04-480 IN THE

Supreme Court of the United States
———— METRO-GOLDWYN-MAYER STUDIOS INC., et al., Petitioners, v. GROKSTER, LTD., and STREAMCAST NETWORKS, INC., Respondents. ———— On Writ of Certiorari to the United States Court of Appeals for the Ninth Circuit ———— BRIEF FOR RESPONDENTS ————
MICHAEL H. PAGE MARK A. LEMLEY KEKER & VAN NEST, LLP 710 Sansome Street San Francisco, CA 94111 (415) 391-5400 Counsel for Grokster CHARLES S. BAKER PORTER & HEDGES, LLP 700 Louisiana Street, Ste. 3500 Houston, TX 77022 (713) 226-0676 Counsel for StreamCast MATTHEW A. NECO WENDY MILLAR GOODKIN STREAMCAST NETWORKS, INC. 20969 Ventura Blvd. Woodland Hills, CA 91364 (818) 887-8610 Counsel for StreamCast
WILSON-EPES PRINTING CO., INC. – (202) 789-0096 – W ASHINGTON, D. C. 20001

RICHARD G. TARANTO H. BARTOW FARR, III FARR & TARANTO 1220 19th Street, NW Washington, DC 20036 (202) 775-0184 Counsel for Respondents CINDY A. COHN Counsel of Record FRED VON LOHMANN ELECTRONIC FRONTIER FNDTN. 454 Shotwell Street San Francisco, CA 94110 (415) 436-9333 Counsel for StreamCast

QUESTION PRESENTED Whether the court of appeals correctly ruled, on the only issue before it, that respondn ’ ir u o o t current et d tbt n f h s si i e versions of their file-sharing software does not render respondents secondarily liable for every direct infringement of pti e ’oyi tcm ie b ue o t sf a . et nr cpr h o m td y sr fh ot r io s gs t s e w e


ii PARTIES TO THE CASE AND RULE 29.6 STATEMENT The parties to the proceeding are fully listed in the MGM pti e ’ r fai. h pr topr i o r pnet et nr bi ( i T e a n croao fe odn io s e t ) e tn s StreamCast Networks, Inc., is Stirling Bridge, Inc., which is not a publicly traded company. Respondent Grokster Ltd. has no parent company, and no publicly traded company owns 10% or more of its stock. The named defendants in the underlying district court cases—but not parties to the particular district court ruling at issue or in the court of appeals cases on certiorari review here—include Kazaa B.V., Niklas Zennstrom, LaGaliote BV, Janus Friis Degnbol, Indigo Investment BV, Sharman Networks Ltd., and LEF Interactive Pty Ld.

TABLE OF CONTENTS Page QUESTION PRESENTED............................................ PARTIES TO THE CASE AND RULE 29.6 STATEMENT ........................................................... TABLE OF AUTHORITIES......................................... STATEMENT ............................................................... A. Factual Background ............................................. 1. R sodn ’ ota ..................................... epnet S f r s w e 2. Benefits of Decentralized Peer-to-Peer File Sharing ............................................................. B. District Court Proceedings................................... C. The Court of Appeals Ruling............................... SUMMARY OF ARGUMENT..................................... ARGUMENT................................................................. I. R sodn ’Dsi t nO T e S f a I epnet ir u o f hi ot r s s tb i r w e Lawful Under The Express Rule Of Sony v. Universal ................................................................. A. Sony Protects General Distribution Of A Product If It Is Capable Of Significant Noninfringing Uses...................................................... Governing Principles and Existing Procedents . The Product-Distribution Rule .......................... B. The Sony Rule Protec R sodn ’S f t epnet ot s s ware ................................................................... C. The Sony Rule Safeguards Innovation In Products With Legitimate Uses ......................... D. Pti e ’ rpsl Wol At T eSony et nr Pooa io s s u lr h d e Rule And Impair Its Rationale........................... Predominant Use................................................ (iii) i ii v 1 2 2 6 8 12 13 15


15 16 18 21 24 26 27

iv TABLE OF CONTENTS Page Profit Motive...................................................... Redesign ............................................................ II. Other Possible Bases For Secondary Liability Are Not Applicable As The Case Comes To This Court A. Claims Of Contributory Infringement Liability For Urging Infringement Or Assisting Specific Known Acts Of Infringement Are Not Before This Court .......................................................... B. R sodn ’L c O C n o O S epnet ak f ot l f pecific s r Acts Of Infringement Precludes Vicarious Liability ............................................................. III. Any Expansion Of Copyright Liability In The Peerto-Peer Context Should Come From Congress......... A. Stare Decisis Weighs Heavily Against Loosening The Innovation-Protecting Sony Rule.................................................................... B. The Overall Effects Of Peer-To-Peer File Sharing Are Still Uncertain ............................... C. For The Technology-Specific Issues Raised, Only Congress Can Choose Among The Many Possible Technology-Specific Solutions ........... D. Congress Has Regularly Addressed Copyright O nr Technology-Specific Concerns........... w e’ s CONCLUSION ............................................................. 29 30 33


37 40

40 42

46 48 50

v TABLE OF AUTHORITIES CASES In re Aimster Copyright Litig., 334 F.3d 643 (7th Cir. 2003), cert. denied, 540 U.S. 1107 (2004) .. Alloc, Inc. v. ITC, 342 F.3d 1361 (Fed. Cir. 2003), cert. denied, 124 S. Ct. 2390 (2004) ...... A & M Records v. Napster, 239 F.3d 1004 (9th Cir. 2001)........................................................... A & M Records v. Napster, 284 F.3d 1091 (9th Cir. 2002)........................................................... AT&T Corp. v. Iowa Utilities Bd., 525 U.S. 366 (1999)................................................................. Black & Decker Disability Plan v. Nord, 538 U.S. 822 (2003) ................................................. Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141 (1989) .......................................... Campbell v. Acuff-Rose Music Inc., 510 U.S. 569 (1994)................................................................. Carson v. American Brands, Inc., 450 U.S. 79 (1981)................................................................. Central Bank of Denver, N.A. v. First Interstate Bank of Denver, N.A., 511 U.S. 164 (1994) ...... Clark v. Martinez, 125 S. Ct. 716 (2005) .............. C.R. Bard, Inc. v. Advanced Cardiovascular Systems, 911 F.2d 670 (Fed. Cir. 1990).................. Dawson Chemical Co. v. Rohm & Haas Co., 448 U.S. 176 (1980) ................................................. Eldred v. Ashcroft, 537 U.S. 186 (2003) ............... Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002).................. Friends of the Earth, Inc. v. Laidlaw Environmental Servs., Inc., 528 U.S. 167 (2000)........... Fromberg, Inc. v. Thornhill, 315 F.2d 407 (5th Cir. 1963)........................................................... Page 4 28 4 4 26 41 25 29, 30 34 26, 39 24, 40 28 18, 28 47 25 37 36

vi TABLE OF AUTHORITIES—Continued Page G r wn P b g C r.v C l b Att es i ul h ’ op . o m i rs u a is Mgmt., Inc., 443 F.2d 1159 (2d Cir. 1971)........ 26 Golden State Transit Corp. v. Los Angeles, 493 U.S. 103 (1989) ................................................. 40 Henry v. A.B. Dick Co., 224 U.S. 1 (1912)............ 36 Kalem Co. v. Harper Bros., 222 U.S. 55 (1911) ... 17 Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996).......................................................... 25 Meyer v. Holley, 537 U.S. 280 (2003)................... 14, 39 Milwaukee v. Illinois, 451 U.S. 304 (1981)........... 41 Neal v. United States, 516 U.S. 284 (1996)........... 40 Patterson v. McLean Credit Union, 491 U.S. 164 (1989)................................................................. 40 Rohm & Haas Co. v. Dawson Chemical Co., 599 F.2d 685 (5th Cir. 1979), a ’, 448 U.S. 176 fd f (1980)................................................................. 37 Shapiro, Bernstein & Co. v. H.L. Green Co., 316 F.2d 304 (2d Cir. 1963) ..................................... 38 Sony Corp. of America v. Universal City Studios, 464 U.S. 417 (1984) .......................................... passim Teleprompter Corp. v. CBS, 415 U.S. 394 (1974)................................................................. 47 Universal City Studios Inc. v. Sony Corp. of America, 480 F. Supp. 429 (C.D. Cal. 1979), rv , 659 F.2d 963 (9th Cir. 1981), rv , 464 e’ d e’ d U.S. 417 (1984) .............................................16, 18, 29 Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, 540 U.S. 398 (2004) .............. 24, 41 Warner-Lambert Co. v. Apotex Corp., 316 F.3d 1348 (Fed. Cir. 2003) ........................................ 36 Yamaha Motor Corp., U.S.A. v. Calhoun, 516 U.S. 199 (1996) ................................................. 34

vii TABLE OF AUTHORITIES—Continued STATUTES, RULES, AND Page CONGRESSIONAL MATERIALS Audio Home Recording Act of 1992, 17 U.S.C. §§ 1001-1010..................................................... 49, 50 Copyright Act of 1909, § 1, 35 Stat. 1075............. 17 Copyright Act of 1976, as amended, 17 U.S.C. § 106 .................................................................. 13 § 108 .................................................................. 48 § 109 .................................................................. 48 § 110 .................................................................. 48 § 111 .................................................................. 49 § 114 .................................................................. 49 § 115 .................................................................. 49 § 116 .................................................................. 49 § 117 .................................................................. 49 § 118 .................................................................. 49 § 119 .................................................................. 49 § 122 .................................................................. 49 § 504 .................................................................. 32 Digital Millenium Copyright Act of 1998 (DMCA), 17 U.S.C. §§ 512, 1201-1205............. passim 28 U.S.C. § 1292(a) ............................................... 34 28 U.S.C. § 1292(b)............................................... 34 35 U.S.C. § 271(b)................................................. 36 35 U.S.C. § 271(c) .............................................18, 28, 36 Fed. R. Civ. P. 54(b).............................................. 11 H.R. Rep. 1476, 94th Cong., 2d Sess. (1976)........ passim OTHER MATERIALS A e FdnT l & R d A tt T sm n m r e’ e . . ai rs , et oy o is i (2001), .... W. Baumol, The Free-Market Innovation Machine: Analyzing the Growth Miracle of Capitalism (2002) ..............................................



viii TABLE OF AUTHORITIES—Continued Page Brief of Universal Studios in Sony Corp. of America v. Universal Studios (81-1687 Universal Br.) ......................................................16, 18, 19 British Broadcast Corp., entertainment/4150747.stm ............................... 43 Brodley, The Economic Goals of Antitrust: Efficiency, Consumer Welfare, and Technological Progress, 62 N.Y.U. L. Rev. 1020 (1987)................................................................. 25 Bureau of National Affairs, 73 U.S.L.W. 2438 (Feb. 1, 2005)..................................................... 45 D. Chisum, Patents (2004) § 17.03[3]....................................................... 23 § 17.04[3]....................................................... 37 § 17.04[4]....................................................... 36 CNET, RIAA+files+754+new+ file-swapping+suits/21101027_35494259. html.................................................................... 45 Fabrikant, Cable and Film Divisions Bolster News Corp. Profit, N.Y. Times, Feb. 3, 2005 ... 43 W. Fisher, Promises to Keep: Technology, Law, and the Future of Entertainment (2004)............ passim Geist, N m es D nt C uc A a s u br o’ rnh gi t n Downloading, Toronto Star, Nov. 29, 2004 ...... 44 Address by Patrick Gelsinger, Intel, Aug. 24, 2000, speeches/ pg082400.htm.................................... 8, 22 P. Goldstein, Copyright (2d ed.)............................ 33 Harvard Berkman Center, Content and Control (Jan. 7, 2005) ..................................................... passim P. Huber, O w ls eeg (1994)....................... r e’R vne l 8

ix TABLE OF AUTHORITIES—Continued Page Internet Week, Feb. 9, 2005, www.internetweek. com/showArticle.jhtml?articleID=59302454 .... 46 Johnson, Good News in Hollywood. Shhh., N.Y. Times, Jan. 31, 2005.......................................... 43 Kusek, Music Like Water, Forbes, Jan. 31, 2005, 2005/0131/042.html........................................... 47 J. Lardner, Fast Forward: Hollywood, the Japanese, and the VCR Wars (Mentor ed. 1988).................................................................. 42 Lemley & Reese, Reducing Digital Copyright Infringement Without Restricting Innovation, 56 Stan. L. Rev. 1345 (2004).........................32, 46, 47 J. Litman, Digital Copyright (2001)...................... 46 Courtney Love, Courtney Love Does the Math, Salon, 2000/06/14/love ................................................. 22 J. Menn, All the Rave: The Rise and Fall of Sa n an gs as r hw F ni ’N pt (2003)...................... n e 8 Music Industry Way Off Track With Song and Dance About Falling Sales, Sydney Morning Herald, Mar. 29, 2004, au/articles/2004/03/28/1080412234274.html .... 44 Mui s Bi t F t e The Economist, s ’ r h r u r, c ge u Oct. 28, 2004, displayStory. cfm?story_id=3329169................ 44 Oberholzer & Strumpf, The Effect of Filesharing on Record Sales: An Empirical Analysis, cigar/papers/FileSharingMarch2004.pdf ...................................... 44

x TABLE OF AUTHORITIES—Continued Page R cri I ut A s o A ec, eod g n sy s n f m r a n d r ’ i Stats.pdf ......................................................... ..... Petition for Certiorari, No. 03-1579 .................. F. Scherer, Innovation and Growth: Schumpeterian Perspectives (1984) .................. F. Scherer & D. Ross, Industrial Market Structure and Economic Performance (3d ed. 1990).................................................................. J. Schumpeter, Capitalism, Socialism, and Democracy (HarperPerennial ed. 1976) ............ United States, Amicus Br. 3, in Napster (9th Cir. Sept. 8, 2000), napsteramicus.pdf .............................................. U.S. Entertainment Industry: 2003 MPA Market Statistics............................................................., article.asp?articleID=9535&catType=NEWS ... Vivendi-Universal, vu/en/files/PR050201_Revenues_for_2004_Q 4_2004.pdf......................................................... Washington Post, Jan. 20, 2005............................. Wu, C pr h sC m ui t n P ly 103 oyi t o m n a os o c, g’ ci i Mich. L. Rev. 278 (2004) .................................. 10A Wright, Miller & Kane, Federal Practice and Procedure: Civil 3d § 2728 (1998)............. 16 Wright, Miller, & Cooper, Federal Practice and Procedure: Jurisdiction 2d § 3924.1 (1996).................................................................

43 45 45 25

26 25

49 44 43

44 44 28 9


STATEMENT Petitioners urge three grounds for secondary liability. Only two are properly before this Court, and they are meritless. First: ne t hai o “otbt yi r gm n” U dr h ed g f cn i o n i e et e n r u r fn , petitioners urge that respondents should be liable for providing their software to users, with general knowledge that many will use it for infringing purposes. That contention is foreclosed by Sony Corp. of America v. Universal City Studios, 464 U.S. 417 (1984), which held that no secondary liability may attach to distribution of an article of commerce, even wt “osut e ko l g o i r g g s,sog st i cnt cv” nwe e fn i i uea l a i h r i d fn n n is capable of significant noninfringing uses. The software here meets that standard. In this situation, copyright holders must pursue not the distributor of the product, but those who misuse it. Petitioners are readily able to discover misuses and to attack them in direct infringement suits. Second: l i oi “otbt yi r gm n”pt As n k g cn i o n i e et eo v n r u r fn , i tioners urge that respondents are liable for various alleged past activities—over and above the usual incidents of ongoing distribution of their software—to encourage or to aid known infringement. But this claim is not before this Court. The district court limited its summary judgment ruling to respondn ’ et then-current file-sharing software, and it certified only s that ruling for interlocutory appeal. The rest of the case remains in the district court, including claims based on past versions of software or past activities that allegedly went beyond the normal offering and promotion of the software. Third: et nr asrt tepnet a “i r ul Pti e s th r odn r v a os io s e a s s e ci y lb ”frd et nr gm n cm ie b t ue o i l o i c i i e et o m td y h sr f ae r fn s t e s their software. But vicarious liability cannot apply here, not only because of Sony’ r sn s e oing, but because any such a liability would require t tepnet present product give h r odn ’ a s s them control over individual infringing acts, which it does not. Pti e ’c i , ht epnet sol hv det nr lm t r odn hu ae e io s a a s s d

2 signed their products differently to gain control, demands an inquiry that copyright law does not and should not require. For petititioners to prevail, therefore, this Court would have to change the clear Sony rule. It should not do so, and not only because of stare decisis. The suggested changes in the general law of secondary liability would profoundly impair Sony’ esn a po co o poutnoao,ot s s tl rt t n f rdc i vt n t h ei ei n i e detriment of future creators among others. Only Congress is institutionally suited to consider the challenges presented by decentralized peer-to-peer file searching and sharing, to assess its real-world effects, to decide when legislative intervention in market experimentation is advisable, and to choose among possible context-specific legal regimes, as it has done throughout the Copyright Act. The ruling below should therefore be affirmed. A. Factual Background 1 R so dnsS f ae The software that respondents . ep n et ot r. ’ w distribute (see JA78-91, JA92-124, Pet. App. 41a-44a; see also JA440-65) is not necessary for computer users to share files. Ordinary email, instant messaging, web browsers, and other software allow users to send and receive files, including music adv e fe.Whtota l e epnet ad t t t n i o is d l a sf r i r odn ’ dso h w ek s s a capability is, at bottom, a mechanism for efficiently finding other computer users who have files a user is seeking. The ue cm ad t sf a t sn ot d yuhv afe sr o m ns h ot r o ed u “o o ae i e w e l f t i X ” esgst w i o e cm u r cn e u n ? m s e,o h h t r o pt s onected a rg a c h e to the Internet can automatically respond. The software is wholly neutral as to the type of file sought, whether it contains words, data, photographs, music, video, or other content.1 Software to search for information on line, like software to enable sharing of files, is itself hardly new. Yahoo, Google,
T e“ ”i t sa hqe m gt et fenm o aw r t t h X n h er ur i b h i a e r od h e c y h e l a apa i “ dx g pr o t fe(fncld“ e dt ) For per n i ei ” as f h i ot ae m t a ” s n n t e l e l a a. example, songs encoded in the prevalent MP3 format commonly include identifying information such as title, artist name, and album name.

3 and others enable searci .T oe“er eg e, hw h g hs sa h ni s o n c n ” ever, focus on the always-o “e e ”o t Wol Wi n sr r n h vs e r d d e Web: a search request produces a list of responsive Web pages, which users, by mouse click, can transfer to their computers. The software at issue here extends the reach of searches beyond centralized Web servers to the computers of ordinary users who are on line: searchers send a request to an already-online computer, which propagates the request to computers to which it is connected, and so on for several steps until (through this exponential process) a large responsive base is reached; in only seconds, the search generates a lt fr pni ol eue ’fe—each user can fully i o e os e n n sr is s s v i s l control what files are made accessible to others (JA253; contra MGM Br. 4, 25, 46)—and files can then be downloaded by mouse click. Pet. App. 42a, 44a. The software is cld“erope (2 ) eas ieal od a ue ae pe t er P P bcue t nb s ri r sr l ” e ny s to communicate with each other directly. P2P software is compact and has proven easy to write. Numerous versions are available—which together overshadow r pnet vrosn sg—some by commercial e odn ’ e i i uae s s sn ventures (e.g., LimeWire, BearShare, eDonkey, Kazaa, Ares), others by hobbyists (e.g., Bit Torrent, Freenet, Gnucleus). Some leave the searching to other tools and refine the downloading function. Others refine what and how information is presented to users or t m as fi i o e ue ’ h en o f d g t r sr e nn h s files. For example, P2P software can be designed to search i i s f ei a dfe o i i da ue ’ omputers; or n c o ds nt is n n v ul sr c de g e l di s it can be designed to organize subgroups of computers and create a common (ever-changing) index for that subgroup on a a i l ue s o pt —selected automatically, without prc a sr cm u r tu r ’ e t sf a c a r i e et nbsd n at si m m h ot r r t ’ n r n o,ae o f o l e e e w e e o s tv i cr k ory capacity or time on line—so that only the common index need be searched, not every computer in the subgroup. Pet. App. 6a-7a; JA254-55. T et m “e ok ad“e i ”i t cn x o t s h e s nt r” n sr c n h ot t fh r w ve e e i case, although common in informal or specialized references

4 (including by respondents), can easily conjure a misleading p t e A t “e ok:epnetd nt w o cn o iu . so nt r” r odn o o o n r ot l cr w s s r a“e ok i t f ia sneo, a, tehn cm nt r” n h a lr es fsy a e poe o w e mi l pany that owns and operates a set of physical lines and associated equipment. Al ntok m asi t peet l“e r” en n h r n w e s setting is the group of users of a particular computer language (protocol) who happen to be on line at any given moment. T e ntok cags rm i t toi t ta ue s n h “e r” hne f w o n a t n a , s sr i sn sn s g on or drop off: it is more a self-forming community of comm nl gaesekr wt nha n d t c t na“e o a ug pae i i er g ia e h n s h i sn a nt w r”n h f ia l sne T eagaes pkn vr ok i t a lra es. h l ug isoe oe e mi y n physical equipment that is not owned or run by respondents, btnl e t ue s o pt ( an HP or Dell or IBM u i u sh sr cm u re.g., cd e ’ e m ci rni Mi oot Wi o s na It o A ah e un g c sfs n w o n n l r MD n n r ’ d e processor), the copper wire, coaxial cable, optical fiber, or wireless band of a telephone or cable or wireless company, and the equipment of Internet service providers like AOL. T ee “e i ” a a o i a ied g iue R h t m sr c cn l g e m s ai p t . e r ve s v l n cr spondents supply users a piece of software, as Microsoft provides software to users or Sony provided users a video tape recorder. The product then functions for users to find and transfer files without any need for continuing provision by respondents of further goods or services to the users. In this key respect, respondents differ from the original Napster, which, after supplying software, operated a server that continually responded to each individual file-transfer request with the very information that was essential to that transfer: the location of available files, even ones Napster allegedly knew were not authorized for copying.2 As the district court
See Pet. App. 5a; A & M Records v. Napster, 239 F.3d 1004 (9th Cir. 2001) (Napster I); A & M Records v. Napster, 284 F.3d 1091 (9th Cir. 2002) (Napster II). The Aimster service at issue in In re Aimster Copyright Litig., 334 F.3d 643 (7th Cir. 2003), cert. denied, 540 U.S. 1107 (04,i t N pt sri ( d “cu” r S ” e i )cn n20)l eh as re c a a Sor o “X sr c ,ot u k e e ve n ve i ally furnished requesting users information necessary to individual file transfers after receiving individual requests for files.Ide, i s r ned Am t ’ es

5 found, respondents (unlike Napster) could disappear tomorrow and the software would continue to allow users to find and transfer files. Pet. App. 45a.3 Respondents have post-distribution contacts with users, but none that plays a role in finding and transferring files. Like Microsoft (and Sony, presumably), respondents engage in routine activities like offering updated versions for users who w n t m ad( G os r cs) rv i t hi l uath e n i rkt ’ ae poi n e n a sp n es dg c c pr Moevrw e ue g o l er pnet sf a ot r e hn sr o n i ,e odn ’ot r . o , s n s s w e automatically and periodically contacts servers run by respondents that send advertisements (for which respondents are pi t t ue s o pt .T e de i m n t nm s o a )o h sr cm u r h avrs etr s i i d e ’ e te a sn plays no part whatever in file searching or in file transfer, the direct infringement. It occurs whenever a user is on line with the program launched on the computer, regardless of whether the user is active at all. JA87-88.4

servers actually found the file and instructed the host computer to send it to the requester. See W. Fisher, Promises to Keep: Technology, Law, and the Future of Entertainment a 12(04 ( i N pt adSor t 1 20) “k as r n cu, le e Am t w s‘ n azd i t snet th cm ayso nsre i s r a c t le’ n h es h t o pn’ w e r e e ri e a e vs managed all search requests, and thus closing down those servers would dso t ntok) et yh e r” r e w . Pti e ’ xe fsbl y p e t twt u r pnet peet nr epr it a l oi d h , i ote odn ’ r io s tr d n a h s s s ec,hisf a w u “er [ ” vri eJ 47 bth ol enet rot r ol dga e oe t ,A 5, u t n x e w e d d] m e y planation he gave said much less: that bugs might not later be fixed if respondents disappeared, id. At deposition he acknowledged the obvious: if any such bug did not impair the file-sharing functionality at the outset, it would not somehow come to do so over time; the code does not change, so there is no spontaeu “er ao.J 19-202. nos dga t n A 16 di ” Post-distribution contacts include other incidental communications (Pet. App. 15a, 18a; JA453-61), but the record is undisputed that, for the versions of the software at issue, none of these contacts, including those reli t t “u . l fe st c ayeiseal r pnett an o h at x ” i a iat l x t nb s e odn o tg e om l u l s, e s s monitor or control what files users search for or download or upload, and nn inee fr sr fe oe s edd o ue ’i-sharing activities. JA72-73, 87-90, 109-12, s l 1172-9 (et nr eprs 5 pti e ’xe ’concessions), 1207-08. io s t
4 3

6 A nm e o d f et po cl ( st f o pt -lanu br f ie n “rt o ” a e o cm u r fr o s e gaecnet n,i t “ye etr s r rt o ”o ug ovn osl e h hpr x t nf po cl r i k e t a e o , http, for the World Wide Web) exist for P2P software. One, called FastTrack, was developed by Kazaa B.V., which, with affiliated and/or successor entities, also wrote P2P software using that protocol. Grokster is a licensee of that software, whose source code Grokster does not control or even see. Pet. App. 41a. StreamCast (under an earlier name, MusicCity) oi nl lesdK za sf a , u i Fbur 20 r i l i ne aa’ ot r btn era 02 g ay c s w e y Kazaa cut off StreamCast, which, while retaining the name Morpheus, then switched to P2P software that uses the “ntl po cla e t n atr k Id. at 44a. Wheregu l” rt o r hrh Fs a . ea o t a Tc as the FastTrack protocol is proprietary to certain Kazaabsdeti ,h gu l po cls oe, i.e., generally ae n tst ntl rt o i“pn ie e ea o ” available for others to use and modify. Id. at 6a; JA95. Within each prot o t d f et bad”o t sf a o l h ie n “r s fh ot r c , e fr n e w e interoperate: their users can share with each other. 2. Benefits of Peer-to-Peer File Sharing. The defining ca c rt o r pnet sf a ii eal g f ic hr tii fe odn ’ot r st nb n o d et a e sc s s w e s i r user-to-user communication without central control. The decentralization offers large efficiency benefits, independent of ayi u o l a lb i. I “i icn yr u[ ]t n s e f e l i it t s n i t e c s h s g a ly g fa l d e e
The versions of the software at issue do not provide users the location o a it n n cn c( ruh ro sproe” o i t t f fsol e ot tt og “ t ue ds) rn a -messaging r i a h o n sn connections (Pet. App. 42a-45a) and do not depend on any user registration or log-in. JA73-74, 84, 100-01, 773-74. Petitioners mischaracterize the record in accusing respondents of willfully removing earlier Kazaaimposed registration and log-in requirements. MGM Br. 10. Kazaa caused the change for both respondents—when it cut off StreamCast from the Fs r k rt o (rm t g t a C ss h to h gu l po cl at a po clpo p n Sr m at si t t ntl rt o T c o i e ’ f e ea o , which does not use such features), JA84, 708-22, 1058-59, JER8715, and when it unilaterally changed the software it licensed to Grokster (which had no control over the change in its software), JA73-74, 773-74. Even as to the earlier software, evidence submitted by both respondents and petitioners stated that file-sharing continued to function even when the registration server was unavailable or a user name was deleted (a new one could be created). See JA73, 201, 575-76, 1058-59; JER8710, 8724.

7 distribution costs of public domain and permissively shared art and speech, as well as reducing the centralized control of that d tbt n Pt p.6. r t shm e e bnf. ir uo. e A p1a Ce o t sl s eet si i ” . ar e v i The economic benefits derive from several undeniable facts: there is tremendous storage and communication bandwidth capacity already present in the mass of end-ue ’ omputers sr c s and connections. If those capacities can be harnessed to be accessible by any individual user seeking material of any form, i.e.,f a rlogt y ue aed r i s no e ue ’ im ti suh b a sr lay e d o t r sr ea r se h s computers that can be accessed over already-in-place communication lines, then it is a wasteful redundancy also to store the material on a group of central servers, and to install and operate high-volume communication lines into and out of those servers, especially at high enough capacity to ensure service at times of maximal usage. Both storage and bandwidth capacity are saved by not centrally duplicating what is already efficiently available in the end-user base. The benefits of decentralized peer-to-peer arrangements (compared to server-based arrangements), if set up well, go w lbyn cssv g.O e s rbs es:na et l e eod otai s n i“ ut s i cn a l n o n ” rized system, a crash at the central repository (of memory, or of communication links) takes down the system; in a peer-topea agm n ay i lue s ii l l v vast numerr ne et n s g sr d fu ye es r , n e ’ fc t a br o o e acs b . A o e i “cl it”i.e., the e f t r ces l s hs i e nt r s sa b i , h a ly organic ability to self-adjust to different levels of demand for particular data. At the low end, peer-to-peer arrangements fs r m l“i e cm uie i e s di seii or ot s a n h” o m n i n r t n pc l e l c ts t e e azed arcane material; for example, early fans of an unsigned performing artist (one not signed by the tight-knit world of record labels that control the scarce pre-Internet distribution resources) can share their interest and, indeed, benefit the artist through mouth-to-mouth promotion. Peer-to-peer arrangements also provide efficiency benefits at the high end of demand: the more any particular material is sought, the more end-user computers will have it, and render it accessible over the millions of diverse communication lines serving those

8 computers, all without bulking up storage and bandwidth capacity on a central server. Even Napster had scalability problems, though it centralized only indices (not the music) on its servers. J. Menn, All the Rave: The Rise and Fall of Sa n an gs as r (2003). hw F ni ’N pt 247 n e Achieving such desirable functionality with substantial cost savings is practically a definition of economic progress. All such efforts, of course, present challenges in actual implementation, making experimentation necessary. But peer-topeer file sharing arrangements indisputably offer promising economic benefits. And the benefits are not only economic. Such arrangements enhance political freedom in those places where suppression of particular material might be attempted. When access to material relies on a central server, suppressors can attack the hub (disable or delete material from the server, snip the wires). Peer-to-peer arrangements, by eliminating the hub, allow access to samizdat material stored on any of millions of computers connected through millions of links. It is harder to suppress a book if 50 million homes, not only central libraries, have copies. See P. Huber, O w ls r e’ l Revenge 170-78, 182, 239-40 (1994). In short, easy-to-use peer-to-peer file-sharing software has obvious benefits, economic and otherwise, for creators as well as consumers. See Address by Patrick Gelsinger, Intel Chief Technology Officer, Aug. 24, 2000, Condemning its distribution as unlawful, as petitioners urge, would cause real social harm. B. District Court Proceedings Pti e ’si aee t tr pnet ( d K za et nr u s lgd h e odn a io s t l a s s n aarelated persons not before this Court, Pet. App. 7a n.4) should be liable for every infringement committed by users of their software. JA15-64. In September 2002, respondents moved for summary judgment: Grokster in full; StreamCast only as to whether its then-current software (the gnutella-based

9 Morpheus product introduced in March 2002) was secondarily infringing, the critical question for going-forward purposes. (Petitioners also sought summary judgment, but they do not mention their motion here, MGM Br. 15, reflecting the narrow scope of the ruling at issue.) The district court rendered a carefully circumscribed ruling in April 2003 (and reiterated the limits in June 2003, JA12132) Iep i dt tt a “os e i ] n w e e t 1. t xln h iw s cni rn ol ht rh ae a d [g y h e current versions o G os r ad t a C ss rdc ad f rkt ’ n Sr m at pout n es e ’ s services subject ehr a yt lb i . PtA p 2a I i e pr o i it” e p. 8. t t t a ly . epes ec dd“ e uso w e e ehr e nats xr l xl e t qet n ht r i e D f dn i sy u h i h t e liable for damages arising from past versions of their softw r o f m o e psat ie. Id. All questions about a , rr t r at cv i ” e o h i ts software and particular activities from before April 2003 (or perhaps before the September 2002 motions—it makes no difference here) were left for further litigation. Wi i t df e l i,h cut r t r pnet t n h e nd i t t or ga e e odn ’ h e i ms e nd s s m t n ( d ei pti e ’ o o,o h et t it overo osa dn d et nr m t nt t x n i n e io s i e e lapped). Id. at 24a-25a.5 T ecut o o e t sC ut h or fl w d h ors l i ’ Sony dc i ad h Nn Cr is eio n t i h i u ’Napster decisions. sn e t ct Rejecting contributory infringement, the court found key f tntngni d pt R sodn “l r ko t t a s o i eu e i u : epnet c a y nw h c n s e s el a many if not most of those individuals who download their sf a sbeun y uei t i r g cpr h ”(e ot r usqet s t o n i e oyi t Pt w e l fn gs . 6 App. 36a; id. at 31a) ; u “ e a sbt tl oi r g g btt r r us n ann fni he e ai ni n

A t aybodr set o G os r o pti e ’ o ost s o n rae apc f rkt ’ r et nr m t n,h s es io s i e court effectively deferred the issues until further proceedings, denying the motions (without prejudice). 10A Wright, Miller & Kane, Federal Practice and Procedure: Civil 3d § 2728, at 525-26 (1998). Pti e r et l asrt t a l s9%” fh ue fepn et nr e a d s th “te t 0 o t s o r oio s p e y e a a e s dn ’ ota w s o i r gm n MG B., -3 (citing Pet. App. et sf r a frn i e et s w e fn . M ri2 4a; JA439). The district court nowhere so found. Nor did the Ninth Circi w i , th ce pg, o dm r yt t et nr “ u , h h a t id ae nt e l h pti e allege that t c e t e e a io s over 90% of the files exchanged . . . involves copyrighted material. Pt ” e . App. 4a (emphases added). That is twice re oe f mpti e ’ lm m vd r et nr c i . o io s a


10 ue fr e nat sf a ,peetn ft e id. at 33a, sso D f dn ’ot r ” r n ad u r ( e s w e s u 3a ad Pa tfd nt i u t t e nat sf a 4) n “ln f o o d pt h D f dn ’ot r , i is s e a e s w e is being used, and could be used, for substantial noninfringing proe” id. at 34a). The court also found that users of reupss ( sodn ’ ota “onc t t r pcv ntok,e pnet sf r cneto h e et e e rsss w e e s i w lect which files to share, send and receive searches, and download files, all with no material involvement of Defendn . Id. a 4a A ar u : If either Defendant closed at” s t 5. s e l “ st their doors and deactivated all computers within their control, users of their products could continue sharing files with little or no interruption. Id. (emphasis added). Noting that ” r pnet“ae ne ae e ott ao as t g sr e odn hv udr kn f r o vi s sn ue ” s s t f s d ii s in infringement (id. at 46a) and offered users some technical support and software updates (id. at 46a-47a), the court conc dd htepnet i t ed “r nti icn y i l e t r odn ,n h n,a o s n i t d u a s s e e g fa l f ferent from companies that sell home video recorders or copy
All petitioners can legitimately point to is the recognition that, today, the “at a ry fh fe a ecagd lgl . Pt p.a vsm j i o t is r xhne ieay e A p8. ot el e l l” . Pti e ’ n pr n g ei new soeeprss d, h h et nr ol e et e v ec a n xe ’ t y w i io s y c a d t u c did not cvrSr m at sf a , u ol t Fs r kpo cl oe t a C ss ot r bt n h at a rt o e ’ w e y e T c o (Grokster, Kazaa, and others). JA434-49. The expert did not assert, or prove, 90% infringement. He said only that (a) 75% of the files in his sm l w r pti e ’cpr h d w rs( 49—some of which a p e et nr oyi t ok J 3) e e io s ge A could well be shared without infringing (as fair use)—and (b) another 1% w r “kl cpr h d ( 49—some of which is certainly 5 e l e oyi t ” J 3) e i y ge A available for sharing without objection by the rights-holders (groups like Phish and the Grateful Dead). T e xe ’sm l g m r vrhd uh eprs a p n, oe e a sb t i o , stantial weaknesses (JER7355-81): e.g., the numbers represent files offered, not transferred; only English-language words were searched, thus excluding many video-game and software files (and foreign-language materi)at r e frsa n; h sa hec dd “dl m t i a u oi d o hr g t er xl e au ” a r l l h z i e c u t ea (identified by sex-related words in their titles), much of which may be lawful and authorized for sharing; the results were skewed toward music files, which have more searchable data; of all the music files from Grokster users, 95% came from only four users. JER7378-80. Notably, even a 10% figure for noninfringing downloads, applied to pti e ’ . b l nf ue( M B. 2, en 20m lo l fl et nr 2 ii i r MG io s 6 lo g r1)m as 6 ii a u ln w downloads each month—significant by any sensible measure.

11 machines, both of which can be and are used to infringe copyr h . Id. at 48a. i t” gs Regarding vicarious liability, the court observed that it was ud pt t t D f dn dr eaf ac l ee trm ni u d h “ e nat e v s e a e s i i ni bnf f n a i o t i r g g odc o t ue o t isf a ,eas h n i i cnut fh sr fh rot r bcue e fn n ” e s e w e “ s n i n poot no D f dn ’ de in r eu a i ic t rpro f e nat avrs g e ne g fa i e s ti v dpns pnt i r gm n ” PtA p 4a, 50a. (Neieed uo h n i e et e p. 9 e fn . . ther the district court nor the Ninth Circuit found that respondents could not sustain their small businesses without the infringing activity. Contra MGM Br. 3.) The court found, however, that the required element of control and supervision under the existing arrangement with users was undisputedly m s n: t r i n am s b ei nebfr t C ut i i “ e s o d i i e v ec e e h or sg h e sl d o e indicating that Defendants have the ability to supervise and control the infringing conduct (all of which occurs after the product has passed to end-ue ) PtA p 5a T e sr . e p. 4. hr s” . e were disputed issues of fact about whether respondents could feasibly and effectively change their software (id. at 52a5a bt i r u lb i cno r t n“ ef tht 3) u v a os i it ant e o t a t a , ci a ly s h c a product could be made such that it is less susceptible to unlawful use, where no control over the user of the product eis Id. at 54a. x t” s. After this ruling on part of the case, petitioners sought an immediate appeal. The district court, in June 2003, amended its April ruling, entered a partial final judgment (Fed. R. Civ. P. 54(b)), and certified the ruling for appeal under 28 U.S.C. § 1292(b). JA1213-21; Opp. App. 1a-10a. The court again emphasized that it was ruling only o “ e cr n vros n t ‘ r te i ’ h ue sn o G os r ad Sr m at sf a ad sri s f rkt ’ n t a C ss ot r n e c . es e ’ w e v e” JA1219-2.T ecut o daa t t et nr “a a0 h ornt gi h pti e hd d e n a io s duced no evidence that Defendants materially facilitate or contribute to the file exchanges that form the basis of these l si” J 11) n t t et nr “ae s n a u s ( 27 ad h pti e hv es tially not w t A a io s e d pt t t e nat sf a hs ur t n pt tl i u d h D f dn ’ ot r a cr n ad o n a s e a e s w e e ei future substantial non-i r g g ss ( 11) n i i ue”J 28. fn n A

12 C. The Court of Appeals Decision The Ninth Circuit affirmed on the limited issue before it— “ait a s gf m peet cv i . PtA p 4.T e lb i ri r r n at ie ” e p. a h i ly in o s i ts . court first ruled that, under Sony, respondents cannot be liable based on the offering and promotion of their software because, as the district court found to be undisputed, the softw r i “aal o sbt tl r o m r ays a s cpb f us n a o cm e il ignificant e e ai cl nn fni ue. Id. at 9a (footnote omitted). Declaraoi r g g ss ni n ” tions show use of the software for sharing public-domain works and many works authorized for distribution, including by commercial performers. Id. at 10a-11a. Whatever the “at a ry o ue a (r i b)the lawful uses are vs m j i ” f ss r o wl e ot e l , subs n a ad“ae o m r a v b i . Id. at 11a, 12a. t tl n hv cm e i i it ” ai c l a ly Sony’t tshr oe e se it e rm t s ef . There was no other basis for contributory infringement, the cut de. “ ] t r t a C s nr rkt maintains oradd [ eh Sr m at o G os r N ie e e cn o oe i e fe” t yd nt r u t o poi ot l vr n x is;h o o “ gle r rv e r d l e e a d acs ; n ue cu cn ne hife ces ad sr ol ot u t ri-sharing activities ” s d i e l (licit and illicit) even if respondents disappeared tomorrow. Id. at 13a-16a. Moreover, although respondents communicate wt t ue o t isf a ,h cm ui t n a “o i h sr fh rot r t o m n aos r nt h e s e w e e ci e t f itenr gm n ” Id. at 15a-16a, 18a. Respondents o a la i i e et cit fn . thus lack the required knowledge of specific infringements at the time of a material contribution. Id. at 8a, 13a. The court of appeals a o a i e t d tc cut l fr d h ir t ors s fm e si ’ rejection of vicarious liability. What respondents undisputedly lack, in the current versions of their software, is any “ i t n ait t spri ’ i i da ue ’ hr go ‘g ad b i o ue s ” n v ul sr sa n f rh ly ve d i s i individual files. Id. at 17a. Napster, in sharp contrast, actually was in the middle of every file transfer. Id. at 18a19a. The law of vicarious liability does not require a defendant to redesign its product to create an ongoing intermediary, conduct-controlling role that it does not currently have. Id. at 19a-2a H r pti e ’ ei ne fh r h 0. e ,et nr “v ec o t i t e io s d eg and ability to supervise [is] little more than a contention that

13 ‘ esf a ie cu b ae dt peet sr f m t ot r t l ol e lr o r n ue r h w e sf d te v s o sa n cpr h d is” at 19a. hr g oyi t fe. Id. i ge l ’ The court ended by noting that its decision was limited to “ e pc i sf a i ue the time of the district court t seic ot r n s at h f w e decision, wt o e cnute a i frigt ni t ” i t r odc r i n o li i n h h h m ng t ao e district court. Id. at 20a-21a (emphasis added). Noting that content markets traditionally have adjusted to disruptive but beneficial new technologies, the Ninth Circuit reiterated this C ut am n i i Sony, 464 U.S. at 456, that legal ors d oio n ’ tn change in the area is properly left to Congress. Pet. App. 21a-22a. SUMMARY OF ARGUMENT Peet oe i l poi s ospoto pti e ’ r n gvr n a rv e n uprfr et nr s ng w d io s claim in this Court that respondents are secondarily liable for all infringing uses of their software. The claim fails under Sony’c a rlt th gnr of i o a rdccno s l ru h t ee l f r g f poutant e e a e a en support secondary liability unless the product has no significant noninfringing uses. Nor can petitioners otherwise prevail on the two grounds for secondary liability that the 1976 Congress recognized when it gave copyright holders the exclusive right to “u oi ”cpi ad o e at (7 at r e oy g n t r c 1 h z n h s USC § 16:“otbt yi r gm n ”w i targets ... 0) cn i o n i e et h h r u r fn , c prc ao i ao e sseic ko nat o i r ga ipt n n nt r pc i nw c f n i e ti i h’ f, s fn m n “i r u lb i ,w i targets defendants that have et v a osi it” h h ; ci a ly c control over t d et nr gr h i c i i e sspecific activities. H.R. e r fn ’ Rep. 1476, 94th Cong., 2d Sess. 61, 159-60 (1976); see Sony, 464 U.S. at 437-38 & n.18. Accordingly, petitioners need a change of law to prevail here. But any legal change in this area should be left to Congress. I. Respondent sf a i i u b hs i icn nn s ot r n s t l a s n i t o’ w e dp a y g fa infringing uses as an article of commerce, and the Sony rule therefore prevents contributory-infringement liability based o r pnet gnr poio o t isf a .Pti n e odn ’ ee l rv i fh rot r et n s s a sn e w e io e ’ a os ru et r vr u a m n —specifically, about whether infrings i g s ing uses predominate or make the product commercially

14 viable for particular defendants, about intent to profit, about possible design changes—are plain departures from the Sony rule, as is the whole attempt to replace a clear rule with an uncertain multi-factor standard. Modifying the Sony rule as petitioners suggest would, in fact, undermine Sony’ cr s oe policy. It would deter investment in innovation by subjecting innovators to standards that are unpredictable in application and expensive to litigate, and put large sectors of the digitaltechnology economy in the hands of entertainment-industry incumbents with a vested interest in preserving their existing business arrangements, to the detriment of both creators and consumers. II. There is no other basis for liability here. Beyond challenging the distribution of software itself, petitioners claim contributory infringement based on past acts of respondents that allegedly encouraged infringement or assisted individual, known acts of infringement. But these contentions, which would not support en-masse liability for all infringing uses of r pnet sf a r a l sa s p otd t l e odn ’ ot r e r e ,r i l u i h i s s w e g ds e m y se e m id cp o t rl go apa I us bu r pnet t soe fh u n n pel s e aote odn ’ e e i . s s s past conduct were not decided below, are not before this Court, and remain to be litigated in the district court. Petitioners also assert vicarious liability, contending that respondents, by adopting a different design for their products, would be able to control, and thus prevent, all the infringing acts committed by users of their software. But, as the Government recognizes, the vicarious-liability standard recognized by Congress does not embrace the provider of a product that, as is indisputable here, gives the provider no control over i cs m r individual uses of the product. Petitiont ut e ’ s o s e ’ds n r ei -change proposal would impermissibly extend s g vicarious liability beyond the principal-agent context without congressional authorization, contrary to Meyer v. Holley, 537 U.S. 280, 285-91 (2003), and would threaten innovation by subjecting product design to expensive and indeterminate judicial second-guessing, contrary to Sony.

15 III. With the Sony standard clearly governing this case, and requiring affirmance, this Court should leave to Congress any legal changes relevant to the peer-to-peer setting. Stare decisis weighs heavily in the statutory context, particularly when Congress is actively engaged in the area. Despite some sky-is-falling pleas, there is no urgent need for judicial action. A dpti e ’ eho g-specific grievance demands, at n et nr t nl y io s c o most, a technology-specific legal response, when there has been enough market experimentation and the myriad facts are reliably understood. This Court could afford petitioners relief only by expanding the general standards of secondary liability, with inevitable collateral harm to numerous industries beyond the present P2P context. Congress, on the other hand, can not only consider the technological and market facts relevant to when and what action is needed, but choose among complex, regulatory, balanced, experimental responses to the specific problem. Congress has repeatedly done just that in other contexts. Sony has served well as the background general rule and should remain so. ARGUMENT I. R so dnsDs iui o T e S f ae s a fl ep n et ir t n f hi ot r I L w u ’ tb o r w Under the Express Rule of Sony v. Universal Sony establishes a clear rule that no secondary copyright liability attaches to the general distribution of a product (or the normal incidents of such distribution) as long as the product has or is capable of substantial noninfringing uses. That simp rl c a yap e hr Pti e ’ ru et l u l r pls e . et nr a m n e e el i e io s g s depart from Sony and would undermine its vital policies. A. Sony Protects General Distribution Of A Product If It Is Capable Of Significant Noninfringing Uses This Court in Sony r et tom v s d s “nr e e c d w oi t i ’ upe j e e uo c dented attempt to impose copyright liability upon the distribut so cpi eu m n a a e ot og ecopyright o f oy g qi et s n f r t i r n p ” f v hl r “ot l vr na ieo cm e et ts o t o e cn o oe a rc f o m r h i nth ds r tl c a e sb c o cpr h po co. 44US a 41 id. at 441 uj t f oyi t rt t n 6 ..t 2; e g ei ”

16 n 1( x ari r t sget sc cn o oe pout . “ t od a o ugs uh ot l vr rdc . 2 er n y ” r ) S n ud pt l ke t tt B t a m ci s w u oy ni u d nw h i e m x ah e “ ol s ey a s a n d be usdt r odcpr h dpor s ( at 426), and e o e r oyi t rga ” id. c ge m survey evidence showed that, almost certainly, most users were, at one time or another, using the machines to make cp so e t ns p t “m si”( a hapor oi t r h i l o t e h t w t e h a m y i f c rga m once, later than when it aired)—though such time shifting w s t pi a ue . . fr ot w e . Id. at 423.7 The a“ e r r s. h m y o m so nr” s Court nevertheless rejected secondary liability, and did so by adopting a clear rule of law. Governing Principles And Existing Precedents. The Court began with the pertinent principles, explaining that copyright protection “ w o y tu r” “a nvreon e a i hl s t oy and hs ee r gi d n s l at c z at r r h t asl ecn o o h w r. Id. at 431, u o’ i to bo t ot l f i ok h s g u r s ” 43 n 3 T eet to po co r u e a“ii l 3 . . h x n f rt t n e i s d f u 1 e ei qr fc t bl c”t tt a be C nr s (o t cut)o t a ne h ihs en oge ’ nth or’j o a a ss e s b strike and to adjust. Id. at 429. Thus, Congress has repeatel eat t gt sl i so rb m r sd y s n i d nc d a e d o t n t pol sa e b “i iy e r e uo e i gf cn cagsi t ho g, sc a p yr i o, ht at hne n e nl y uh s l e p ns po c o ” a a o copiers, cable and microwave television transmission, and audio tape recorders. Id. at 430 n.11, 431. The Court sesdt “ d i ys e c ne oepn t po cos t s h j ia ’ r ut c t xad h rt t n r e e u cr l a e ei afforded by the copyright without explicit legislative guidance” and i “e r c t C nr s hnm j t ho t df e e o oge w e a re nl s en s o c ogi lnoaos lrh m re fr oyi t m t is c i vt n ae t a to cpr h d a r l” a n i t e k ge ea , bcueii C nr st t a “ ei tu oa ait t eas ts oge h hs t n i t nl b i o s a h st i ly accommodate fully the varied permutations of competing interests that are inevitably implicated by such new technoloy Id. at 431; see id. a46 C nr s a ea i “ i g. ” t 5 (oge m y xm n t s s e h
Fully 44% of users had more than 10 tapes in their library, and 42.1% indicated plans for multiple viewings even of programs taped the previous m n .44 ..t 2 n .T e i et o d h “uvy so i t t ot 6 US a44 . h d sn nt t sres hwn h h 4 s e e g a the average Be m x sr w s e en 5 n 3 t e. Id. at 483 n.35; t a ue o n bt e 2 ad 2 a s a w p ” see Universal City Studios Inc. v. Sony Corp. of America, 480 F. Supp. 429, 438-49 (C.D. Cal. 1979); 81-1687 Universal Br. 5 n.11 (69-75% of ue “ a tna ei a e o of sr m i a l g l r i f f s n i r b rs -the-air recordins) g” .

17 new technology, just as it so often has examined other i oaosnh ps” u t k so C nr snt or. n vt n it at bta ifr oge ,oC ut n i e , s s ) Within this framework, the Court described the limited bases for secondary copyright liability in existing precedent. Id. at 434-37. Its own precedent, Kalem Co. v. Harper Bros., 222 U.S. 55 (1911), did not approve liability for merely “up i t ‘ en’oacm lha i r g g cv y spl n h m ast co p s n n i i at i , yg e i fn n i t” btno e a e natht spldt [infringing] work u i l d df dn t “up e he vv e a i ie . 44US a 46 Ide,h df dn s cos t l” 6 .. t 3. ned t e nat at n— sf e e ’ i making a movie of the book Ben Hur, then distributing it to theaters for exhibition—may have been dramatizations, exhibitions, productions, presentations, performances, or representations of the book, directly infringing under § 1(b), (c), or (d) of the 1909 Copyright Act, 35 Stat. 1075. Subsequent lower court precedents for secondary liability, Sony ep i dw r l idi o i “n no g e t n xln , e i t , vl n a ogi r aoae eme n vg n li ship between the direct infringer and the contributory infne a t t et i r g gcnut cur , w e r gr th i h n i i odc ocr d hr i e m e fn n e” e t lt “ a i a oio t cn o t ue f oyi t h ae w sn psi o ot lh s o cpr h d e tr tn r e ge works by others and [in fact] had authorized the use without pr i i . 44US a 47( phasis added).8 The Court e s o ” 6 ..t 3 e m sn m readily concluded that Sony could not be held liable under such pre-existing precedents, because Sony had no direct invl m n wt ue ’t i ad hd nti l ne o o e et i sr a n n a o n u cd r v h s pg fe encouraged infringing taping. Id. at 438. The Court so held dsi t s d sv oos ru etl e et nr a u ep eh t i ’ i ru a m n (k pti e ’ r t e uo g g i io s g ment here) that Sony, knowing that many users would engage in unauthorized copying, was inducing infringement by promoting its product for copying through advertising and genT eC ut o dt asneo lb i o l d rsw od “o h or nt h bec f i it f a l d h o nt e e a ly no prc a d et i ayi r g gat i , cm a n lb i o a a ipt i cy n n n i i cv y o pr g i it f ti e r l fn n i t” i a ly store owner for infringement committed by the concessionaire hired to run t s r s w r oddpr et as part of the store, and liability of a h t e o n e r ea m n e o’ c t m ngr h po o d ta ispr r acs i “c ako l g” aae w o rm t i rs ’ e om ne wt at l nwe e e s tt f h u d of the works they would perform. 464 U.S. at 437 n.18.

18 eral technical assistance. The Court underscored the narrow meaning of “i c i o e et i ayi r g gat i ” d etn l m n wt n n i i cv y r vv h fn n i t (id. a 47 w e i rl t tS n “e a l de nt t 4) hn t u d h oy cr i y os o e a tn i et nl i u[ ”ue t i r g o “upyi pon n oay n c ] sr o n i e r spl t rd t i l d e s fn s ucts to identified individuals known by it to be engaging in continuing infringement.”Id. at 439 n.19 (emphasis added). The Product-Distribution Rule. Having easily dismissed any liability under existing standards, the Court noted that no copyright precedent supported the studi ’ e ad o dm n—to ims ps lb i frsln “qi et i t cnt cv oe i it o ei eu m n wt h osut e a ly lg p h e r i knowledge of the fact that its customers may use that equipm n t m k uat r e cp s f oyi t m t i. eto ae nu oi d oi o cpr h d a r l h z e ge e a” Id. at 439. Turning to patent law as the best source of a rule defining when such copyright liability could be allowed, the Court explained that 35 U.S.C. § 271(c) sets a very stringent test: there is no contributory liability for distributing unpatet a ie “n s t ya ‘nu e fraycm e n d rc s ul s h r usid o n o m r e tl e e e t cial non-i r g gue ” 44US a 41 qo n Dawson n i i s. 6 .. t 4, ut g fn n ’ i Chemical Co. v. Rohm & Haas Co., 448 U.S. 176, 198 (1980); id. ( no use except through practice of the patented “ ‘ m t d” “ h im m s almost be uniquely suited as a e o’ ‘ e t h ; t e ut component of the patented inven o’ ( pae add. t n” e hss de) i ) m That principle determined how to strike the “a ne e en bl c bt e a w acpr h hl r l im t dm n fre et e oyi t o e s e t a e ad o f cv—not g d ’ gi e f i merely symbolic—protection of the statutory monopoly, and the rights of others freely to engage in substantially unrelated a a o cm e e Id. at 442. The Court expressly held: r s fo m r . e c” Accordingly, the sale of copying equipment, like the sale of other articles of commerce, does not constitute contributory infringement if the product is widely used for legitimate, unobjectionable purposes. Indeed, it need merely be capable of substantial noninfringing uses.
81-1687 Universal Br. 12-13. The studios cited evidence that Sony, oe cusl a vr one s dvice, had even refused to include a warning against un’ authorized copying in its advertisements, fearing lost sales. Id. at 45 n.97; see Universal, 480 F. Supp. at 459.


19 Id. The question was not what uses predominated, or whether the product could be modified to reduce infringement, but “ ht rh B t a icpb o cm e ially significant w e e t e m x s aal f o m r h e a e c 10 nn fni ue. Id. oi r g g ss ni n ” The Court held that test readily satisfied by private, noncommercial time-sii “ h t g both (A) because [the studios] fn have no right to prevent other copyright holders from authorizing it for their programs, and (B) because . . . even the unauthorized home time-sii o r pnet por ss h t g fe odn ’ rga i fn s s m l im t f r s. Id. (emphasis added). The sentence, by e t a a ue gi e i ” its structure, indicates two reasons each of which suffices to meet the test. Indeed, that is the only apparent reason the Court discussed the first category at all, because time-shifting, held to be fair use (id. at 447-55), was at the time the “r a ue . . fr ot w e ” id. at 423). (The fair-use pi r s . m y o m so nr ( s question warranted resolution regardless because of its obvious independent importance to users.) The district court decision, which this Court reinstated, had expressly found that Sony satisfied the patent-law standard even if the primary use
The Court adopted that standard, which makes no reference to possible product alter i see t uht d sn nt t t S n m yb ao , vn h g h i et o d h “oy a e tn o e s e a able . . . to build a [video recorder] that enables broadcasters to scramble t s nl fni dapor s n ‘ m t uat r e r od g f h i ao i v ul rga ad j ’h nu oi d e ri o e g di m a e h z c n t m”44 ..t 9. h s d s l r etedly pointed this Court to h . 6 US a44 T e t i a o e a e uo s p t ps b i o “ cnl i l oict n frht e proe 8h os it f t ho g a m d i i ” o t vr ups. 1 e i ly e oc f ao a y 1687 Universal Br. 2 n.4, 48, 54. They cited evidence, as well, that Sony, upon learning that broadcasters could deploy a jamming mechanism, engineered a way around it. Id. at 2-3 n.4, 45 n.97. And they contended that S n’ r ua t cut ac aft i m cai so e t t t r oys e slo on nne ie n ehn m hw d h “ e f e lr g s a he would be little, if any, market for [video recorders] if they could not be ue frnr g g upss Id. at 54. sd o i i i proe. fn n ” Contrary to petitioners (MGM Br. 33-34), the Court articulated its dualuse-product standard with no reference to any possible technological filtering measures, and did so before any discussion of fair use. Moreover, the possible filtering that petitioners stress in this case (id. at 10-11) would, like the technology mentioned in Sony, prevent copying of particular works outright, whether or not the copying was a fair use.

20 (time-shifting) was not fair use. Universal, 480 F. Supp. at 457-62. The first significant permissible use, then, was to copy material authorized for copying. This category, focused on sports, religious, and educational programming (464 U.S. at 44& n 4,h C ut em d“i icn dsi t l 4 . )t orde e s n i t ep e h o 2 e g fa ” t e w percentage of all uses it represented. Id. at 444. The dissent nt t tabs “n 7 % o a B t a ueso eod o d h ,t et ol . f l e m x s it r r e a , y 3 l a c soteeto a k d” id. at 494 n.45), and the surveys pr vn f l i s ( s s l n evidently showed that sports, religious, and educational material constu d“ s t n9 o r od g.8-1687 Unii t l sh %” f e ri s 1 te e a c n versal Br. 52-53. This figure—cm a b t pti e ’ o pr l o et nr ae io s dubiously low 10% figure for legal use here (see note 6, supra)—was enough. Indeed, the Court explained that a cpr h hl r m ynt r a u oyi t o e “ a o pe i nless the relief that he g d vl seeks affects only his programs, or unless he speaks for virtually all copyright holders with an interest in the outcm . Id. at 446 (emphasis added). oe ” The dissent in Sony (besides disagreeing about fair use) proposed a standard for secondary liability that highlights t po cvns o t C ut rl T e i et o t l h rt t ees fh ors u . h d sn pi e y e ei e ’ e s nd oj t t t C ut r et no scna lb i “s b c d o h ors e co f eodr i it a ee e ’ j i y a ly l ga t pouts capable of substantial noninfringing o s h rdc i ‘ n e ue. Id. at 498. As to actual uses, the dissent did not pross” ’ pose a predominance test condemning products whenever misuse was a majority use, as the Ninth Circuit in Sony had done. Rather, it said that there should be no secondary liabili a l ga “ s n i n pro o t pout uei t s o s a i ic t ot n fh rdc s s s y n g fa i e ’ nn fni ” id. at 491), evidently approving liability only oi r g g ( ni n i“iul a o t pout ue . . it i r g,i.e.,i f v t l l fh rdc s s . r ay l e ’ so n i e fn ” “ f no one would buy the product for noninfringing purposes a n. Id. With even the 9% figure for authorized copying le o ” uncertain, the dissent urged a remand to determine the proportion of infringing use. Id. at 492-93.

21 B. The Sony R lP oetR so dnsS f ae ue rt s ep n et ot r c ’ w Both courts below correctly recognized that respondents in this case had easily established not only that their software was capable of significant noninfringing uses, but that it was being so used. The district court explained that Sr m at ta C ss e ’ Mopes ir u r ue t f ite n sa h o pb c rhu “ e ll sd o a la ad er fr ul s g ay cit c i domain materials, government documents, media content for which distribution is authorized, media content as to which the rights owners do not object to distribution, and computer softw r fr h hd tbt ni pr ie, ad“]esm i a o w i ir u o s e td n [h a e s e c si i mt ” t t e f rkt. PtA p 3a cao to evidence omitr o G os r e p. 3 (ti u e” . it ns t ) E e pti e ’i r w u sg hundreds of mile . vn et nr f ue ol ugest d io s g s d lions of legitimate downloads a month. Note 6, supra. Numerous performing artists, from Janis Ian, Bela Fleck, and John Mayer to bands such as Phish, Pearl Jam, and the Dave Matthews Band, have authorized free file-sharing. JA125-29; JER645-51. The Internet Archive relies on P2P file-sharing to distribute many (over 17,000) concert recordings from over 700 musical groups. JA137-42. (See also ACLU Amicus Br.) Computer por s( hr a ”ad rga m “ awr n s e e “ ew r ) a wdl athorized for such distribution. fe a ” r i y u r e e e JA160-61; JER7608. Ventures such as the Prelinger Archives, GigAmerica, Reel Mind, and J!ve Media have built commercial businesses by using peer-to-peer file-sharing to promote and distribute hundreds of thousands of authorized copies of music, video, and video-game content. JA143– 64, 322-24; JER502-05. (See also Harvard Berkman Center, Content and Control (Jan. 7, 2005), App. III.) Project Gutenberg and the Internet Achieve, to reduce their costs, rely on P2P software to share public-domain works (JA130-43, 16971)—such as Shakespeare plays, the King James Bible, the Koran, the Communist Manifesto, and Pa ’ d l us all los io e, t ag fud y et nr epr(JER752.109-13). on b pti e ’xe io s t T ecpb i o r pnet sf a frsl fr e h aait f e odn ’ ot r o tl ut r ly s s w e i h legitimate uses as a communication tool, helping creators as well as consumers, is clear from its inherent economic and

22 other benefits. See pages 6-8, supra; JA149-64; Gelsinger, supra. Even focusing just on musical performers today, those benefits offer otherwise-unavailable opportunities for reaching an audience. Many, perhaps the great majority of, artists receive no royalties from CD sales under major-label recordcompany contracts,11 which serve instead chiefly to attract audiences for concerts or other activities. Free online filesa n o sc a isft e ui(feoddcne so hr g fuh rs ’u r m s o r re ocr, f i tt u c c t non-major-label recordings) may serve that function at much lower expense. Many other artists are not signed at all by the four major labels. For them, using online file-sharing opens w o y e ps b ie fr xoue P Ps ee to m s hl nw os itso epsr 2 ’bnfs f al i li . i sive cost savings, robustness, niche-market service, and scalability clearly make it capable of substantial legitimate uses. Contrary to the Government (US Br. 5, 23-24, 17), nothing in Sony’ c a rl r u e t t oi r g gue sfc s l r u e i s h nn fni ss uf e e e qr a ni n i to make the product commercially viable, let alone for the particular defendant. Cf. MGM Br. 35. This Court reversed the Ninth Circuit in Sony without any such inquiry, and the C ut r e ne o“o m r ays n i n . . . ue” m ors e r c t cm e il i ic t ’ fe c l g fa ss i ports no such requirement. 464 U.S. at 442. The term “o m r a intvnpro t epes te et fh cm e i”s o ee a fh xr s t n o t cl t e s am e Sony rule, id., because the term plays such a limited role: to assure merely that the noninfringing uses add value as an article of commerce, i.e., are among the reasons consumers want the particular product (rather than a substitute with lesser features). Not surprisingly, the source of Sony’ rl su , e pt ta ,m oe n “o m r ayv b ”r u e et a n l i pss o cm e il i l e i m n e w cl ae q r , bt e l r e t a“rc o cm oi o cm e e u m r y e r o a ie r o m d y f o m r ” e fs tl t c being suitable for substantial noninfringing use (35 U.S.C. § 271(c)), excluding only uses that are “ r t e,l sr, f f c d iuoy ae h l
See Fisher, at 19-20, 55-56, 77; Content and Control, at AI-4; Testimony of Amicus AFTRA (2001),; Courtney Love, Courtney Love Does the Math, Salon,

23 i pata o m r yepr et. D C i m Patents m r i l r e l xe m n l cc e i a” . h u , s § 17.03[3] at 17-62 (2004); Intellectual Property Owners A s ( O Amicus Br. 12-13. snI ) ’ P The reasons for rejecting the proposed commercial-viability inquiry are clear. It would senselessly make some businesses liable, and others not liable, for the identical conduct of selling the identical product. And it invokes a concept with no consistent, nonarbitrary meaning or workable application in the business world. Even for a product, the contribt nim kst poi (s“o m r a v b i ” i a u o t ae o rfs i cm e i i it ) n i t t c l a ly particular multi-product firm is rarely a stand-alone proposition, but depends on inherently indeterminate accounting for potentially massive common costs (e.g., R&D, administration, shared equipment and facilities, marketing), equally indeterminate judgments about how it helps sell other products, and other factors. Businesses commonly guess wrong about their products, as the entertainment industry well knows and as recognized in the deferential business-judgment rule. The problems get markedly worse if particular uses of products, not the products themselves, must be evaluated for their contribution to profits. Indeed, as intrinsically uncertain as such questions are in a mature business, they can be positively nonsensical for innovations in digital markets, where venturecapital investments often are made and businesses launched without concrete plans about how to make money, trusting that a good idea will eventually bear f i Pti e ’ui r t et nr j u . io s d cial inquiry, aside from arbitrarily distinguishing firms engaged in the identical conduct, thus presents unknowable risks for investors (and for noncommercial innovators). See Dg aMeiA s ( i ) it il d s n DMA Amicus Br. 12-13. a ’ In short, the products at issue here, like the product at issue in Sony, fall squarely within the protection of Sony’ rl su. e Here, as there, despite known widespread misuse, the products provide legitimate benefits to large numbers of users; and it is the same product capabilities that enable proper and improper use. Here, as there, the products benefit many crea-

24 tors of copyright-protected works and copyright owners themselves. Sony’rlt s rt tt gnr of i ad su h po c h ee l f r g n e u es e a en promotion of the software at issue here, and continued innovation in that software, against patent-like control by copyright holders. Unable to satisfy Sony’ c a rl pti e t t f l s l r u , et nr r o a e e io s y l back (MGM Br. 17-19, 27) on Sony’ osrao t t oy s be t n h cpvi a r h l “ uttk a a ne e en cpr h hl r i ta m ssi bl c bt e a oyi t o e s g w re a w g d’ legitimate demand for effective—not merely symbolic—protection of the statutory monopoly, and the rights of others freely to engage in substantially unrelated areas of comm r . 44US a 42 T i t a et f e e 6 .. t 4. h r t n o Sony is imperc” s em missible. The relied-on statement merely formulates the statutory-construction question.T e ors h C ut answer was that ’ a dual-use product is noninfringing—in short, is part of a “us n ayur a dl eo cm e e—whenever it is sbt tl ne t i f o m r ” a il le n c “ e l . . . cpb o sbt tl oi r g gue. Id. m ry e aal f us n a nn fni ss e ai ni n ” It is the answer, not the question, that establishes the governing rule having precedential force, even for cases presenting different facts, as the Court recently affirmed. Clark v. Martinez, 125 S. Ct. 716 (2005). C. The Sony Rule Safeguards Innovation In Products With Legitimate Uses Copyright liability for general distribution of a dual-use product inevitably harms the legitimate uses of a product as well as impeding its improper uses. Avoidance of that harm as beneficial new products are created is the central substantive rationale for the clear Sony rule. T i C ut a r on e t i pr neo t l ’ h or hs e gi d h m ot c f h a s s c z e a e w po co o “ s t i t t rdcs noao adeo rt t n f rk a n h poue i vt n n cei i kg a n i nm c rwh o “a ga [ g t i et eo noa . o igo t” f sf ur i ]h n n v t i vt” , e dn e c i n e Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, 540 U.S. 398, 407 (2004). It is now well accepted that economic progress is fundamentally dependent on continuing innovation—on what Joseph Schumpeter called the

25 “e ni gl o c av dsut n t og t ho g pr n l a f r t e et co, h uh e nl i e a e ei r i ” r c ocal advances tested and perfected in the marketplace.12 Protection of product innovation is the central commitment of patent law specifically.13 Sony adopted a strict rule precisely to avoid allowing general copyright duties to trench on patent l ’poeac oe t ho g ai oao,ev g oa s rvnne vre nl i ln vt n l i C n w c oc n i an gress to enact any needed technology-specific copyright rules. 464 U.S. at 421, 430 n.11. Sony’ c a rl po c ntuth bnfs of the pars l r u rt t o j t ee t e e es s e i ticular innovation at issue here, including the dramatic lowering of search-and-distribution costs that can provide an audience for many speakers, writers, and performing artists that cannot secure scarce entertainment-industry resources and attention. The rule protects beneficial innovations like phonographs, radio, cable television, the VCR, photocopiers, ad cs t dcsS n’ Wa m n TV , D and DVD ui as t ek, oys l a, io C o ee k bre ,n A p ’io.A d t rt tyt unr ad pl sP d n ipo c e s e e s -unrealized, or unimagined, future improvements and innovations that will benefit creators and consumers tomorrow. Abandonment of the protective Sony rule would threaten innovation in obvious ways—as indicated, e.g., by the already-filed (light-green) amicus briefs of DiMA, IPO, and
J. Schumpeter, Capitalism, Socialism, and Democracy 84 (HarperPerennial ed. 1976); see F. Scherer, Innovation and Growth: Schumpeterian Perspectives v (94 ( eho g a cag hshd adwl i 18) “ cnl i l hne a a, n i i t oc l continue to have, much more of an impact on material well-being than the niceties of static resource allocation to which microeconomists devote most of their aet n) Bol , tn o” rd y The Economic Goals of Antitrust: t i ; e Efficiency, Consumer Welfare, and Technological Progress, 62 N.Y.U. L. R v12,06 18)“ cnl i l rgessh s g m si pr e.0012 (97 ( eho g apor it i l otm ot oc s e ne tant factor in the growth of real output in the United States and the rest of t i utazd w r ” W.B u o The Free-Market Innovation h n sii e d r le ol ) d; am l , Machine: Analyzing the Growth Miracle of Capitalism (2002). 13 Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722, 730-32 (2002); Markman v. Westview Instruments, Inc., 517 U.S. 370, 390 (1996); Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U.S. 141, 146 (1989).

26 the Business Software Alliance (BSA), who represent large segments of the economy. See also IEEE Amicus Br. Innovation is a multi-s ppoeso “ vn o, n er t rcs f i et n et pe e n i r 14 nusi i et etdvl m n add fs n er p n s n ee p et n iui . All of h, v m , o , f o” this costs money.15 The greater the risks and costs of either liability or litigation, the less investment will flow into the costly steps of innovation, to the detriment of future creators and consumers alike. A departure from Sony’ rl would su e dramatically raise these risks and costs, which cannot be managed by the courts. Cf. Central Bank of Denver, N.A. v. First Interstate Bank of Denver, N.A., 511 U.S. 164, 189 (1994) (rejecting aiding and abetting 10b-5 cause of action, ci “ne a t ad xes eigt n) in ucr i y n ecs v li i ” tg tn i t ao . D. Petitioners P o oa Wo l At T eSony Rule ’ rp sl s ud l r h e And Impair Its Rationale Pti e ’ ru eta w u i o e dpr r f m et nr a m n l ol n l a ea ue r io s g s l d vv t o the Sony rule, not an application of it. Initially, petitioners seem to suggest by their quotation of G r wn ul C r. es i P b g op h ’ v. Columbia Artists Mgmt., Inc., 443 F.2d 1159, 1161-62 (2d Cr 17) adr et r e ne t “s sn”i r gi 91, n e a d e r cs o as t g n i e . p e fe ii fn ment, that liability attaches to providing tools that materially aid copying, with general “nweg o t i r g g cv ko l e fh n i i at d e fn n i i ,r hrhn nowledge of specific infringing acts. MGM t” a e t k y t a Br. 17,18, 23, 26, 27. That suggestion misreads Gershwin
F. Scherer & D. Ross, Industrial Market Structure and Economic Performance 616 (3d ed. 1990); id. at 616-2 Shr , t ( net 0 ce ra 6 “ c h e o e necessary inventions are available, the development of new products or processes to the innovation stage depends largely upon the allocation of human and material resources in order to solve through costly trial and error the detailed problems of technical advane)B u o a1. c” am l t 0 ; , Shr & R s a 65( ehi l noaosd nt a l e ce r e os t 1 “ cn a i vt n o o f l i , T c n i l k manna from heaven. They require effort—the creative labor of invention, dvl m n t t gad n out n n t sem o eoo ile ) ee p et e i ,n i rdco i o h t a f cnm ci . ; o , sn t i t er f” cf. AT&T Corp. v. Iowa Utilities Bd., 525 U.S. 366, 429 (1999) (Breyer, J cnur gi pradd sn n i pr (o n t “ ei et et . ocrn n a n i et g n a ) nt g h t n s n , i t s i t i e h v m ncs r t poue o p x eho g ai oaos) ees yo rdc cm l t nl i ln vt n” a e c oc n i .
15 14

27 (which involved specific knowledge) and contradicts Sony. 464 U.S. at 437 n.18 (discussing Gershwin). Sony itself provided an essential tool for copying, which it promoted, with general knowledge of infringing use. Pages 17, 19 & nn.9,10; Sony dissent, 464 U.S. at 458-59 (Sony promoted taping shows and building libraries). The Court squarely held that conduct insufficient to support liability. Petitioner l s w ei a u et a eul adpr s e s ep g r m n r qay ea ’ s n g s e l ture from Sony. For one thing, petitioners refuse to state a clear rule of law at all, but present a mishmash of considerations, whereas Sony sharply separated the general offering of a dual-use product from participation in specific known acts of infringement and then stated a clear rule of law for the former situation. In any event, petitioners depart from Sony in the particular considerations they advance as potential bases of liability: (a) that infringing uses of a product predominate; (b) that the product offeror intends to profit from increased use, including misuse, of the product; and (c) that the product could be redesigned to reduce misuse. MGM Br. 23, 26, 27, 30-38, 40-44; Songwriter Br. 10-14; see also US Br. 13-21. Each of these variations would alter Sony’ s clear rule. And they would undermine its rationale of protecting innovations that have legitimate uses, by introducing debilitating uncertainty and risk to would-be investors in the vast realm of digital products, which are inherently useful for copyrighted content. Predominant Use. By its plain terms, the Sony rule protects products regardless of whether illegitimate uses predominate at some moment (which can change drastically over time). 464 U.S. at 442. Even as to present uses, the Court br dlb i a l ga t pouts wdl ue” i a e i it so sh rdc i“ i y sd wt r a ly n e e h out infringement—which says nothing about whether other uses predominate. Id. The Court then lowered the threshold for protection slfr e t pout edol b “aal tlut r h rdc ne ny e cpb i h: e e o sbt tl oi r g gue. Id. That standard, highf us n a nn fni ss ai ni n ” lighted by the dissent, precludes any predominance inquiry,

28 which could hardly be conducted as to yet-unestablished uses. A predominance inquiry a oii os t t i t C ut l sn nie wt h ors s c sn h e ’ holding that all i nee t cni rw s“ ht rt t edd o os e a w e e h d h e Betamax is capable of commercially significant noninfringing ue,not the totality of possible uses. Id. ss ” A predominance inquiry, moreover, is foreign to the patentlaw standard that Sony expressly borrowed, as this Court itself described that standard. Page 18, supra.16 Condemning distribution of a dual-use product under a predominance test is incompatible with the central idea underlying the capableof-substantial-noninfringing-use standard in both the patent and copyright contexts: that the legitimate uses, when more than insignificant, should not be sacrificed by condemning the product. And, in applying its standard, the Court in Sony relied on and protected the category of authorized copying, at most only 9% of uses. Page 20, supra. Sony’ rl t s su h e u involves no predominance inquiry. Nor should the Court require such an inquiry now. Even aside from stare decisis, a predominance inquiry would threaten innovation by embroiling investors in incurably unreliable and costly litigation. See Wu, C pr h s o oyi t C m g’ munications Policy, 103 Mich. L. Rev. 278, 349 (2004). At the early investment stage, there are no usage data available at all, and no one may have any reliable idea what uses will predominate once the product is launched. Thereafter, measuring usage may be difficult, expensive, and uncertain, as this case illustrates (see note 6, supra); so too will assessing what uses are legal, because copyright standards, particularly fair464 U.S. at 441; see Menell Amicus Br. 10-11 (arguing against borrown f m 3 USC §21c bt o sget gt t pi a ue i r g o 5 ... 7( , u nt ugsn h “r r s” ) i a m y standard can be found there); Dawson, 448 U.S. at 199 (for patent-law cn i t yi r gm n s na i“oue xethog pate f otb o n i e et t dr s n s ecp t uh r i o r u r fn ,a d r cc t pt t m t d)Alloc, Inc. v. ITC, 342 F.3d 1361, 1374 (Fed. Cir. h a n d e o” e ee h ; 2003) (same), cert. denied, 124 S. Ct. 2390 (2004); C.R. Bard, Inc. v. Advanced Cardiovascular Sys., 911 F.2d 670, 674 (Fed. Cir. 1990) (same).

29 use standards, are notoriously full of gray areas and expensive to litigate, as Sony and Campbell v. Acuff-Rose Music Inc., 510 U.S. 569 (1994), indicate. In any event, and decisively, the mix of uses is highly likely to change greatly over time, making judicial assessments of predominance a form of guesswork and precluding any meaningful finality. Beyond that, the facts themselves are readily subject to control by a tightly knit set of incumbent businesses (studios and labels): by deciding either to authorize or not to authorize use of their content, they can themselves dictate the proportion of lawful uses. The mix of VCR uses, for example, obviously changed greatly when the studios agreed to license their movies for cassette rental and purchase. How is a court to factor in the effect on usage data, present and future, of the incentive to protect existing business arrangements against threats from new technologies? More to the point, how is an investor to predict, long in advance, the likely outcome of all the uncertain judicial inquiries? A predominance inquiry w u “ r k ao i pate DiMA Amicus Br. 25; BSA ol w e hvcn r i . d a cc ” Amicus Br. 13-16; IPO Amicus Br. 13; IEEE Amicus Br. 12-14. Profit Motive. Equally problematic, and foreign to Sony, would be an inquiry into whether the product seller intends to profit from increased use, including infringing use. The Court in Sony did not rely on lack of profit motive; nor could it have. Sony plainly profited from greater use, including infringing use. The Betamax was more valuable the more it could be used, so that Sony could sell more and perhaps even charge more. Moreover, Sony sold more tapes the more the machine was used; and library-building, not time-shifting, drove the need for tapes. See Sony dissent, 464 U.S. at 486; Universal, 480 F.Supp. at 462 (tapes); notes 9, 10, supra. Not surprisingly, Sony’ c a rl t n nt t lo poi s l r u u s o a a n rfe e r l t making intent. To condemn a profit motive would, in fact, paralyze investment in any product with potential infringing uses.

30 Investors and their companies would routinely have it held against them that they seek to make money off increased use o t pout H w vrre h qi t t ‘ ] m nbt fh rdc o ee t t u h “ no a u a e . u e p a [ b cha ee w o ,xeto m ny ” l ked vr rt ecp fr oe, Acuff-Rose Music, o e ’ 510 U.S. at 584, investors in innovation typically hope (eventually) to make money off the resulting products. They will fully recognize that the more consumers use a product, the more valuable it is to them, the more they will buy, and the more they will be willing to pay (in cash or other value). It is therefore practically inherent in a dual-use product that the sellers will profit from misuse as well as proper use, and they will be subject to a finding of intent to profit from misuse, c an a a otui r ll a rk udrpti e ’ r t g n l s n e a e l i ne et nr ei m vs g s io s 17 approach. Redesign. Finally, the Sony rule protects, and must protect, distribution of products without regard to whether they might be redesigned to reduce infringement. Possible modifications t S n’B t a w r epes ad eet l cld o h o oy e m x e xr l n r a d ae t t s a e sy p ey l e C ut aetn See note 10, supra. Yet the Court anors tn o. ’ t i nounced its test and applied it to rule for Sony without finding ay ed o nu en sc r ei s Pti e ’ ri o n ne t i i i o uh e s n. et nr ug g f qr t d g io s n such an inquiry, like their other suggestions, is thus flatly contrary to Sony’ epesrl w i , hnapout a s xr u , h h w e s e c rdc hs significant legitimate uses, requires that copyright holders target misuses, not the tool, or seek relief from Congress. The prospect of litigation over possible design changes would inhibit investment in innovation. Anyone contemplating a new product, or investing in the development of one,
The claim that respondents sought users that previously had used Napster (MGM Br2)ai f m r e i t “atcnut o bfr . 5,s e r e rn o ps odc nt e e d o frg ” o this Court, adds nothing. To market consumer P2P software at all, with its tremendous legitimate benefits, is necessarily to seek former Napster users—80 million people, at its height (Fisher, at 111)—who may employ the technology for wholly legitimate uses. Of course, the industrysupported pay service initially called PressPlay, acquired by Roxio, has recently taken the Napster name precisely to attract Napster users.

31 would face the real prospect that one or more copyright owners—potentially threatened in their existing business arrangements—would sue to launch uncertain and expensive litigation over how the product might have been designed differently. What design changes are possible, and how would they actually affect misuse? What would they cost to implement, and (relatedly) how close to perfection in eliminating misuse must the design changes be? How would the ds ncags de e a eth pout pi o l iei hne avr l f c t rdc s r e re t g sy f e ’ c gi mate functionality? Many of these essential questions will be empirically unanswerable, turning on mere predictions about untested marketplace responses and the costs of implementing something new and untried. Most if not all will trigger complex disputes among experts, including those relying on third-party companies (like some amici here) touting their wares as design changes. JA181, 224, 228 (cited MGM Br. 11). The district court here found a genuine dispute over the costs and benefits of filtering alternatives (Pet. App. 52a53a), and this case is anything but idiosyncratic. As the G vrm n r on e, uha i u y“ ol oe et e gi s sc n n i w u n c z qr d have the undesirable effect of chilling technological innovation and constraining the product development options of dvl e o sf a ado e d ileho g s U ee pr f ot r n t r i t t nl i . S o s w e h ga c o e ” 18 Br. 19-20 n.3 ; see DiMA Amicus Br. 2-3. Congress, in related legislation, itself has recognized the overriding difficulties with demanding such design changes. The 1998 Digital Millenium Copyright Act (DMCA) added protections for cpr h hl r ec p o e ot (7USC § 1201oyi t o e ’ nr t n f r 1 ... § g ds yi f s 1205) and safe-harbor provisions for Internet service companies (§ 512). In the former context, Congress repudiated inquiries into possible design changes firms might make to
The Government, having recognized this obvious point, fails to recognize that the same chilling effect is present under an approach that treats redesign as one factor in a secondary-liability standard—especially one that adds uncertainty to the legal regime. See US Br. 19-20.

32 a cpr h hl r ec p o e ot ( 10( () i i oyi t o e ’ nr t n f r § 21c 3) n d g ds yi f s ) ; the latter, it likewise repudiated inquiries into possible changes the service companies might make to assist in monitoring infringement (§ 512(m)). The Sony rule should not now be altered to add such design-change inquiries. The just-discussed investment risks presented by all of pti e ’pooe m d i t n t Sony—ranging from et nr rpsd oic i s o io s f ao predominance to redesign—are anything but speculative. Copyright holders have been nothing if not aggressive in suing (Lemley & Reese, Reducing Digital Copyright Infringement Without Restricting Innovation, 56 Stan. L. Rev. 1345, 1347, 1384 (2004); Fisher, at 133), and litigation over the issues petitioners propose to inject would be uncertain and expensive. That is so not only for the companies involved, but for individuals: in Napster and other cases, copyright holders have directly sued investors, officers and directors, and even lawyers, embroiling them in costly litigation long after the company went bankrupt. Litigation costs aside, the prospect for investors includes the dramatic consequences of guessing wrong about all of the issues petitioners would inject into the liability standard. Large investments in developing a product can be wiped out by enjoining distribution (or by litigation costs alone). See Fisher, at 133 (features of Replay TV dropped after suit and subsequent bankruptcy). As important, the nondiscretionary “tu r dm gs o a i r gm n i o e” §54a s t oy a ae fr l n i e et n l d ( 0( , at l fn s vv ) (c)), which the plaintiff can elect instead of proving any actual damages or lost profits, can be ruinous, given that a mass-market product is commonly used by literally millions of people. Statutory damages are set at $750 to $30,000 ( dc lt $0 it i r gr w s o a a and had no r ui eo 20 fh n i e “ a ntw r e b e fn e reason to believe”iw s nr g g. §54c 1 (2) (emt ai ii) fn n 0( (), ) phasis added). The text and legislative history indicate that s t oy dm gsa ol a r ad t tt “ i m m tu r a ae r b gt y n h h m n u at e i o a e i s t oydm gs o ec w r m sb a a e” ne o tu r a ae fr ah ok ut e w r d oc fr at d each different individual infringer of the work (seemingly

33 once for each non-identical group of liable infringers). 1976 House Report at 162; P. Goldstein, Copyright § (2d ed.). Given the millions of users of mass-market products, the mandatory damages quickly reach hundreds of millions or billions of dollars. (A single iPod may hold 3,000-15,000 songs; Apple has sold more than 10 million.) The rigid, severe remedial regime of the Copyright Act underscores the importance of Sony’c a rlto safeguarding investment in s l ru e e innovation across numerous industries. II. Other Possible Bases For Secondary Liability Are Not Applicable As The Case Comes To This Court Congress in 1976 recognized two grounds for secondary copyright liability. As described in the 1976 legislative history and in Sony, 464 U.S. at 437 n.18, both rested on involvement in specific acts of infringement, through participation in or actual supervisory power over them.19 Neither ground supplies any basis for secondary liability on the only issue before this Court—w e e r pnet gnr sply ht re odn ’ ee lup h s s a of their software renders them liable en-masse for all infringing uses.

The17 H ueR pr d cs n “otbt yi r gr”gv 96 os eot i us g cn i o n i e , ae , s i r u r fn s as its sole explanation a classic case of direct participation in infringement (of the public-pr r ac r h i t ea p g e)“o ea p , e om ne i t n h xm l i n: F r xm l a f g, e e v e person who lawfully acquires an authorized copy of a motion picture would be an infringer if he or she engages in the business of renting it to o e fr upss f nu oi dpb c e om ne Id. at 61. The t r o proe o uat r e ul pr r ac. hs h z i f ” sm r otd cs n “ ] a os lait ” o dt r et no a a e e r i us g [ i r u [i ly nt h e co f n p , s i vc i ]bi, e e j i am nm n “ ee p t por t so a et lh etsc a a ed et t xm t h rpio f n s b sm n uh s o e er ai , ballroom or night club, from liability for copyright infringement committ b a i eedn cn at ,uh s n r et l dr Id. at 159. e y n n pnetot c rsc a a oc sae e ” d d r o h r a . T e o m te xlnd “ ob hl a h C m ie ep i : T e e related or vicarious infringer in t ae d the case of performing rights, a defendant must either actively operate or supervise the operation of the place wherein the performances occur, or control the content of the infringing program, and expect commercial gain from the operation and either direct or indirect benefit from the infringing pr r ac. Id. at 159-60. e om ne f ”


34 A. Claims Of Contributory Infringement Liability For Urging Infringement Or Assisting Specific Known Acts Of Infringement Are Not Before This Court A t iscn bs fr cn i t y nr gm n” esh reod ai o “otb o i i e et pt e s r u r fn , i tioners assert that respondents, beyond engaging in normal distribution of their product, affirmatively urged infringing uses and otherwise participated in individually identified, specific, known acts of infringement. E.g., MGM Br. 25-29. But this contention is simply outside the case in its posture in this Court, which concerns only ongoing en-masse liability fr sr i r gm n bsdo t sf a ie , o ay o ue ’n i e et ae n h ot r t l nt n s fn s e w e sf isolated past acts with isolated effects. The district cut ol rl gw s of e t t “uors n u n a cni d o h cr ’ y i n e r tsf a ad e i s fepnet ec d g lps e ” ot r n sr c o r odn ,xl i a at n w e ve s s un l versions of the software and past activities. Pet. App. 28a. Only that ruling was certified for appeal. JA1213-31. The Ninth Circuit limited its decision to that ruling, i.e. to respondn ’ ur tota ad hiogi rli i oe t n et cr n sf r n t r no g o n t pr i . s e w e e n e s ao That limit was jurisdictionally required by 28 U.S.C. § 1292(b). Yamaha Motor Corp., U.S.A. v. Calhoun, 516 U.S. 199, 205 (1996).20 Nothing else is before this Court.
The grant of partial summary judgment to respondents is not properly treated as a refusal of an injunction appealable under 28 U.S.C. § 1292(a)(1), see Carson v. American Brands, Inc., 450 U.S. 79, 84 (1981); 16 Wright, Miller, & Cooper, Federal Practice and Procedure: Jurisdiction 2d § 3924.1, at 175-80, 193-96 (1996), but even if it were, review would be limited to the only arguable refusal of an injunction that exists—t d tc cut l idrl gt th cr n sf a ad h ir t ors i t u n h t ur t ot r n e si ’ me i a e e w e services supported no secondary liability. See id. § 3921.1, at 25-27, 39, 42. (The denial o pti e ’ u m r j g et o o, i n r f et nr sm a u m n m t n wt o e io s y d i h quest for a preliminary injunction, is clearly outside Section 1292(a)(1). Id. § 3924.1, at 172-73.) In the Ninth Circuit, the MGM petitioners recognized the limited scope of the issue, i.e.,h l fl s o r pnet “ ur t e i so t i t a u es fe odn ’ ‘ r n vro ’ fh r e w n s s c e sn e products and services,ntere vros fhintok or other past ” o “a i e i o t r e rs lr s n e w activities. C MG B. ( paiadd.I sei a i m d t ” A M r7 e hs de) n ek g n m eie m s n a appeal, they affirmatively asserted that what the district court had decided

35 Everything else remains for litigation in the district court regardless of the outcome here. Within the scope of the ruling at issue, then, there plainly was no proof of the required participation in direct infringement. Sony readily rejected any direct-participation liability even though Sony was supplying the means for infringement, had general knowledge that many if not most users were engaged in unauthorized copying (including illicit librarying), advertised the product for use in copying, refused to include warnings in advertising, provided instructions on use, sought to profit, and (presumably) warrantied and updated its product (all to make copying easier). Petitioners have shown nothing more here. They have not shown that (within the limits of the ruling at issue) respondents urged, encouraged, requested, argued fro po o d “ dcd) n i r g g c F r lt i o, r rm t ( nue” ay n i i at o a h r e i fn n . le efforts to intimate otherwise, petitioners have identified no such communication to users promoting unauthorized sharing o cpr h dfe ( e l e a “rm t nl a r l f oyi t is t y a l s po o oa m t is ge l h b i ea ” certain pitches to would-be advertisers, not to users). MGM Br. 6-8, 25.21 In any event, in the Ninth Circuit petitioners presented no such evidence, but claimed only knowledge (even that, at the en-masse level, which is what matters), and
(h l at udrh C pr h A t f rkt ’ cr n ss m ad “ e e ly ne t oyi t c o G os r ur t yt ” n t gi e g es e e “ el at udrh A t f ta C ss ur t ss m )s distinct t e ly ne t c o Sr m at cr n yt ” i “ h gi e e ’ e e and severable” rm w aihd o dc e ( h C utxr s r e e f o htt a nt ei d “ e orepe l e r d d t sy s v dc i o c i sno i G os r ad ta -C sspast versions of eio n lm i l n rkt ’ n Sr m at ‘ sn a vvg es e ’ their softa , . o o e psat ie’ .D C. M B. nSco w r . . r t r atcvi ” . tMG e h its ) ro et n i 1292(b) Certification at 3 (first emphasis added). E e a t “at cv i , w i pti e cea e a t oe vn s o ps at ie ” h e et nr i n m i o n i ts l io s t l user (JA808), they ignore the anti-infringement warnings in the very user less fr a i vros fepnet sf a )hyce J 38 i ne ( ere e i o r odn ’ ot r t i. A 3c o lr s n s s w e e t 39, 344, 999-10, 06 T e ir t ornt r pnet e ott 00 10. h d tc cut o d e odn ’ f r o si e s s f s avoid assisting infringement, Pet. App. 46a: e.g., Grokster warned users upon receiving infringement notices from petitioners—who then stopped sending notices, JA75-77, 325-29, 334-35. See JA200, 575 (Morpheus).

36 thus cannot do so here. See CA MGM Br. 28-29, 48; CA Lieber Br. 12-19.22 Nor can petitioners claim here that respondents, knowing of any particular act of impending infringement, lent material assistance to that act. Decisively, petitioners argued in the Ninth Circuit only that respondents were liable for the mass of misuse collectively, which is the only issue of consequence. Petitioners did not argue that, even if there is no general liability, there were certain instances of particularized knowledge for which respondents should be liable. Indeed, the only instances of even-arguable particularized knowledge cited to the Ninth Circuit were five emails in which respondents answered requests for technical help that referred to named works. CA MGM Br. 29, citing JA811, 813, 820, 941, 943. Those emails, and the slightly larger current set (MGM Br. 8, citing JA809-17, 820, 941-5) a “at cnut 4, r ps odc e ” excluded from the ruling at issue. Pet. App. 28a.23

In the patent-law context, 35 U.S.C. § 271 carefully distinguishes conduct-seic i ue et ( 21b)rm cn i t y nr gm n pc i “ dcm n § 7()f f n ” o otb o i i e et r u r fn based on product distribution (§ 271(c)). The latter expressly precludes lb i fra sfhr ia sbt tl oi r g g s” a i Sony); i it o sl it es “us n ann fni ue ( n a ly e e ai ni n s t fr e r u e t th df dn wt t “pc i i et t po h om r e i s h t e nat i h seic n n o re qr a e e , h e f t ” duce infringement, actively “rm t o ecuae t i r g gue po o r nor ” h n i i s. e g e fn n Warner-Lambert Co. v. Apotex Corp., 316 F.3d 1348, 1364 (Fed. Cir. 2003). See MGM Br. 29 (patent-l s na :‘ t e t s r t e t a t dr “ cv s p a a n o w a d ai e e k ecuae )Henry v. A.B. Dick Co., 224 U.S. 1, 48 (1912) (inducement nor ’ ; g” w e i r g g ue“ vkd b avrsm n ) Fromberg, Inc. v. hr n i i s i oe y de i et ; e fn n n te ” Thornhill, 315 F.2d 407 (5th Cir. 1963) (defendant instructed customers how to commit infringement); Chisum, § 17.04[4]. As amicus IPO stresses (Br. 9 n )t “b cvl po o oabhv r r u e ets . ,h oj t e rm t nl eai ” e i m n i 7 e ei y i o qr needed t ao “ cii e et neim ti oao. o vi a h l g f co l t a n vt n d ln f gi e n i ” 23 E e t “atei neai f mbi so yhs v et ek vn h ps v ec,s er e g pt ,a ei n w ae ” d d o n t d nesses. For example, petitioners did not prove copyrights in the works mentioned in the handful of emails from February or March 2002 (JA94154). See Pet. App. 46a n.9; JER59-62. Nor did petitioners prove that the emails actually assisted any infringing act; and some gave advice about the noninfringing act of playing already-downloaded files. JA809-13,


37 In any event, any such isolated events are critically different from conduct inherent in a mass distribution of the software. Only the latter could, as petitioners demand, generate secondary liability for every infringing use of the software and support commensurately broad relief. With narrower kinds of conduct, such as a statement urging use of the product for infringement, the harm would be dramatically narrower, and liability and relief would be commensurately narrow, under the basic principle that relief must be tailored to the wrong. Friends of the Earth, Inc. v. Laidlaw Environmental Servs., Inc.,2 US 17 13 20)“ dr cut 58 .. 6, 9 (00 ( ee lor f a s sol a t esr ‘ h f m n o r i n bodrhn hu i o nue “ e r i fee o rae t d m t a g lf a r u e b t pei f t ’24 For that reason—and e id y h r s a s ) qr e ce c”. because other companies can avoid any particular acts not inherent in the operation of the software—the district court properly recognized that the important issue was liability bsdo t “ur t e i , adpti e egr sp ae n h cr n vro ” n et nr ae y eae e sn io s l rated the ruling on that issue for immediate appeal and limited their claim in the Ninth Circuit. Nothing more is before this Court. B. R so dns L c Of Control Of Specific Acts Of ep n et ak ’ Infringement Precludes Vicarious Liability Wh e et nr c i o “i r u lb i ” s rpr i pti e ’ lm f v a osi it ipoe y l io s a ci a ly l bfr t C utr a i cr n vros f epnet e e h or( gr n ur t e i o r odn ’ o e e dg e sn s s software), MGM Br. 42-50, this claim fails on the merits. As the Government rightly recognizes, vicarious liability in

816, 820 (Morpheus); JA949-50 (Grokster). Such issues remain in the district court. 24 I pt ta ,e ei fr i ui ” nr gm n “ant e x n a n l r d s o “ dc g i i e et cno b ee w m e n n fn pne s a t et lhec s ecn o oe [ s p cm oi . add o s o s b s xl i ot l vr a t l o m d y ai uv r ] ae t” Chisum, § 17.04[3]; Rohm & Haas Co. v. Dawson Chemical Co., 599 F.2d 685, 703 n.24 (5th Cir. 1979), a ’, 448 U.S. 176 (1980) (rejecting fd f “ j co fri i t df dn s a o s p ss c m r sl i i u t n ob d g h e nat sl f t l ,i e e a s nni dn e e ’ e ae n e e nt rnfl f oi “ j co aa s cn ni t ‘ t e i o w ogu ; a r g i u t n gi t ot u g o a i l n” v n nni n i n cv y dc’nr gm n ) uei i e et . fn ”

38 copyright law has not encompassed, and should not now ecm aspti e ’ lm U B.9 n.3, 30 n.6. no ps et nr c i . S r1-20 , io s a At most, vicarious liability applies to a product provider only if the existing produt i st ot l vr sr d et c g e icn o oe ue ’ i c v r s r infringements (as a night-club proprietor has over the performances it offers to its patrons, whether or not it merely promises to refrain from exercising that control). 1976 House Report 159-60 (quoted note 19, supra); Sony, 464 U.S. at 437 n.18; Shapiro, Bernstein & Co. v. H.L. Green Co., 316 F.2d 304, 306 (2d Cir. 1963). Vicarious liability does not apply just because a change in an existing product could give the defendant control over direct infringements. US Br. 19-20 n ( h ‘gtn ait t spri ’ l et f i r u . “ e r h ad b i o ue s e m n o v a os 3 T i ly ve e ci liability . . . has never . . . been held to be satisfied by the mere fact that the defendant could restructure its relations or its product to obtain such an ability. Rather, the law of vicarious liability looks at the extent of control the defendant at l ps s s )3 n.6. After all, in the landlord-tenant c ay os s ., 0 ul e e” cases where no vicarious liability attached (Sony, 464 U.S. at 437 n.18), the lease could have been changed to provide for l d r cn o Pti e ’ o p i aotnet e t a l d ot l et nr cm ln bu i n vs o no r . io s at c i avoid control (MGM Br. 38-41)—m d i t i“otbt y aen h r cn i o e ru r infringem n d cs o—directly challenges the traditional et i us n ” s i limits on vicarious liability, which embody a policy balance, ntet nr oe opti e ’ n-sided approach. io s It is beyond genuine dispute that, as their software actually opera s r pnet l k cn o oe ue ’co e o t , e odn a ot l vr sr hi s f e s s c r s c whether to engage in illicit copying. Both courts below so r on e. PtA p 1a 5a Pti e ’ ru et e e gi d e p. 7, 4. et nr a m n dc z . io s g pends entirely on the claim that respondents could change their software to acquire a control over individual uses not today present. The vicarious liability standard Congress recognized in 1976 embraced no such claim (note 19, supra), and expanding the standard now would be improper.

39 That is so, first, as a matter of institutional role. Recently, this Court explained that vicarious liability traditionally applies only to employers and other principals, rendering them liable for certain acts of employees or other agents carried out on their behalf. Meyer v. Holley, 537 U.S. at 28586 (citing cases). The Court held that vicarious liability should not be expanded to unusual circumstances without specific congressional direction. Id. at 286–1 “ ]ut 9. [ or C s ordinarily should determine that matter in accordance with traditional principles of vicarious liability—unless, of course, Congress, better able than courts to weigh the relevant policy cni r i s hs nt c dt cut d f et . Id. at os e t n, a i r t h or ie n y d ao su e e s f r l” 290-91. That principle fully applies to the copyright setting and precludes any judicial expansion of vicarious liability beyond the circumstances recognized in 1976, leaving any expansion to the policy-weighing discretion of Congress. See also Central Bank, supra (rejecting expansion of implied 10b-5 cause of action to aiding and abetting). Even were that not so, the Court should reject efforts to reconfigure the vicarious-liability standard. Pti e ’ etnr io s argument would substantially undo Sony by changing labels, though Sony itself notes that its reasons for protecting dual-use products bear on vicarious liability as well as contributory infringement. 464 U.S. at 435 n.17. As discussed (pages 3032, supra), secondary liability, under any name, would seriously threaten innovation (to the detriment of creators and consumers) if it turned on assessments of design changes. That is true whether the court is asked to decide that the product developer should have adopted a different design or should not have changed the design of a previous product: innovation in changing products is as important as in creating them in the first place. The Court in Sony correctly refrained from requiring this kind of second-guessing, and in the absence of direction from Congress—which can assess relevant policy risks, e.g., effects on innovation, on product-design incentives, on

40 cpr h o nr i et e t li tds n s e t hbl oy gt w e ’n n vso ig e ei i uso ob i s c i ta g s e new technologies—the Court should continue to do so now. III. Any Expansion of Copyright Liability In the Peer-ToPeer Context Should Come from Congress With the Ninth Cr isj g et or t ne eii i u ’ u m n cr c udr x t g ct d e sn law, this Court should leave legal changes in this setting to Congress. Adhering to stare decisis, the Court should reject pti e ’ rpsdgnr epni o scna lb et nr pooe ee l xas n f eodr i i io s a o y al ity, which would have harmful effects on settings not even before this Court. Only Congress is able adequately to consider and craft a technology-specific legal response to pti e ’eho g-specific grievance, as it has done time et nr t nl y io s c o and again for similar grievances. A. Stare Decisis Weighs Heavily Against Loosening The Innovation-Protecting Sony Rule The Sony rule is necessarily a statutory construction. See page 13, supra; 464 U.S. at 431 (all copyright liability is solely a creature of statute). Sony states a clear rule of law, not a case-specific result limited to the circumstances presented. Such a statutory construction does not change when a new case presents facts different from those which gave rise to it (Clark, supra), and it is strongly entitled to stare decisis. Neal v. United States, 516 U.S. 284, 295-96 (1996); Patterson v. McLean Credit Union, 491 U.S. 164, 172-173 (1989); see Golden State Transit Corp. v. Los Angeles, 9 US 13 12 18)“ rlo l t tsh 43 .. 0, 1 (99 ( u fa h it A e w a e product of judicial interpretation of a vague, ambiguous, or incomplete statutory provision is no less binding than a rule t ts ae o t p i m ai o a tu . . h ibsd n h ln en g f s t e ) a e a n a t” 25
Neal, 516 U.S. at 295-96, stresses that, to assure Congress a fixed target for possible statutory change, statutory precedents are judicially altered only “ hnt i e ei dvl m n o t l hs r oe w e h n r n g ee p et fh a a ‘ m vd e tv n o e w e or weakened the conceptual underpinnings from the prior decision, or where the later law has rendered the decision irreconcilable with competi l a dc i s r o c s” N sc j ti t neisnt s ae n e l otn o pli . o uh u ic i x ti h cs. g g re ie ’ sf ao s i

41 Precedent is even more strongly to be respected in an area where Congress is actively engaged. In Black & Decker Disability Plan v. Nord, 538 U.S. 822, 831-32 (2003), the C ut xlndt t j ia i oao”i prcl l i orep i h “ d i n vt n s a i a y n a e a u cl n i tu r apor t w e “t r ee l c r a eggd ci prpie hr o e f r at s r nae, in a e h da o e ” tg precedent where Congress was the federal actor so engaged (Milwaukee v. Illinois, 451 U.S. 304 (1981)). Similarly, in Trinko, the Court refused to expand monopolization liability because (a) such liability had not previously embraced the duty plaintiffs there alleged, (b) the proposed new liability raised serious questions both about dampening investment i et e ad h cut cm a t entu oai aai n n vs n t or’o pr i i i t nln pcy c i e s av st i c t to make and experimentally adjust the required judgments, and (c) the matter was elsewhere addressed by Congress in an industry-specific manner. 540 U.S. at 407-15. That logic applies here, with differences of detail but not of rationale: pre-existing liability would have to be extended to achieve pti e ’eu , h h a e sr u i et et et nr r l w i r ss e osn s n io s s t c i i v m -deterrence and institutional-incapacity problems, and (with no agency having relevant delegated authority) Congress itself has addressed problems in the area (e.g., the 1998 DMCA), regularly enacts industry-specific measures for these problems, and is actively considering pleas for additional legislation. The reliance interests protected by stare decisis are present here. Patent law itself has relied on the strict Sony rule. IPO Amicus Br. 13-14. And for 20 years, the flourishing consumerelectronics, computer, and telecommunications industries, to name just a few, have been able to innovate without inhibition by any of the legal standards petitioners propose. For example, Apple developed the iPod without facing liability based on legal challenges that its primary use (each iPod holds thousands of songs) was for illegal copying. The same
Nor, in this case, has any agency been vested with implementing authority entitling it to fill gaps or resolve ambiguities in any relevant aspect of copyright law. The United States nowhere claims entitlement to deference.

42 is true for Sony and its CD burner (and, before that, its Betamax). Similarly, Internet service providers and others developed and deployed high-speed services—and the routers that move traffic on the Internet—without facing liability based on legal challenges that their primary use (measured, e.g., in bits transmitted) is for illegal transmission of music or video. Such reliance interests demand preserving Sony, not altering it. B. The Overall Effects Of Peer-To-Peer File Sharing Are Still Uncertain T t et tht et nr p at c a am r po o h x n t pti e ’ l o r t e e a io s e e e oe rplaintiff standard of secondary copyright liability rests on a “k i f l g pe i , M B. 2 they simply have sy s ai ” r s MG ln m e r1-13, not established the premise. Undoubtedly, there is a large volume of file-sharing, there are consumers who obtain songs illegally that they otherwise would buy (though many would not), and the record industry saw a downturn in sales in 20002003, before rebounding in 2004. Beyond those facts, however, there is substantial uncertainty about what harm is being caused to copyright holders. Still less can any such effect be weighed against the harm that liability expansion would inflict on the massive consumer-electronics, computer, telecommunications, and other industries. See BSA Amicus Br. 4-5. At present, there is no urgency for judicial action that could ps b j ty d p cm n o C nr s rli s di os l u i a i l e et f oge ’ o n t y g i y sf sa ss e u n industry and technological facts that bear on selecting among, and deciding when if ever to intervene in the marketplace to implement, any of the many possible legal solutions to the technology-specific problem petitioners present. Claims of destruction must be met with skepticism. The studios cried wolf in Sony. See J. Lardner, Fast Forward: Hollywood, the Japanese, and the VCR Wars 279 (Mentor ed. 1988) (MPAA president Jack Valenti told Congress t t t h “e a h VCR is to the American film producer and the American pb c sh B s n t nl it t w m nhm a n” ul a t ot sag rso h o a o e l e) i e o r e e o . By saving the VCR from legal condemnation, the Court pre-

43 served a new channel for distribution of video content that has been a fabulous success for the industry. Revenues from sales and rentals of home videos are now more than twice the revenues from in-theater exhibition—though the latter have risen as well. Fisher, at 32-33 (2002 figures: $9.52 billion for box-office receipts, $24.42 billion for home rental/purchase). Claims of calamity today are likewise subject to serious doubt. As to the video-content industry, which is thriving, there is no meaningful evidence of harm. Fisher, at 32-3 ( h 3“ e T 26 m v i ut i nt e sm t ac ) Displacement of oi n sy s o yt y p m t.. e d r o i” DVDs, let alone of movie-going (with its communal, eveningout experience) and television viewing, has been severely limited by the size of movie files for downloading, most user s ’ inability to watch downloaded movies except on a small computer screen, and the comparatively poor quality of movie downloads. Not surprisingly, petnr sr y nt6supra) t oe ’uv (o , i s i e e turned up not one movie shared by a Grokster user. JER7379. The music industry presents a more serious claim of harm, bth et t fr eb hr iucr i Pti e ’ e u t x n o t cal a s ne a . et nr ky e e a e m tn io s fact is the drop in record industry sales in the period 20002003, after substantial prior increases in sales. MGM Br. 13 n.11; Fisher, at 32. Yet the downturn indisputably ended in 2004, which saw an increase in CD sales.27 The business of
See Content & Control, at AI-5 & n.22; (2004 home video market reached record $24.1 billion); Johnson, Good News in Hollywood. Shhh., N.Y. Times, Jan. 31, 2005, at C1, C8; Fabrikant, Cable and Film Divisions Bolster News Corp. Profit, N.Y. Times, Feb. 3, 2005, at C4 (petiti e pr t e sC r. r eu i qa e ed g20 rs, o r a n N w op s e ne n ur r ni 04 oe n e ’ v t n “r e b hm -v e sl ” f$. b l nu f m$. b l n) di n y o e i o a s o “1 7 ii ,p r 1 7 ii ” v d e 8 lo o 3 lo . Industry group Recording Industry Association of America (RIAA) reports an increase in units and dollars for the first half of 2004 over 2003. Nielsen Sound Scan (the main industry data source) reports 2004 domestic retail sales of albums rising over 2003 sales. BBC, The Universal Music Group (parent of several peti27 26

44 selling music on line also took off in 2004, proving that userpy dw l d e i sl e pl sT n are not deterred as o n a sr c (k A p ’iues) o ve i e from successful competition (contra MGM Br. 41; Napster Amicus Br.): such downloads exploded from 20 million in 2003 to 200 million in 2004. Washington Post, Jan. 20, 2005, at E2. The 2004 turnaround confirms the substantial reasons to question how much of the 2000-2003 downturn is traceable to downloading rather than to, e.g., abnormally large sales before 2000 as consumers replaced their record collections with CDs, subsequent reduction in CD offerings, diminished interest in those offerings, a retailing shift to Wal-Mart and other stores that stock fewer CDs, and burgeoning competition for entertainment time and dollars from DVDs, Internet browsing, and video games; even one internal industry study attributed only 1/4 to 1/3 of the sales drop to online copying. See Geist, N m esD ntC uc A a s D w l d g u br o’ rnh gi t o n ai , n o n Toronto Star, Nov. 29, 2004, at D02; Mui s r h r u r, s ’Bi t F t e c ge u The Economist, Oct. 28, 2004, An important study by economists from Harvard and the University of North Carolina found no statistically significant negative correlation between levels of CD sales and file-sharing. Oberholzer & Strumpf, The Effect of Filesharing on Record Sales: An Empirical Analysis, 28 FileSharing_March2004.pdf. T e “v ec t t fe h ei ne h i d a l
tioners, representing 1/4-1 o m s sl )eot“t n sl go t / f ui a sr r sog a s rwh 3 c e p s r e ” both for 2004 overall and the last quarter (compared to 2003). _for_2004_ Q4_2004.pdf. See also U.S. Entertainment Industry: 2003 MPA Market Statistics at 55, 57 (tremendous 1999-2003 growth of Internet, video game, and home video viewing); Music Industry Way Off Track With Song and Dance About Falling Sales, Sydney Morning Herald, Mar. 29, 2004, www.smh. (CD sales grew as P2P became popular).

45 sharing has caused losses to the music industry is controvers l Content and Control, at 6. i. a” The future economics of music sales remain unsettled. CDs offer high fidelity and certain conveniences, but still embody high distribution costs. Paid online services like iTunes offer reliability and quality (the song files are not mislabeled or corrupted), convenience, legality. Many artists with limited access to the existing labels can increase sales by gaining exposure from file-sharing. See JA125-29 (Janis Ian declaration); page 22, supra. And there is no telling how sales of various sorts might increase if the industry actively e bae P Ps fc ni .Content and Control, App. III m r d 2 ’ e iec s c fi e (i us g“2 Soe” B t t sot d cs n P P t s) o h hr s i r . h e -term effects, and the longer-term ones, thus remain unclear. See JA149-64. Finally, adverse effects on the recording industry are limidb t i ut ’ ait wt t G vrm n t aak t y h n sys b i , i h oe et o tc e e d r ly h e n , t illicit file-sharing directly. First: Although the record here does not establish what is realistic, the industry may be able to inhibit or limit copying by using target “poi ”o e sof g r d n technologies like those in certain new CDs or in iTunes. See JA158; Fisher, at 129-30. Second: In any event, contrary to the central policy claim that direct actions are impractical for encryption or other reasons (MGM Br. 22, 25), the industry and Government can readily pursue direct infringers. The Government has filed criminal charges, and obtained guilty pleas, against infringing users of peer-to-peer software. 73 U.S.L.W. 2438 (Feb. 1, 2005). The record industry has sued more than 8,000 users, whose liability for statutory damages is daunting. CNET News, +new+fileswapping+suits/21101027_35494259.html; www. The RIAA has touted to this Court the effectiveness of direct enforcement, enabled by its 24-hu at a d“ ec we t t i oe fe or u m t w br l ” h d cvr i o e a r a s s l sharers. 03-1579 Pet. 27 n.18 (suits reduced infringing downloads by half). The Internet, while enabling copying, a o“ cit dt t n o m ss,ed even the movie l f ite e co” f i el ing s a la s e i u a

46 industry to sue direct infringers.29 For that reason, and because P2P software is so widely available from hobbyists and from firms in countries where legality is established, pti e ’ bt cl i t tu gd etn et nr asato c h si i ci io s r g a n r fringers is either impractical or less efficacious than suing software providers (MGM Br. 22) is doubly inapplicable in this context. C. For The Technology-Specific Issues Raised, Only Congress Can Choose Among The Many Possible Technology-Specific Solutions Not only is urgent judicial action unnecessary; it would be singularly inappropriate. To begin with, too many facts relevant to any legal response even for this particular context are not reliably knowable in this case, including technological facts and industry facts that are fast-changing. The relevant facts include the effects of P2P software, both positive and negative, on copyright owners and creators, both present and future; the impact of expanding secondary liability on innovation and investment in the massive industries involving digital technology; and the real-world possibilities for technological or business arrangements that can capitalize on the extraordinary possibilities for cost reduction represented by online distribution, including peer-to-peer arrangements. These and other critical facts cannot be reliably established in this forum. As this Court has often recognized, Congress is better suited to collecting and assessing such facts and deciding when market experimentation has so matured as to make legal intervention advisable. Sony, 464 U.S. at 429-31;

J. Litman, Digital Copyright 25 (2001); Lemley & Reese, at 139192; id. a 19 n 2 (I A et a t t“ 0 pr n o ue t 39 . 0 RA 2 sm t h ‘ i e a 1 e et f sr c s a r pni e o 9 pr n o t i r gm n ” It nt e, e. r e os lfr 0 e et fh n i e et ) n reWek Fb e s b c e fn ’; e 9, 2005, (MPAA anuc gFbur 20 si: ‘ep who have been stealnoni era 05 u s“ ole n y t P i or oi blv t y r aoy os nt It ntadw u nt n u m v s eee h a nnm u o h n re n ol ’ g e i e e e e , d be held responsible for their actions. They are wrong. We know who they a ,n w wlg a e t m a t s si wlpoe ” r ad e i o f rh ,sh e u s i rv. ) e l t e e t l ’.

47 see Eldred v. Ashcroft, 537 U.S. 186, 204-05, 207 n.15, 21213, 222 (2003); Teleprompter Corp. v. CBS, 415 U.S. 394, 414 (1974). As important, this Court is institutionally constrained in what action it can take. The only way to grant petitioners relief here is to loosen the general standards for secondary copyright liability, within the confines of the general remedial regime of the Act. That constraint requires rejection of petitionersc i ,o al sto esn. ’lm frte tw r os a a a First, this Court cannot, through adjudication within the present copyright regime, assess and act on the full range of issues and possible measures that must be considered in sensibly seeking to nurture the potential of, while alleviating difficulties caused by, online file-sharing. The right solution may be to do nothing for a while, as the market continues to experiment; or it may be to craft a solution that addresses specific carefully defined technologies, grants compulsory licenses, creates royalty payment systems, provides safe harbors for particular activities, or makes offsetting tradeoffs in one area to balance measures in another. The challenge is essentially legislative, requiring wide-ranging inquiry, assessment, and experimentation (as the breadth of amici’poos rps als indicates) that this Court cannot meaningfully undertake. See Fisher, at 134-258 (discussing possible legal changes); Kusek, Music Like Water, Forbes, Jan. 31, 2005, www.forbes. com/columnists/free_forbes/2005/0131/042.html; Content and Control, supra; Lemley & Reese, at 1405-25. Second, anything this Court were to do to expand general secondary liability would inherently apply to a vast array of settings not before this Court. In the present age of extensive digitization, vast sectors of the economy involve products specifically designed to handle digitized content, much of which is copyrighted (given the breadth and automatic character of copyright protection): consumer electronics, computer hardware and software, and telecommunications devices and

48 services, among others. Innovation in all those industries has flourished in reliance on Sony. Newly exposing all such industries to serious risks of secondary copyright liability (and the expense of litigation over such issues) threatens that success and would cede innovation across the digital-technology realm to the substantial control of incumbent copyright-owner industries, with their inherent incentive to obstruct new technologies that threaten present business arrangements. Congress, of course, could react to a decision by this Court. But in the meantime the legal basis for the flourishing pattern of innovation will have been undermined; and in any event, any congressional action is overwhelmingly likely to be industry- or setting-specific, given the history of copyright law and the historic lobbying power of the movie and record industries. Thus, a legislative response would almost certainly leave any newly expanded general principles of secondary liability in place only for the settings not before this Court. Any impetus for legal change based on arguments about the present setting should not perversely lead to alteration of the legal rules for other settings, about which even less can be known. Any alteration of Sony should be left to Congress from the outset. D. Congress Has Regularly Addressed Copyright OwnesT cn l y r’ eh o g-Specific Concerns o The Copyright Act is replete with complex provisions tailored to specific problems as they have arisen for 100 ya .T e c s ay uhpoios tsnt n t t er h A t m n sc rv i aet o ol o h s ’ sn t y e range of possible technology-specific measures but to Conges at e s m t n fepni ly o ae n nee r ’ cv as p o o r os it t t ay edd ss i u i s bi k action. Title 17 contains special provisions for libraries (§ 108) and rental of phonorecords, including CDs (§ 109(b)). Detailed rules allow certain performances and displays in schools, restaurants, and other designated places. § 110. Special rules involving compulsory licenses and royalty

49 payments apply to cable television and satellite systems (§11 19 12, opb cbodat gsueo sns § 1, 1, 2) t ul racsn’ s f og i i (§ 118), and coin-operated phonorecord players (jukeboxes) (§ 116). Special rules protect certain copying of computer programs. § 117. Complex rules involving compulsory licenses and royalties apply to digital audio transmissions (interactive or non-interactive). § 114. A special compulsorylicense regime governs recording of songs and other nondramatic musical works. § 115. Complicated safe-harbor provisions protect certain online services, carefully differentiating those which merely transmit material and those which host it or take certain other actions. § 512. Complex provisions govern actions to circumvent encryption measures copyright holders might adopt to control access to their works. §§ 1201-1205. Many of these provisions post-dtts ors a h C ut e i ’ decision in Sony, which has served well as the background rule for congressional action. See Opp. 4-5 (detailing enactments in 1988, 1992, 1995, 1997, 1998, and 2002). An example of a balanced, interdependent regime is the 1992 Audio Home Recording Act, codified at §§ 1001-1010, enacted after Sony was sued for introducing digital audio tape recorders. Congress evaluated the contentions of the music and consumer-electronics industries and enacted (in the Gove m n s od) a a fl dvl e adf e bl cd r et w rs “ cr u y ee pd n i l a ne n ’ e l o ny a l iav cm rm s. U Amicus Br. 3, in Napster (9th e s t e o po i ” S g li e Cir. Sept. 8, 2000), pdf. It required copy-limiting circuitry in certain digital audio recorders (§ 1002) and, in exchange for protecting the making and selling of the devices and their noncommercial use to record music, § 1008, required manufacturers to pay royalties into a fund for ultimate distribution to copyright owners, §§ 1003-1007. The large record companies and movie studios can hardly complain of impediments to congressional consideration of their present arguments. They suffer from no collectiveaction problems such as often afflict the unorganized general

50 public, and they are established interests rather than newcomers. Even in the past decade, petitioners have secured congressional attention (MGM Br. 13 nn.11,12) and legislation (e.g., the just-described 1992 Act, the 1998 DMCA) on issues relating to copyright in the digital setting. Congressional consideration of their current arguments is actively under way. See Opp. 13-15. Congress is the proper forum to consider any legal change in the Sony rule, which requires affirmance in this case. CONCLUSION The judgment of the Ninth Circuit should be affirmed. Respectfully submitted.
MICHAEL H. PAGE MARK A. LEMLEY KEKER & VAN NEST, LLP 710 Sansome Street San Francisco, CA 94111 (415) 391-5400 Counsel for Grokster CHARLES S. BAKER PORTER & HEDGES, LLP 700 Louisiana Street, Ste. 3500 Houston, TX 77022 (713) 226-0676 Counsel for StreamCast MATTHEW A. NECO WENDY MILLAR GOODKIN STREAMCAST NETWORKS, INC. 20969 Ventura Blvd. Woodland Hills, CA 91364 (818) 887-8610 Counsel for StreamCast March 1, 2005 RICHARD G. TARANTO H. BARTOW FARR, III FARR & TARANTO 1220 19th Street, NW Washington, DC 20036 (202) 775-0184 Counsel for Respondents CINDY A. COHN Counsel of Record FRED VON LOHMANN ELECTRONIC FRONTIER FNDTN. 454 Shotwell Street San Francisco, CA 94110 (415) 436-9333 Counsel for StreamCast