delivered the opinion of the Court.
The costs that may be awarded to prevailing parties in lawsuits brought in federal court are set forth in 28 U. S. C. § 1920. The Court Interpreters Act amended that statute to include “compensation of interpreters.” §1920(6); see also § 7, 92 Stat. 2044. The question presented in this case is whether “compensation of interpreters” covers the cost of translating documents. Because the ordinary meaning of the word “interpreter” is a person who translates orally from one language to another, we hold that “compensation of interpreters” is limited to the cost of oral translation and does not include the cost of document translation.
I — <
This case arises from a personal injury action brought by petitioner Kouichi Taniguchi, a professional baseball player in Japan, against respondent Kan Pacific Saipan, Ltd., the owner of a resort in the Northern Mariana Islands. Petitioner was injured when his leg broke through, a wooden deck during a tour of respondent’s resort property. Initially, petitioner said that he needed no medical attention, *563but two weeks later, he informed respondent that he had suffered cuts, bruises, and torn ligaments from the accident. Due to these alleged injuries, he claimed damages for medical expenses and for lost income from contracts he was unable to honor. After discovery concluded, both parties moved for summary judgment. The United States District Court for the Northern Mariana Islands granted respondent’s motion on the ground that petitioner offered no evidence that respondent knew of the defective deck or otherwise failed to exercise reasonable care.
In preparing its defense, respondent paid to have various documents translated from Japanese to English. After the District Court granted summary judgment in respondent’s favor, respondent submitted a bill for those costs. Over petitioner’s objection, the District Court awarded the costs to respondent as “compensation of interpreters” under § 1920(6). Explaining that interpreter services “cannot be separated into ‘translation’ and ‘interpretation,’” App. to Pet. for Cert. 25a, the court held that costs for document translation “fal[l] within the meaning of ‘compensation of an interpreter,’” ibid. Finding that it was necessary for respondent to have the documents translated in order to depose petitioner, the court concluded that the translation services were properly taxed as costs.
The United States Court of Appeals for the Ninth Circuit affirmed both the District Court’s grant of summary judgment and its award of costs. The court rejected petitioner’s argument that the cost of document translation services is not recoverable as “compensation of interpreters.” The court explained that “the word ‘interpreter’ can reasonably encompass a ‘translator,’ both according to the dictionary definition and common usage of these terms, which does not always draw precise distinctions between foreign language interpretations involving live speech versus written documents.” 633 F. 3d 1218, 1221 (2011). “More importantly,” the court stressed, this construction of the statute “is more *564compatible with Rule 54 of the Federal Rules of Civil Procedure, which includes a decided preference for the award of costs to the prevailing party.” Ibid. The court thus concluded that “the prevailing party should be awarded costs for services required to interpret either live speech or written documents into a familiar language, so long as. interpretation of the items is necessary to the litigation.” Id., at 1221-1222.
Because there is a split among the Courts of Appeals on this issue,1 we granted certiorari. 564 U. S. 1066 (2011).
HH f — <
A
Although the taxation of costs was not allowed at common law, it was the practice of federal courts in the early years to award costs in the same manner as the courts of the relevant forum State. Alyeska Pipeline Service Co. v. Wilderness Society, 421 U. S. 240, 247-248 (1975). In 1793, Congress enacted a statute that authorized the awarding of certain costs to prevailing parties based on state law:
“That there be allowed and taxed in the supreme,.circuit and district courts of the United States, in favour of the parties obtaining judgments therein, such compensation for their travel and attendance, and for attornies and counsellors’ fees ... as are allowed in the supreme or superior courts of the respective states.” Act of Mar. 1, 1793, ch. 20, §4, 1 Stat. 333.
*565Although twice reenacted, this provision expired in 1799. Alyeska Pipeline, supra, at 248, n. 19; Crawford Fitting Co. v. J.T. Gibbons, Inc., 482 U. S. 437, 439 (1987). Yet even in the absence of express legislative authorization, the practice of referring to state rules for the taxation of costs persisted. See Alyeska Pipeline, 421 U. S., at 250.
Not until 1853 did Congress enact legislation specifying the costs allowable in federal court. Id., at 251. The impetus for a uniform federal rule was largely the consequence of two developments. First, a “great diversity in practice among the courts” had emerged. Ibid. Second, “losing litigants were being unfairly saddled with exorbitant fees for the victor’s attorney.” Ibid. Against this backdrop, Congress passed the 1853 Fee Act, which we have described as a “far-reaching Act specifying in detail the nature and amount of the taxable items of cost in the federal courts.” Id., at 251-252; The substance of this Act was transmitted through the Revised Statutes of 1874 and the Judicial Code of 1911 to the Revised Code of 1948, where it was codified, “without any apparent intent to change the controlling rules,” as 28 U. S. C. § 1920. 421 U. S., at 255.
Federal Rule of Civil Procedure 54(d) gives courts the discretion to award costs to prevailing parties. That Rule provides in relevant part: “Unless a federal statute, these rules, or a court order provides otherwise, costs — other than attorney’s fees — should be allowed to the prevailing party.” Rule 54(d)(1). We have held that “§ 1920 defines the term ‘costs’ as used in Rule 54(d).” Crawford Fitting, 482 U. S., at 441. In so doing, we rejected the view that “the discretion granted by Rule 54(d) is a separate source of power to tax as costs expenses not enumerated in § 1920.” Ibid.
As originally configured, § 1920 contained five categories of taxable costs: (1) “[f]ees of the clerk and marshal”; (2) “[f]ees of the court reporter for all or any part of the stenographic transcript necessarily obtained for use in the case”; (3) “[f]ees and disbursements for printing and witnesses”; *566(4) “[f]ees for exemplification and copies of papers necessarily obtained for use in the case”; and (5) “[djocket fees under section 1923 of this title.” 62 Stat. 955. In 1978, Congress enacted the Court Interpreters Act, which amended § 1920 to add a sixth category: “[ejompensation of court appointed experts, compensation of interpreters, and salaries, fees, expenses, and costs of special interpretation services- under section 1828 of this title.” 28 U. S. C. § 1920(6); see also § 7, 92 Stat. 2044. We are concerned here with this sixth category, specifically the item of taxable costs identified as “compensation of interpreters.”
B
To determine whether the item “compensation of interpreters” includes costs for document translation, we must look to the meaning of “interpreter.” That term is not defined in the Court Interpreters Act or in any other relevant statutory provision. When a term goes undefined in a statute, we give the term its ordinary meaning. Asgrow Seed Co. v. Winterboer, 513 U. S. 179, 187 (1995). The question here is: What is the ordinary meaning of “interpreter”?
Many dictionaries in use when Congress enacted the Court Interpreters Act in 1978 defined “interpreter” as one who translates spoken, as opposed to written, language. The American Heritage Dictionary, for instance, defined the term as “[o]ne who translates orally from one language 'into another.” American Heritage Dictionary 685 (1978). The Scribner-Bantam English Dictionary defined the related word “interpret” as “to translate orally.” Scribner-Bantam English Dictionary 476 (1977). Similarly, the Random House Dictionary defined the intransitive form of “interpret” as “to translate what is said in a foreign language.” Random House Dictionary of the English Language 744 (1973) (emphasis added). And, notably, the Oxford English Dictionary defined “interpreter” as “[o]ne who translates languages,” but then divided that definition into two senses: “a. [a] translator of books or writings,” which it designated *567as obsolete, and “b. [o]ne who translates the communications of persons speaking different languages; spec, one whose office it is to do so orally in the presence of the persons; a dragoman.” 5 Oxford English Dictionary 416 (1933); see also Concise Oxford Dictionary of Current English 566 (6th ed. 1976) (“[o]ne who interprets; one whose office it is to translate the words of persons speaking different languages, esp. orally in their presence”); Chambers Twentieth Century Dictionary 686 (1973) (“one who translates orally for the benefit of two or more parties speaking different languages:... a translator (obs.)”).
Pre-1978 legal dictionaries also generally defined the words “interpreter” and “interpret” in terms of oral, translation. The; then-current edition of Black’s Law Dictionary, for example, defined “interpreter” as “[a] person sworn at a trial to interpret the evidence of a foreigner... to the court,” and it defined “interpret” in relevant part as “to translate orally from one tongue to another.” Black’s Law Dictionary 954, 953 (rev. 4th ed. 1968); see also W. Anderson, A Dictionary of Law 565 (1888) (“[o]ne who translates the testimony of witnesses speaking a foreign tongue, for the benefit of the court and jury”); 1 B. Abbott, Dictionary of Terms and Phrases Used in American or English Jurisprudence 639 (1878) (“one who restates the testimony of a witness testifying in a foreign tongue, to the court and jury, in their language”). But see Ballentine’s Law Dictionary 655, 654 (3d ed. 1969) (defining “interpreter” as “[o]ne who interprets, particularly one who interprets words written or spoken in a foreign language,” and “interpret” as “to translate from a foreign language”).
Against these authorities, respondent relies almost exclusively on Webster’s Third New International Dictionary (hereinafter Webster’s Third). The version of that dictionary in print when Congress enacted the Court Interpreters Act defined “interpreter” as “one that translates; esp: a person who translates orally for parties conversing in different *568tongues.” Webster’s Third 1182 (1976).2 The sense divider esp (for especially) indicates that the most common meaning of the term is one “who translates orally,” but that meaning is subsumed within the more general definition “one that translates.” See 12,000 Words: A Supplement to Webster’s Third 15a (1986) (explaining that esp “is used to introduce the most common meaning included in the more general preceding definition”). For respondent, the general definition suffices to establish that the term “interpreter” ordinarily includes persons who translate the written word. Explaining that “the word ‘interpreter’ can reasonably encompass a ‘translator,’ ” the Court of Appeals reached the same conclusion. 633 F. 3d, at 1221. We disagree.
That a definition is broad enough to encompass one sense of a word does not establish that the word is ordinarily understood in that sense. See Mallard v. United States Dist. Court for Southern Dist. of Iowa, 490 U. S. 296, 301 (1989) (relying on the “most common meaning” and the “ordinary and natural signification” of the word “request,” even though it may sometimes “double for ‘demand’ or ‘command’ ”). The fact that the definition of “interpreter” in Webster’s Third has a sense divider denoting the most common usage suggests that other usages, although acceptable, might not be common-or ordinary. It is telling that all the dictionaries cited above defined “interpreter” at the time of the statute’s enactment as including persons who translate orally, but only a handful defined the word broadly enough to encompass *569translators of written material. See supra, at 566-568. Although the Oxford English Dictionary, one of the most authoritative on the English language, recognized that “interpreter” can mean one who translates writings, it expressly designated that meaning as obsolete. See supra, at 566-567. Were the meaning of “interpreter” that respondent advocates truly common or ordinary, we would expect to see more support for that meaning. We certainly would not expect to see it designated as obsolete in the Oxford English Dictionary. Any definition of a word that is absent from many dictionaries and is deemed obsolete in others is hardly a common or ordinary meaning.
Based on our survey of the relevant dictionaries, we conclude that the ordinary or common meaning of “interpreter” does not include those who translate writings. Instead, we find that an interpreter is normally understood as one who translates orally from one language to another. This sense of the word is far more natural. As the Seventh Circuit put it: “Robert Fagles made famous translations into English of the Iliad, the Odyssey, and the Aeneid, but no one would refer to him as an English-language ‘interpreter’ of these works.” Extra Equipamentos E Exportaȩão Ltda. v. Case Corp., 541 F. 3d 719, 727 (2008).
To be sure, the word “interpreter” can encompass persons who translate documents, but because that is not the ordinary meaning of the word, it does not control unless the context in which the word appears indicates that it does. Nothing in the Court Interpreters Act or in § 1920, however, even hints that Congress intended to go beyond the ordinary meaning of “interpreter” and to embrace the broadest possible meaning that the definition of the word can bear.
If anything, the statutory context suggests the opposite: that the word “interpreter” applies only to those who translate orally. As previously mentioned, Congress enacted § 1920(6) as part of the Court Interpreters Act. The main *570provision of that Act is §2(a), codified in 28 U. S. C. §§ 1827 and 1828. See 92 Stat. 2040-2042. Particularly relevant here is § 1827. As it now reads, that statute provides for the establishment of “a program to facilitate the use of certified and otherwise qualified interpreters in judicial proceedings instituted by the United States.” § 1827(a). Subsection (d) directs courts to use an interpreter in any criminal or civil action instituted by the United States if a party or witness “speaks only or primarily a language other than the English language” or “suffers from a hearing impairment” “so as to inhibit such party's comprehension of the proceedings or communication with counsel or the presiding judicial officer, or so as to inhibit such witness' comprehension of questions and the presentation of such testimony.” § 1827(d)(1).3 As originally enacted, subsection (k) mandated that the “interpretation provided by certified interpreters . . . shall be in the consecutive mode except that the presiding judicial officer . . . may authorize a simultaneous or summary interpretation.” § 1827(k) (1976 ed., Supp. II); see also 92 Stat. 2042. In its current form, subsection (k) provides that interpretation “shall be in the simultaneous mode for any party ., . and in the consecutive mode for witnesses,” unless the court directs otherwise. The simultaneous, consecutive, and summary modes are all methods of oral interpretation and have nothing to do with the translation of writings.4 Taken together, these provisions are a strong *571contextual clue that Congress was dealing only with oral translation in the Court Interpreters Act and that it intended to use the term “interpreter” throughout the Act in its ordinary sense as someone who translates the spoken word. As we have said before, it is a “ ‘normal rule of statutory construction’ that ‘identical words used in different parts of the same act are intended to have the same meaning.’ ” Gustafson v. Alloyd Co., 513 U. S. 561, 570 (1995) (quoting Department of Revenue of Ore. v. ACF Industries, Inc., 510 U. S. 332, 342 (1994)).5
The references to technical terminology in the Court Interpreters Act further suggest that Congress used “interpreter” in a technical sense, and it is therefore significant that relevant professional literature draws a line between “interpreters,” who “are used for oral conversations,” and “translators,” who “are, used for written communications.” Zazueta, supra n. 4, at 477; see also M. Frankenthaler, Skills for Bilingual Legal Personnel 67 (1982) (“While the translator deals with the written word, the interpreter is concerned with the spoken language”); Brislin, Introduction, in Translation: Applications and Research i (R. Brislin ed. 1976) (explaining that when both terms are used together, translation “refers to the processing [of] written input, and interpretation to the processing of oral input” (emphasis deleted)); J. Herbert, Interpreter’s Handbook 1 (2d ed. 1952) (“In the present-day jargon of international organisations, the words translate, translations, translator are used when the immediate result of the work is a written text; and the words inter*572pret, interpreter, interpretation when it is a speech delivered orally”). That Congress specified “interpreters” but not “translators” is yet another signal that it intended to limit § 1920(6) to the costs of oral, instead of written, translation.6
In sum, both the ordinary and technical meanings of “interpreter,” as well as the statutory context in which the word is found, lead to the conclusion that § 1920(6) does not apply to translators of written materials.7
C
No other rule of construction compels us to depart from the ordinary meaning of “interpreter.” The Court of Appeals reasoned that a broader meaning is “more compatible with Rule 54 of the Federal Rules of Civil Procedure, which includes a decided preference for the award of costs to the prevailing party.” 633 F. 3d, at 1221. But we have never held that Rule 54(d) creates a presumption of statutory construction in favor of the broadest possible reading of the costs enumerated in § 1920. To the contrary, we have made clear that the “discretion granted by Rule 54(d) is not a power to evade” the specific categories of costs set forth by Congress. Crawford, Fitting, 482 U. S., at 442. “Rather,” *573we have said, “it is solely a power to decline to tax, as costs, the items enumerated in §1920.” Ibid. Rule 54(d) thus provides no sound basis for casting aside the ordinary meaning of the various items enumerated in the costs statute, including the ordinary meaning of “interpreter.”
Our decision is in keeping with the narrow scope of taxable costs. “Although ‘costs’ has an everyday meaning synonymous with ‘expenses,’ the concept of taxable costs under Rule 54(d) is more limited and represents those expenses, including, for example, court fees, that a court will assess against a litigant.” 10 C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure § 2666, pp. 202-203 (3d ed. 1998) (hereinafter Wright & Miller). Taxable costs are limited to relatively minor, incidental expenses as is eyident from § 1920, which lists such items as clerk fees, court reporter fees, expenses for printing and witnesses, expenses for exemplification and copies, docket fees, and compensation of court-appointed experts. Indeed, “the assessment of costs most often is merely a clerical matter that can be done by the court clerk.” Hairline Creations, Inc. v. Refalas, 664 F. 2d 652, 656 (CA7 1981). Taxable costs are a fraction of the nontaxable expenses borne by litigants for attorneys, experts, consultants, and investigators. It comes as little surprise, therefore, that “costs almost always amount to less than the successful litigant’s total expenses in connection with a lawsuit.” 10 Wright & Miller §2666, at 203. Because taxable costs are limited by statute and are modest in scope, we see no compelling reason to stretch the ordinary meaning of the cost items Congress authorized in § 1920.
As for respondent’s extratextual arguments, they are more properly directed at Congress. Respondent contends that documentary evidence is no less important than testimonial evidence and that it would be anomalous to.require the losing party to cover translation costs for spoken words but not for written words. Brief for Respondent 20. Respondent also observes that some translation tasks are not entirely oral or entirely written. Id., at 20-24. One task, called “ ‘sight *574translation,”’ involves the oral translation of a document. Id., at 21. Another task involves the written translation of speech. Ibid. And a third task, called “ ‘document comparison,’ ” involves comparing documents in the source and target language to verify that the two are identical. Id., at 21-22. Respondent argues that a narrow definition cannot account for these variations and that a bright-line definition of “interpreter” as someone who translates spoken and written words would avoid complication and provide a simple, administrable rule for district courts.
Neither of these arguments convinces us that Congress must have intended to dispense with the ordinary meaning of “interpreter” in §1920(6). First, Congress might have distinguished between oral and written translation out of a concern that requiring losing parties to bear the potentially sizable costs of translating discovery documents, as opposed to the more limited costs of oral testimony, could be too burdensome and possibly unfair, especially for litigants with limited means. Cf. Fleischmann Distilling Corp. v. Maier Brewing Co., 386 U. S. 714, 718 (1967) (noting the argument “that since litigation is at best uncertain one should not be penalized for merely defending or prosecuting a lawsuit, and that the poor might be unjustly discouraged from instituting actions to vindicate their rights if the penalty for losing included the fees of their opponents’ counsel”). Congress might also have concluded that a document translator is more akin to an expert or consultant retained by a party to decipher documentary evidence — like, for instance, a forensic accountant — than to an interpreter whose real-time oral translation services are necessary for communication between litigants, witnesses, and the court.8
*575Second, respondent has not shown that any of the hybrid translation/interpretation tasks to which it points actually arise with overwhelming frequency or that the problem of drawing the line between taxable and nontaxable costs in such cases will vex the trial courts. It certainly has not shown that any such problems will be more troublesome than the task of sifting through translated discovery documents to ascertain which can be taxed as necessary to the litigation. In any event, the present case does not present a hybrid situation; it involves purely written translation, which falls outside the tasks performed by an “interpreter” as that term is ordinarily understood.
* * *
Because the ordinary meaning of “interpreter” is someone who translates orally from one language to another, we hold that the category “compensation of interpreters” in § 1920(6) does not include costs for document translation. We therefore vacate the judgment of the United States Court of Ap- . peals for the Ninth Circuit and remand the case for further proceedings consistent with this opinion.
It is so ordered.
Compare BDT Products, Inc. v. Lexmark Int'l, Inc., 405 F. 3d 415, 419 (CA6 2005) (holding that document translation costs are taxable under § 1920(6) because the “definition of interpret expressly includes to ‘translate into intelligible or familiar language’ ” (quoting Webster’s Third New International Dictionary 1182 (1981))), with Extra Equipamentos E Exportando Ltda. v. Case Corp., 541 F. 3d 719, 727-728 (CA7 2008) (holding that document translation costs are not taxable under § 1920(6) because an interpreter is “normally understood [as] a person who translates living speech from one language to another”).
A handful of other contemporaneous dictionaries used a similar formulation. See Funk & Wagnalls New Comprehensive International Dictionary of the English Language 665 (1977) (“[o]ne who interprets or translates; specifically, one who serves as oral translator between people speaking different languages”); 1 World Book Dictionary 1103 (C. Barn-hart & R. Barnhart eds. 1977) (“a person whose business is translating, especially orally, from a foreign language”); Cassell’s English Dictionary 617 (4th ed. 1969) (“[o]ne who interprets, esp. one employed to translate orally to persons speaking a foreign language”).
This provision remains substantially the same as it appeared when first enacted. See 28 U. S. C. § 1827(d)(1) (1976 ed., Supp. II); see also 92 Stat. 2040.
The simultaneous mode requires the interpreter “to interpret and to speak contemporaneously with the individual whose communication is being translated.” H. R. Rep. No. 95-1687, p. 8 (1978). The consecutive mode requires the speaker whose communication is being translated to pause so that the interpreter can “convey the testimony given.” Ibid. And the summary mode “allow[s] the interpreter to condense and distill the speech of the speaker.” Ibid.; see generally Zazueta, Attorneys Guide to the Use of Court Interpreters, 8 U. C. D. L. Rev. 471, 477-478 (1975).
The dissent agrees that context should help guide our analysis, but instead of looking to the Court Interpreters Act, it looks to “the practice of federal courts both before and after § 1920(6)’s enactment.” Post, at 579 (opinion of Ginsburg, J.). The practice of federal courts after the Act’s enactment tells us nothing about what Congress intended at the time of enactment. And federal-court practice before the Act under other provisions of § 1920 tells us little, if anything, about what Congress intended when it added paragraph (6). We think the statutory context in which the word “interpreter” appears is a more reliable guide to its meaning.
Some provisions within the United States Code use both “interpreter” and “translator” together, thus implying that Congress understands the terms to have the distinct meanings described above. See, e. g., 8 U. S. C. § 1555(b) (providing that appropriations for the Immigration and Naturalization Service “shall be available for payment of . . . interpreters and translators who are not citizens of the United States”); 28 U. S. C. § 530C(b)(1)(1) (providing that Department of Justice funds may be used for “[pjayment of interpreters and translators who are not citizens of the United States”).
Our conclusion is buttressed by respondent’s concession at oral argument that there is no provision in the United States Code where it is clear that the word extends to those who translate documents. Tr. of Oral Arg. 39; see also Brief for Petitioner 32 (“And the Code is wholly devoid of any corresponding definition of ‘interpreter’ extending to the translation of written documents”). As respondent acknowledged, either the word is used in a context that strongly suggests it applies only to oral translation or its meaning is unclear. See Tr. of Oral Arg. 38.
The dissent contends that document translation, no less than oral translation, is essential “to ecpiip the parties to present their case clearly and the court to decide the merits intelligently. ” Post, at 579. But a document translator is no more important than an expert or consultant in making sense of otherwise incomprehensible documentary evidence, yet expenses *575for experts and consultants are generally not taxable as costs. To be sure, forgoing document translation can impair a litigant’s case, but document translation is not indispensable, in the way oral translation is, to the parties’ ability to communicate with each other, with witnesses, and with the court.