form of abstention. See also Bauer , 620 F.3d at 717 (describing Pullman abstention as another path to same result).

The majority's approach also overlooks the fact that we now have experience with this statute, litigated with able counsel on both sides and full discovery, including depositions and statements from internationally recognized scientific experts. It turns out that interpreting this law is a mess. In fact, the whole process makes a mockery of the standard that any "normal person" could figure out what is included in "any other part of an aborted fetus." 6

It is not the job of courts to tell legislators what processes to follow toward the final enactment of the statutory language of their choice. Indiana is absolutely permitted to, as one legislator put it, move up from being "49 out of 50" in terms of abortion restrictions, and "attempt[ ] to be the most restrictive state in the country ... shooting for 50." Senate Committee Hearing at 1:47:22-1:47:32 (Feb. 24, 2016). But it is our job to insist that the effort remain within constitutional parameters: that a new criminal law give fair notice of what is criminal and what is not, and that the legislature make the key choices rather than delegate them to prosecutors and courts. That simply did not happen with § 1.5. I would affirm the district court's injunction against its enforcement as written.

Takings Claim : Finally, we should not decide the merits of Indiana University's claim for taking of private property without compensation. That claim simply is not ripe yet. The majority mistakenly asserts that the statute "does render valueless any fetal tissue, derived from abortions, owned by Indiana University." Ante at 543. That is not consistent with the position taken by the law's defenders. They insist that Indiana University remains free to conduct research with materials on hand, and even to transfer it at the request of the NIH. Supra at 548-49.

More fundamental, the legal status of Indiana University vis-à-vis the state government is more nuanced than the majority acknowledges. For example, it is well-established that the university is an arm of the state for purposes of the Eleventh Amendment and is a state actor for purposes of the Fourteenth Amendment and other constitutional provisions. See Medlock v. Trustees of Indiana University , 738 F.3d 867 , 871 (7th Cir. 2013) ; Power v. Summers , 226 F.3d 815 , 818 (7th Cir. 2000). But Indiana law does not always treat the university as the state, particularly when it comes to property. See Sendak v.Trustees of Indiana University , 254 Ind. 390 , 260 N.E.2d 601 , 603-05 (1970) (university not subject to constitutional bar to ownership of private stock); State Board of Accounts v. Indiana University Foundation , 647 N.E.2d 342 , 352 (Ind. App. 1995) (university is "state" for purposes of State Board of Accounts statute but not in accepting and administering private gifts). The majority seems to imply that state legislators could simply confiscate private gifts and grants to the university (as distinct from a private foundation associated with it) and use them for more general state-government purposes. That would be a troubling prospect for many donors and grantors. We would do better to say that the university's takings claim is simply not yet ripe because the actual scope of the challenged statute remains in so much doubt.