Rowland v. California Men's Colony, Unit II Men's Advisory Council

Justice Souter

delivered the opinion of the Court.

Title 28 U. S. C. § 1915, providing for appearances informa pauperis, authorizes federal courts to favor any “person” meeting its criteria with a series of benefits including dispensation from the obligation to prepay fees, costs, or security for bringing, defending, or appealing a lawsuit. Here, we are asked to decide whether the term “person” as so used applies to the artificial entities listed in the definition of that term contained in 1 U. S. C. § 1. We hold that it does not, so that only a natural person may qualify for treatment in forma pauperis under § 1915.

I

Respondent California Men’s Colony, Unit II Men’s Advisory Council (Council), is a representative association of prison inmates organized at the behest of one of the petitioners, the Warden of the Colony, to advise him of complaints and recommendations from the inmates, and to communicate his administrative decisions back to them. The general prison population elects the Council’s members.

In a complaint filed in the District Court in 1989, the Council charged the petitioners, state correctional officers, with *197 violations of the Eighth and Fourteenth Amendments in discontinuing their practice of providing free tobacco to indigent inmates. The Council sought leave to proceed in forma pauperis under 28 U. S. C. § 1915(a), claiming by affidavit of the Council’s chairman that the warden forbad the Council to hold funds of its own. The District Court denied the motion for an inadequate showing of indigency, though it responded to the Council’s motion for reconsideration with a suggestion of willingness to consider an amended application containing “details of each individual’s indigency.”

On appeal, the Council was allowed to proceed in forma pauperis to enable the court to reach the very question “whether an organization, such as [the Council], may proceed in forma pauperis pursuant to 28 U. S. C. § 1915(a).” No. 90-55600 (CA9, July 20, 1990). The court requested that a lawyer represent the Council pursuant to 28 U. S. C. § 1915(d).1

The Court of Appeals reversed, 939 F. 2d 854 (CA9 1991), noting that a “person” who may be authorized by a federal court to proceed in forma pauperis under § 1915(a) may be an “association” under a definition provided in 1 U. S. C. § 1. The Council being an “association,” it was a “person” within the meaning of § 1915(a), and could proceed in forma pau-peris upon the requisite proof of its indigency. The court found it adequate proof that prison regulations prohibited the Council from maintaining a bank account, and, apparently, from owning any other asset.

We granted certiorari, 503 U. S. 905 (1992), to resolve a conflict between that decision and the holding in FDM Manufacturing Co. v. Scottsdale Ins. Co., 855 F. 2d 213 (CA5 1988) (per curiam) (“person,” within the meaning of § 1915(a), includes only natural persons). We reverse.

*198 II

A

Both § 1915(a), which the Council invoked in seeking to be excused from prepaying filing fees, and § 1915(d) employ the word “person” in controlling access to four benefits provided by § 1915 and a related statute. First, a qualifying person may “commenc[e], prosecut[e] or defen[d]... any suit, action or proceeding, civil or criminal, or appeal therein, without prepayment of fees and costs or security therefor.” 28 U. S. C. § 1915(a). Second, a court may in certain cases direct the United States to pay the person's expenses in printing the record on appeal and preparing a transcript of proceedings before a United States magistrate. § 1915(b). Third, if the person is unable to employ counsel, “[t]he court may request an attorney to represent [him].” § 1915(d). And, fourth, in an appeal, the United States will pay for a transcript of proceedings below “if the trial judge or a circuit judge certifies that the appeal is not frivolous (but presents a substantial question).” 28 U. S. C. § 753(f); see ibid, (detailing slightly different criteria for habeas proceedings).

“Persons” were not always so entitled, for the benefits of § 1915 were once available only to “citizens,” a term held, in the only two cases on the issue, to exclude corporations. See Atlantic S. S. Corp. v. Kelley, 79 F. 2d 339, 340 (CA5 1935) (construing the predecessor to § 1915); Quittner v. Motion Picture Producers & Distributors of America, Inc., 70 F. 2d 331, 332 (CA2 1934) (same). In 1959, however, Congress passed a one-sentence provision that “section 1915(a) of title 28, United States Code, is amended by deleting the word ‘citizen’ and inserting in place thereof the. word ‘person.’” Pub. L. 86-320, 73 Stat. 590. For this amendment, the sole reason cited in the legislative history was to extend the statutory benefits to aliens.2

*199 B

The relevant portion of the Dictionary Act, 1 U. S. C. § 1, provides (as it did in 1959) that

“[i]n determining the meaning of any Act of Congress, unless the context indicates otherwise—

“the wor[d] ‘person’ . . . include[s] corporations, companies, associations, firms, partnerships, societies, and joint stock companies, as well as individuals.”

See 1 U. S. C. § 1 (1958 ed.). “Context” here means the text of the Act of Congress surrounding the word at issue, or the texts of other related congressional Acts, and this is simply an instance of the word’s ordinary meaning: “[t]he part or parts of a discourse preceding or following a ‘text’ or passage or a word, or so intimately associated with it as to throw light upon its meaning.” Webster’s New International Dictionary 576 (2d ed. 1942). While “context” can carry a secondary meaning of “[associated surroundings, whether material or mental,” ibid., we doubt that the broader sense applies here. The Dictionary Act uses “context” to give an *200 instruction about how to “determin[e] the meaning of a[n] Act of Congress,” a purpose suggesting the primary sense. If Congress had meant to point further afield, as to legislative history, for example, it would have been natural to use a more spacious phrase, like “evidence of congressional intent,” in place of “context.”

If “context” thus has a narrow compass, the “indication” contemplated by 1 U. S. C. § 1 has a broader one. The Dictionary Act’s very reference to contextual “indication” bespeaks something more than an express contrary definition, and courts would hardly need direction where Congress had thought to include an express, specialized definition for the purpose of a particular Act; ordinary rules of statutory construction would prefer the specific definition over the Dictionary Act’s general one. Where a court needs help is in the awkward case where Congress provides no particular definition, but the definition in 1 U. S. C, § 1 seems not to fit. There it is that the qualification “unless the context indicates otherwise” has a real job to do, in excusing the court from forcing a square peg into a round hole.

The point at which the indication of particular meaning becomes insistent enough to excuse the poor fit is of course a matter of judgment, but one can say that “indicates” certainly imposes less of a burden than, say, “requires” or “necessitates.” One can also say that this exception from the general rule would be superfluous if the context “indicate[d] otherwise” only when use of the general definition would be incongruous enough to invoke the common mandate of statutory construction to avoid absurd results.3 See, e. g., Mc- *201 Nary v. Haitian Refugee Center, Inc., 498 U. S. 479, 496 (1991) (“It is presumable that Congress legislates with knowledge of our basic rules of statutory construction”). In fine, a contrary “indication” may raise a specter short of inanity, and with something less than syllogistic force.

H-1 HH

Four contextual features indicate that “person” in § 1915(a) refers only to individuals, the first being the provision of § 1915(d) that “[t]he court may request an attorney to represent any such person unable to employ counsel.” (Emphasis added.) This permissive language suggests that Congress assumed the court would in many cases not “request” counsel, see Mallard v. United States District Court, Southern District of Iowa, 490 U. S. 296, 301-302 (1989) (holding that § 1915(d) does not authorize mandatory appointments of counsel), leaving the “person” proceeding informa pauperis to conduct litigation on his own behalf.4 Underlying this congressional assumption are probably two others: that the “person” in question enjoys the legal capacity to appear before a court for the purpose of seeking such benefits as appointment of counsel without being represented by professional counsel beforehand, and likewise enjoys the capacity to litigate without counsel if the court chooses to provide none, in the exercise of the discretion apparently conferred by the permissive language. The state of the law, however, leaves it highly unlikely that Congress would have made either assumption about an artificial entity like an association, and thus just as unlikely that “person” in § 1915 was meant to cover more than individuals. It has been the law *202 for the better part of two centuries, for example, that a corporation may appear in the federal courts only through licensed counsel. Osborn v. President of Bank of United States, 9 Wheat. 738, 829 (1824); see Turner v. American Bar Assn., 407 F. Supp. 451, 476 (ND Tex. 1975) (citing the “long line of cases” from 1824 to the present holding that a corporation may only be represented by licensed counsel), affirmance order sub nom. Taylor v. Montgomery, 539 F. 2d 715 (CA7 1976), and aff’d sub nom. Pilla v. American Bar Assn., 542 F. 2d 56 (CA8 1976). As the courts have recognized, the rationale for that rule applies equally to all artificial entities. Thus, save in a few aberrant cases,5 the lower courts have uniformly held that 28 U. S. C. § 1654, providing that “parties may plead and conduct their own cases personally or by counsel,” does not allow corporations, partnerships, or associations to appear in federal court otherwise than through a licensed attorney. See, e. g., Eagle Associates v. Bank of Montreal, 926 F. 2d 1305 (CA2 1991) (partnership); Taylor v. Knapp, 871 F. 2d 803, 806 (CA9) (nonprofit corporation formed by prison inmates), cert. denied, 493 U. S. 868 (1989); Jones v. Niagara Frontier Transportation Authority, 722 F. 2d 20, 22 (CA2 1983) (corporation); Richdel, Inc. v. Sunspool Corp., 699 F. 2d 1366 (CA Fed. 1983) (per curiam) (corporation); Southwest Express Co. v. ICC, 670 F. 2d 53, 55 *203 (CA5 1982) (per curiam) (corporation); In re Victor Publishers, Inc., 545 F. 2d 285, 286 (CA1 1976) (per curiam) (corporation); Strong Delivery Ministry Assn. v. Board of Appeals of Cook County, 543 F. 2d 32, 34 (CA7 1976) (per curiam) (corporation); United States v. 9.19 Acres of Land, 416 F. 2d 1244, 1245 (CA6 1969) (per curiam) (corporation); Simbraw, Inc. v. United States, 367 F. 2d 373, 374 (CA3 1966) (per curiam) (corporation). Viewing § 1915(d) against the background of this tradition, its assumption that litigants proceeding in forma pauperis may represent themselves tells us that Congress was thinking in terms of “persons” who could petition courts themselves and appear pro se, that is, of natural persons only.

The second revealing feature of § 1915(d) is its description of the affidavit required by § 1915(a) as an “allegation of poverty.” Poverty, in its primary sense, is a human condition, to be “[w]anting in material riches or goods; lacking in the comforts of life; needy,” Webster’s New International Dictionary 1919 (2d ed. 1942), and it was in just such distinctly human terms that this Court had established the standard of eligibility long before Congress considered extending in forma pauperis treatment from “citizens” to “persons.” As we first said in 1948, “[w]e think an affidavit is sufficient which states that one cannot because of his poverty ‘pay or give security for the costs . . . and still be able to provide’ himself and dependents ‘with the necessities of life.’” Adkins v. E. I. DuPont de Nemours & Co., 335 U. S. 331, 339. But artificial entities do not fit this description. Whatever the state of its treasury, an association or corporation cannot be said to “lac[k] the comforts of life,” any more than one can sensibly ask whether it can provide itself, let alone its dependents, with life’s “necessities.” Artificial entities may be insolvent, but they are not well spoken of as “poor.” So eccentric a description is not lightly to be imputed to Congress.

*204 The third clue is much like the second. Section 1915(a) authorizes the courts to allow litigation without the prepayment of fees, costs, or security “by a person who makes affidavit that he is unable to pay such costs or give security therefor,” and requires that the affidavit also “state the nature of the action, defense or appeal and affiant’s belief that he is entitled to redress.” Because artificial entities cannot take oaths, they cannot make affidavits. See, e. g., In re Empire Refining Co., 1 F. Supp. 548, 549 (SD Cal. 1932) (“It is, of course, conceded that a corporation cannot make an affidavit in its corporate name. It is an inanimate thing incapable of voicing an oath”); Moya Enterprises, Inc. v. Harry Anderson Trucking, Inc., 162 Ga. App. 39, 290 S. E. 2d 145 (1982); Strand Restaurant Co. v. Parks Engineering Co., 91 A. 2d 711 (D. C. 1952); 9A T. Bjur & C. Slezak, Fletcher Cyclopedia of Law of Private Corporations §4629 (Perm. ed. 1992) (“A document purporting to be the affidavit of a corporation is void, since a corporation cannot make a sworn statement”) (footnote omitted).

Of course, it is true that courts have often coupled this recognition of a corporation’s incapacity to make an affidavit with a willingness to accept the affidavit of a corporate officer or agent on its behalf even when the applicable statute makes no express provision for doing so. See, e. g., In re Ben Weiss Co., 271 F. 2d 234 (CA7 1959). Any such accommodation would raise at least three difficulties in this particular statutory context, however. There would be, first, the frequent problem of establishing an affiant’s authorization. The artificial entities covered by “person” in the Dictionary Act include not only corporations, for which lines of authority are well established by state law, but also amorphous legal creatures like the unincorporated association before us here. A court may not as readily determine whether a member of such an association, even a member styled as “president” or “chairman” or whatnot, has any business purporting to bind it by affidavit. Next, some weight should probably be given *205 to the requirement of § 1915(a) that the affidavit state the “affiant’s belief that he is entitled to redress” (emphasis added). “He,” read naturally, refers to the “affiant” as the person claiming informa pauperis entitlement. If the affi-ant is an agent making an affidavit on behalf of an artificial entity, however, it would wrench the rules of grammar to read “he” as referring to the entity.6 Finally, and most significantly, the affidavit requirement cannot serve its deterrent function fully when applied to artificial entities. We said in Adkins that “[o]ne who makes this affidavit exposes himself ‘to the pains of perjury in a case of bad faith.’ . . . This constitutes a sanction important in protection of the public against a false or fraudulent invocation of the statute’s benefits.” Adkins, supra, at 338 (quoting Pothier v. Rodman, 261 U. S. 307, 309 (1923)), The perjury sanction thus serves to protect the public against misuse of public funds by a litigant with adequate funds of his own, and against the filing of “frivolous or malicious” lawsuits funded from the public purse. 28 U. S. C. §§ 1915(a), 1915(d). The force of these sanctions pales when applied to artificial persons, however. Natural persons can be imprisoned for perjury, but artificial entities can only be fined. And while a monetary sanction may mean something to an entity whose agent has lied about its ability to pay costs or security, it has no teeth *206 when the lie goes only to belief of entitlement to redress.7 So far, then, as Congress assumed that the threat of a perjury conviction couid deter an impoverished “person” from filing a frivolous or malicious lawsuit, it probably assumed that the person was an individual.

The fourth clue to congressional understanding is the failure of § 1915 even to hint at a resolution of the issues raised by applying an “inability to pay” standard to artificial entities. It is true, of course, that because artificial entities have no use for food or the other “necessities of life,” Congress could not have intended the courts to apply the traditional “inability to pay” criterion to such entities. Yet no alternative standard can be discerned in the language of § 1915, and we can find no obvious analogy to the “necessities of life” in the organizational context. Although the most promising candidate might seem to be commercial-law “insolvency,” commercial law actually knows a number of different insolvency concepts. See, e. g., 11 U. S. C. § 101(32) (1988 ed., Supp. III) (defining insolvency as used in the Federal Bankruptcy Code); Kreps v. Commissioner, 351 F. 2d 1, 9 (CA2 1965) (discussing a type of “equity” insolvency); Uniform Commercial Code § 1-201(23), 1 U. L. A. 65 (1989) (combining three' different types of insolvency). In any event, since it is common knowledge that corporations can often perfectly well pay court costs and retain paid legal counsel in spite of being temporarily “insolvent” under any or all of these definitions, it is far from clear that corporate insolvency is appropriately analogous to individual indigency.8

*207 If § 1915 yields no “inability to pay” standard applicable to artificial entities, neither does it guide courts in determining when to “pierce the veil” of the entity, that is, when to look beyond the entity to its owners or members in determining ability to pay. Because courts would necessarily have to do just this to avoid abuse, congressional silence on the subject indicates that Congress simply was not thinking in terms of granting informa pauperis status to artificial entities.

While the courts that have nonetheless held § 1915 applicable to artificial entities have devised their own tests for telling when to “pierce the veil” for a look at individual members or owners, none of their tests is based on the language of § 1915 or on any assumption implicit in it. For example, the leading opinion on the subject, a dissent from a majority opinion that never reached the issue, appears to frame the issue as whether the individual shareholders of a corporation “have adopted the corporate form as a subterfuge to avoid the payment of court costs.” S. O. U P., Inc. v. FTC, 146 U. S. App. D. C. 66, 68, 449 F. 2d 1142, 1144 (1971) (Bazelon, C. J., dissenting) (footnote omitted). While this test certainly emphasizes why we could hardly hold that a court should never look beyond the organization to its individuals, it stems from nothing in §1915 suggesting that entities claiming to have slight assets should be treated in forma pauperis unless they were organized to cheat the courts.9

*208 The Council makes the argument, apparently accepted by the court below, that however difficult it might be to formulate comprehensive rules for determining organizational eligibility to file informa pauperis, we are excused from facing the difficulty in this case, because the Council’s circumstances would make it eligible under any set of rules. But we cannot construe tlie statute very well by sidestepping the implications of deciding one way or the other, and even if we did assume that some narrow band of eligibility escaped the contrary contextual indicators, it is not wholly clear that the Council could conclusively establish informa pauperis entitlement. It is not obvious, for example, why the Council’s inability to maintain a separate bank account should conclusively establish pauper status under § 1915, any more than a bank account with a one-cent balance would be conclusive. Account or no account, the Council, like thousands of other associations, appears to have no source of revenue but the donations of its members. If members with funds must donate to pay court fees, why should it make a conclusive legal difference whether they are able to donate indirectly through an intermediate bank account, or through one member who transmits donations by making a payment to the federal court when the Council files a complaint?10 Thus, recognizing the possibility of an organizational in forma pauperis status even in the supposedly “extreme” case of the Council would force us to delve into the difficult issues of policy and administration without any guidance from § 1915. This con *209 text of congressional silence on these issues indicates the natural character of a § 1915 “person.”11

IV

We do not forget our cases holding that the broad definition of “person” in 1 U. S. C. § 1 applies in spite of incongruities as strong, or stronger, than those produced by the four contextual features we have noted in § 1915. But in each of these cases, some other aspect of statutory context independently indicated the broad reading. In Wilson v. Omaha Tribe, 442 U. S. 653, 666 (1979), for example, we held that a statutory burden of proof on a “white person” involved in a property dispute with an Indian applied to the artificial “persons” listed in the Dictionary Act as well as to individuals. Because a wholly legal creature has no color, and belongs to no race, the use of the adjective “white” to describe a “person” is one of the strongest contextual indicators imaginable that “person” covers only individuals, and if there had been no more to the context at issue in Omaha Tribe, we would have to concede that our decision in that case is inconsistent with our conclusion here. But Omaha Tribe involved another important, countervailing contextual indication. The larger context of the whole statute and other laws *210 related to it revealed that the statute’s purpose was “to protect Indians from claims made by non-Indian squatters on their lands,” id., at 665, and we recognized that construing the disability placed on “white persons” by the statute as extending only to individuals would virtually frustrate this purpose. “[I]n terms of the protective purposes of the Acts of which [the property dispute provision was] a part, it would make little sense to construe the provision so that individuals, otherwise subject to its burdens, could escape its reach merely by incorporating and carrying on business as usual.” Id., at 666.

United States v. A & P Trucking Co., 358 U. S. 121 (1958), is a comparable case, involving two criminal statutes applying to truckers, one of which expressly applied to partnerships, and the other of which imposed criminal liability on “whoever” knowingly violated Interstate Commerce Commission regulations on transporting dangerous articles. The issue was whether partnerships could violate the statutes. We noted that the statutes required proof of knowing violations, and that a partnership at common law was deemed not to be a separate entity for purposes of suit. Id., at 124. Nonetheless, given that “[t]he purpose of both statutes [was] clear: to ensure compliance by motor carriers, among others, with safety and other requirements laid down by the Interstate Commerce Commission in the exercise of its statutory duty to regulate the operations of interstate carriers for hire,” id., at 123-124, we concluded that it would make no sense if motor carriers could avoid criminal liability for violating the trucking regulations “merely because of the form under which they were organized to do business,” id., at 124 (footnote omitted).

Thus, in both Omaha Tribe and A & P Trucking Co., we found that the statutes in question manifested a purpose that would be substantially frustrated if we did not construe the statute to reach artificial entities. Section 1915, however, manifests no such single purpose subject to substantial frus *211 tration by limiting the statutory reach to natural persons. Denying artificial entities the benefits of § 1915 will not in any sense render nugatory the benefits that § 1915 still provides to individuals. Thus, Omaha Tribe and A & P Trucking Co. confirm our focus on context, but turned on contextual indicators not present here.12

V

The Council argues that denying it informa pauperis status would place an unconstitutional burden on its members’' First Amendment rights to associate, to avoid which we should construe § 1915 broadly. See, e. g., NLRB v. Catholic Bishop of Chicago, 440 U. S. 490, 500 (1979) (“[A]n Act of Congress ought not be construed to violate the Constitution if any other possible construction remains available”). We find no merit in this argument. It is true that to file a suit in forma pauperis, not in the Council’s name, as such, but under the title “X, Y, and Z, known as the Council v. Rowland,” X, Y, and Z would each need to file an affidavit stating that he met the indigency requirements of § 1915. Nothing, however, in § 1915 suggests that the requirements would be less burdensome if the suit were titled “The Council v. Rowland”; even if we held that an association could proceed in forma pauperis, our prior discussion shows that a court could hardly ignore the assets of the association’s members in making the indigency determination. Because the extension of § 1915 to artificial entities need not lighten its practi *212 cal requirements, the limitation of § 1915 to individuals puts no unconstitutional burden on the right to associate in the manner suggested.

VI

The judgment of the Court of Appeals is reversed, and the case is remanded with instructions that the case be remanded to the District Court, where the motion for leave to file informa pauperis must be denied.

So ordered.

For a description of § 1915(d) and its relationship to § 1915(a), see infra, at 198, 203.

The House Report noted three reasons for “extend[ing] the same privilege of proceedings in forma pauperis as is now afforded citizens.” H. R. Rep. No. 650, 86th Cong., 1st Sess., 2 (1959). First, “[i]t is the opinion of *199 the Department of Justice that this proposal would be consonant with the ideas or policies of the United States.” Ibid. Second, “the Judicial Conference of the United States in recommending this legislation pointed out that the distinction between citizens and aliens as contained in existing law may be unconstitutional.” Ibid. Third, “it may also be in violation of various treaties entered into by the United States with foreign countries which guarantees [sic] to their citizens access of the courts of the United States on the same terms as American citizens.” Ibid.; see also S. Rep. No. 947, 86th Cong., 1st Sess., 2 (1959) (quoting the portion of the House Report containing these three reasons). None of these reasons supports extension of § 1916 benefits to artificial entities, or suggests that anyone involved with drafting or evaluating this legislation was thinking of such an extension. The House debate on the bill contains a discussion about the deportation of alien criminals, a matter which obviously concerns only natural persons, see 106 Cong. Rec. 13714 (1969) (remarks of Rep. Gross and Rep. Rogers); otherwise, the congressional debates provide no additional information, see ibid.; id., at 18909 (remarks of Sen. Eastland).

This rule has been applied throughout the history of 1 U. S. C. § 1 and its predecessors. See, e. g., Green v. Bock Laundry Machine Co., 490 U. S. 504, 510-511 (1989); Trans Alaska Pipeline Rate Cases, 436 U. S. 631, 643 (1978); Commissioner v. Brown, 380 U. S. 563, 571 (1965); Helvering v. Hammel, 311 U. S. 504, 510-611 (1941); United States v. Katz, 271 U. S. 354, 357 (1926); Caminetti v. United States, 242 U. S. 470, 490 (1917); United States v. Kirby, 7 Wall. 482, 486-487 (1869).

This assumption reflects a reality well known within the legal community. See, e. g., Turner, When Prisoners Sue: A Study of Prisoner Section 1983 Suits in the Federal Courts, 92 Harv. L. Rev. 610, 617 (1979) (study of 42 U. S. C. § 1983 cases filed by prisoners in five districts found that the “overwhelming majority” of cases were filed informa pauperis, and that “almost all” the cases were filed pro se).

Two federal cases cited by respondent are the only two, of which we are aware, to hold that artificial entities may be represented by persons who are not licensed attorneys: United States v. Reeves, 431 F. 2d 1187 (CA9 1970) (per curiam) (partner can appear on behalf of a partnership), and In re Holliday’s Tax Services, Inc., 417 F. Supp. 182 (EDNY 1976) (sole shareholder can appear for a closely held corporation), affirmance order sub nom. Holliday’s Tax Services, Inc. v. Hauptman, 614 F. 2d 1287 (CA2 1979). These cases neither follow federal precedent, nor have themselves been followed. See, e. g., Eagle Associates v. Bank of Montreal, 926 F. 2d 1305, 1309-1310 (CA2 1991) (criticizing and reiusing to follow Reeves); Jones v. Niagara Frontier Transportation Authority, 722 F. 2d 20, 22, n. 3 (CA2 1983) (distinguishing and narrowing Holliday’s Tax Services).

On occasion, when a party is a minor or incompetent, or fails to cooperate with appointed counsel, or is for some other reason unable to file a timely affidavit, we will accept an affidavit from a guardian ad litem or an attorney. By accepting such an affidavit, we bend the requirement that the affiant state that “he” is indigent and that “he” believes “he” is entitled to relief. In such a case, however, it is clear that the party himself is a “person” within the meaning of § 1915. The only question is whether Congress intended to deny § 1915 benefits to such a person who for some reason peculiar to him is' disabled from filing an affidavit. It is quite a different question whether Congress intended to extend § 1915 to entities that, by their nature, could never meet the statute’s requirements.

We are not ignoring the fact that the individual who made the affidavit as the entity’s agent could still be prosecuted for perjury. However, this is clearly a “second-best” solution; the law does not normally presume that corporate misbehavior can adequately be deterred solely by threatening to punish individual agents.

One plausible motive for Congress to include artificial entities within the meaning of “person” in § 1916 would be to aid organizations in bankruptcy proceedings. But the fact that the law has been settled for almost 20 years that § 1916(a) does not apply to bankruptcy proceedings, see *207 United States v. Kras, 409 U. S. 434, 440 (1973), would seem to foreclose speculation about such a motive.

Two other decisions allowing organizations to proceed in forma pau-peris appear to place importance on the “public interest” character of the organization or the litigation in question. See River Valley, Inc. v. Dubuque County, 63 F. R. D. 123, 125 (ND Iowa 1974) (noting that the corporation at issue “was formed ... for the purpose of assisting the poor and underprivileged”); Harlem River Consumers Cooperative, Inc. v. Associated Grocers of Harlem, Inc., 71 F. R. D. 93, 96 (SDNY 1976) (finding that “[t]here is a public interest quality to the stated goal for which the corporation was formed” and that “there is a public interest aspect to any private suit for treble damages under the antitrust laws”). The language of *208 § 1915, however, suggests indifference to the character of the litigant and to the type of litigation pursued, so long as it is not frivolous or malicious.

There is no evidence in the record suggesting that an inmate would not be allowed to donate part of the Council’s court costs directly from his personal account to the court, or that the inmates could not coordinate such donations.

Justice Thomas asserts that, by drawing an inference from congressional silence, we “deparft] from the definition of ‘context’ set out at the beginning of [our] opinion.” Post, at 221, n. 9. It is not from some dimensionless void, however, that we draw our conclusion. Rather, it is from a pointed silence in the face of obvious problems created by applying to artificial entities the text of § 1915, in this case the requirement that the person seeking in forma pauperis status be “unable to pay” costs, fees, and security.' As the dissent is willing to affirm without itself addressing these problems, it is apparently confident that workable, uncontroversial solutions can be drawn from the statute. Yet the rule it would affirm (that an unincorporated association is “unable to pay” whenever its “chairman” says that it cannot maintain a bank account in its own name) does not inspire confidence.

Justice Thomas suggests that our reference to statutory purpose here is inconsistent with our interpretation of “context” in 1 U. S. C. § 1. Post, at 213-214, n. 1. A focus on statutory text, however, does not preclude reasoning from statutory purpose. To the contrary, since “[s]tatutes . . . are not inert exercises in literary composition[, but] instruments of government,” United States v. Shirey, 359 U. S. 255, 260 (1959) (per Frankfurter, J.), a statute’s meaning is inextricably intertwined with its purpose, and we will look to statutory text to determine purpose because “the purpose of an enactment is embedded in its words even though it is not always pedantically expressed in words,” id., at 261.