Michael Pietsch to DFW, December 22, 1994 (second-read letter + cut-suggestion list)

TL;DR

The actual editorial letter the composition legend orbits: Pietsch's December 22, 1994 second-read response to the full manuscript, on Little, Brown letterhead — two pages of reaction plus a four-page list of suggested cuts keyed to manuscript pages 496–1603 — in scans of Wallace's own HRC archive copy, carrying his working marginalia in four inks. The letter frames every cut as a question ("chapters and scenes that maybe can come out without killing the patient"), and cross-checking the list against the print digest shows most flagged scenes SURVIVED: the MAS*H story, the mattress scene, Idris blindfolded, the Ennet-House viewer night, Ewell's money-stealers confession, the lobster memory, the boys in the tunnel, the Luria–Tine puppet romance, and the grave-dig dream (which Pietsch describes as "digging w/Joelle in grave" and asks to be made "more explicit" — print keeps it, with Joelle in it, and keeps it oblique). The verified executions are surgical: the entire jamais-vu thread (marginalia: "Eliminate all jamais vu"; zero print occurrences), Lyle's wise-old-bee fable, the hairy-woman myth, the vegetable-woman eavesdropping scene, the Roberval story, and the duck-pond draining — replaced by exactly the engineer-kidnapping solution Pietsch proposed (print pp. 620–626). This is ij-cuts's primary document, and it also witnesses, from December 1994: the Orin/Avril question legible on the page ("Orin's among those who slept with Avril?"), Luria P—'s designed shadowiness, the moved-objects (wraith) thread Pietsch couldn't decode on two reads, the multi-explanation Hal-breakdown design, and the notes-at-the-end capitulation ("You're right that notes should go at the end.").

Key claims

• The cut list is scene-surgery phrased as questions, not a 300-page demand. 46 numbered entries over ms pp. 496–1603, most ending in "?" or "Cut." on single scenes/threads; the covering rationale is reader-facing: "I'm still hoping there are ways to make the novel much shorter, not because any one piece of it isn't wonderful but because the longer it is the more people will find excuses not to read it." [p2] The tone is explicitly non-lethal: "chapters and scenes that maybe can come out without killing the patient." [p2]
• Verified EXECUTED cuts (absent from the printed novel, verified against its full text): the jamais-vu thread (typed entries at 533 and 683; red margin note "J46 - 50 Eliminate all jamais vu"; "jamais" occurs zero times in the print fulltext); "643 Lyle's story of wise old bee: cut." (zero print trace; matches Moore's cut inventory); the "myth of hairy woman" from the 764 entry (no print occurrence); "808 Ennett House talk with vegetable woman listening. Cut." (green "Yes"; matches Moore's cut catatonic-eavesdropper item); the Roberval story (524 entry; print retains only a passing name in the Poutrincourt-seminar list, pp. 307–312); and the duck-pond draining (1028) — no such scene in print, and Pietsch's proposed replacement ("just say later that AFR found where Joelle was by kidnapping and interviewing her engineer") IS the published solution: the AFR captures the WYYY student engineer at print pp. 620–626.
• Verified REFUSED / survived (present in the printed novel, verified against its full text): Steeply's father's MAS*H story (764 "Cut … story of Steeply's father's MASH addiction." → print pp. 638–648, whole scene); "818 James's father with mattress. Cut." (green "Yes" — yet print pp. 491–503); "944 Idris Arslanian, blindfolded. Cut." (green "NO" — print pp. 567–574); "1071 Green, Gompert, Day watching cartridges. Cut." (green "NO" — print pp. 648–651, viewer on, Green asleep); Lyle's advice to LaMont Chu (648 entry red-circled — reads as engagement, not acceptance — print pp. 386–390); Hal's splintering-teeth dream (722 — print pp. ~449–450); Ewell's money-stealers club (1354 — print pp. 809–827, "Money-Stealers’ Club"); the lobster memory (1423 — print pp. 883–896, the Beverly lobster-traps memory); the boys in the tunnel (1118 — print pp. 666–673); the Orin–Luria hotel thread (940/956 — print pp. 565–567, 574–575, 596–601); Luria's "love thing with Tine" in Mario's film (683 continuation — print pp. 410–418); the Steeply/Bain letters (1094 — print pp. 663–665 + n.269); Marathe's Pat M. interview (1261 — print pp. 729–755); Molly Notkin's interview (1321 — print pp. 787–795); Joelle interviewed by Steeply (1553 — print pp. 934, 938–941); Matty Pemulis (1149 — print pp. 682–686); the Fackelmann ending (1603 — print pp. 972–981); Green's parents story (957 — print pp. 575–589, the Mauna Loa material). Net: the December list's grain confirms the scope-preserving-cuts verdict from the witness side, in primary documentation.
• The marginalia are a four-ink working system, plausibly a session diary. Green = response layer ("Yes" ×4, "NO" ×2, "True", underlinings of Pietsch's asks, the Orin-calling chronology note); red = tracking layer (checkmarks down the list, circles on contested entries, instruction notes like "Mention cartridge", session dates 1/27, 2/11, 2/16); blue = a distinct pass carrying J-prefixed tokens ("J23 p. 721", "J36/37"(?), "J46") and the date 1/18; black = grouping ovals/lines. The dates read as Jan–Feb 1995 revision sessions — "likely" — i.e., the months between this letter and Wallace's known 16-page February 1995 reply. ~73 marks total (inventory in the transcription file).
• A green "Yes" did not mean execution. The mattress scene (818) is marked "Yes" and survives in print; 940's "Yes" sits beside an entry offering two opposite remedies. This refines the Debraski-note claim (its second point: Wallace circled Yes on suggestions that nonetheless remain in the published book): the assents are written (not circled), and at least one demonstrably did not stick. (The mark-force key from Max — circle = acquiescence, dash = bared teeth — describes the February reply's code, not necessarily these inks.)
• Endnote apparatus, primary-verified: "You're right that notes should go at the end." + "Maybe we'll package it with a bookmark to help readers keep their place." + "I still hope you'll look for many, many notes you can cut. I've suggested xing more of them but will go into that another time." — this letter IS the December-1994 notes-at-the-end capitulation (with bookmark idea) that dtmax-bio ch.6 reports secondhand.
• Lost apparatus witnessed: "I'm also still confused by the placement of the smiley logo and the meaning of the Gaudeamus Igitur headings (and nobody I've polled here recognizes it, except distantly)." — MS-1994 evidently carried a smiley graphic and Gaudeamus Igitur headings; neither survives in the print text as digested (no smiley logo; no such headings). A genuinely new archival datum (recorded in the working notes).
• The physical-manuscript adversarialism, corroborated in passing: "reading the tiny type of the first printout left me so dog tired that my brain wasn't always running" [p1] — the 8-point-type length-disguise (diff-ms-physical-state) as experienced by its victim.
• Dating/context: sent two days before Christmas ("Sorry to be dumping this back just at your Christmas break."), i.e., between the October 1994 ending-objection letter and the February 1995 reply; Wallace's address is 1214 North Fell Ave., Bloomington.

Theories & evidence

The typed layer is Pietsch's; where a block reads theory-force into marginalia, it is flagged as an interpretive reading of the handwritten layer, not letter text. Locators: [p1]– [p6] = letter/scan pages; bare numbers = the cut list's ms-page entries; print page cites are digest-verified.

THEORY: The December 1994 cut list was scope-preserving scene-surgery, and most of what it targeted survived — the primary document behind the cut-count folklore

• Crux: ij-cuts
• Stance: advances
• Evidence:
  • [draft] The list itself: 46 entries over ms 496–1603, phrased as questions and single-scene "Cut."s; rationale "the longer it is the more people will find excuses not to read it" [p2]; "chapters and scenes that maybe can come out without killing the patient" [p2].
  • [text] Survival cross-check (this digestion, print digest): ~19 flagged scenes verified surviving vs. 6 verified executions (ledger in Key claims). The refused list includes the items Pietsch pressed hardest on (duck pond aside).
  • [draft] The green Yes/NO layer: "Yes" at 808 (executed) but also at 818 (NOT executed — print pp. 491–503); "NO" at 944 and 1071 (both held). Assent marks did not bind.
  • [author] Context rider: Wallace's own tour-era "cut by 400 pages or so" (speak-1996, PRINT) is the author origin of the big-cut myth; this document shows what the editing round actually looked like at the scene grain — suggestion lists measured in scenes, not chunks of 400 pages. (MS-page accounting may still reconcile the figures; see the ij-cuts ledger's recorded arithmetic gap.)
• Notes: Complements, from the primary side, Moore's MS-1993 collation and Pietsch's retrospective. The strongest single datum for the "popularity/authority inversion" already recorded at ij-cuts: the editor's actual maximum ask was per-scene surgery.

THEORY: The grave-dig dream — described by the editor as "digging w/Joelle in grave" — existed by December 1994, and its obliqueness survived a direct request for explicitness

• Crux: joi-head (context for ending-missing-year, gately-fate)
• Stance: mentions
• Evidence:
  • [draft] "1520 make concluding dream--digging w/Joelle in grave--more explicit?" [p6]
  • [text] Print keeps the dream (p. 934: Gately, "a very sad kid", the head that is "like Continental-Emergency important", Joelle van D. appearing winged) and keeps it oblique — the dead guy is never named there. The request for "more explicit" was not taken.
• Notes: Independent third-party attestation that Joelle is IN the MS-1994 dream — the same trio-adjacent grouping the theory record's Swartz reconstruction assembles for the missing-year dig (Hal's p. 17 version supplies Gately and Wayne; Gately's p. 934 version supplies the sad kid and Joelle). Also a clean stable-by-design datum: the joi-head evidence base was pressure-tested by the editor and kept oblique on purpose. Pietsch calls it the "concluding dream" at ms 1520 of 1603 — position matches the print placement (p. 934 of 981).

THEORY: The Orin/Avril question was legible to the editor in 1994 — "Orin's among those who slept with Avril?" — and Orin's calls read as approach-to-the-family-secret

• Crux: avril-orin (context for orin-coded-call)
• Stance: entertains
• Evidence:
  • [draft] "1321 Molly Notkin's interview: Orin's among those who slept with Avril?" [p6] — an editor with no access to fan exegesis read the incest question off the manuscript's Notkin interview (print pp. 787–795).
  • [draft] "1180 Who is Avril calling? Orin?" [p5] — the Avril-phone mystery flagged as its own entry (black-ink oval around it).
  • [draft] Pietsch's reading of Orin's calls to Hal: "Orin is calling because he's beginning to try to deal with it." [p2] — "it" = the family knot he glosses as "why dad drank, why mom fooled around, why mom is agoraphobic, why dad killed himself" [p2].
  • [draft]+ruling: green margin chronology note beside that paragraph: "O started calling again ~Sept" / "about a year ago" — Wallace working out the call chronology in his own hand ("about a year ago" confirmed; "~Sept" plausible).
• Notes: The 1321 entry is the earliest dated external witness of the Orin/Avril reading in the record — December 1994, pre-publication. The green chronology note bears on the n.110 call-dating arguments filed under orin-coded-call (call onset ~a year before the frame) — margin layer, interpretive register, cite with the ruling stamp.

THEORY: Luria P—'s shadowiness is authorial design — the editor asked for her twice and print kept her the "enigmatic Québecois fatale" known publicly by initial only

• Crux: avril-afr (the Luria-identity sub-question); context for wayne-afr
• Stance: mentions
• Evidence:
  • [draft] "Their whole supporting cast--Fourier, Broullime, Plourd, Luria--remains indistinct." [p1]
  • [draft] "More on who Luria P is? She remains shadowy throughout." [683 continuation, p3]; the 940 entry's fork: "Can we cut this thread and just leave it that only Steeply has gotten to him? Or spell out a little more clearly, what Luria is doing there." [p4]
  • [draft] "1404 Luria P and Broullime going to interview Orin?" [p6]
  • [text] Print outcome: the thread survives (pp. 565–567, 574–575, 596–601), the Luria–Tine "love thing" in Mario's film survives (pp. 410–418, the "enigmatic Québecois fatale known publicly only as ‘Luria P———,’" per the print text), and Luria is NOT spelled out — both of Pietsch's remedies declined; the shadow is the design.
• Notes: Composition-side support for treating Luria's indistinctness as engineered rather than accidental — relevant background when the avril-afr ledger weighs arguments from Luria's under-specification (including the Luria=Avril contradiction record). The ms name is already "Luria P" in Pietsch's typing: the initial-only surname predates the editing round.

THEORY: An informed 1994 reader guessed Hal's breakdown as "withdrawal, DMZ, or maybe something connected to the whole weird things-move-around aspect" — the multiple-explanation design was already operative in MS-1994

• Crux: hal-condition (also hal-dmz)
• Stance: entertains
• Evidence:
  • [draft] "I finished the book guessing how he got from here to there--withdrawal, DMZ, or maybe something connected to the whole weird things-move-around aspect of the story that I still don't really understand." [p1]
  • [draft] "The revelation that Hal's known all along of his mom's many affairs seems like a key to it all, a part of the puzzle that includes his father's alcoholism and suicide, that Hal's been avoiding with the help of pot." [p1] — previously known to this archive only via dtmax-bio ch.6's quotation; now primary-verified byte-exact.
  • [draft] The continuation: Hal's state persists "because Hal's still avoiding it" [p2] — Pietsch's psychological (avoidance/repression) reading standing alongside his chemical candidates.
• Notes: Cite-as-written variance worth preserving: Pietsch's third candidate is the moved-objects (wraith) mechanism, where Max's account of the YG frame's engineered ambiguity lists DMZ / detox / having seen the Entertainment (diff-yg-opening). The editor's triple and the biographer's triple overlap on two of three — evidence that the ambiguity was real on the page, and unsettled even for its first professional reader.

THEORY: The samizdat's provenance was already obscure to its editor — who licensed the obscurity ("mostly okay") — and the Antitoi fix was "Mention cartridge"

• Crux: designed-holes
• Stance: entertains
• Evidence:
  • [draft] "But I still don't get how the tape appeared in the first place, and why in Canadian hands." "Is this intentional on James's part? Is it somehow delayed revenge against the medical attache?" "Have I missed something or is it meant to be obscure?" [p2]
  • [draft] The license: "This aspect of the novel remains shaggy, which is mostly okay but any clarity you can add will help." [p2]
  • [draft] "754 Story about P-terminals is good, but Fortier, read-only samizdat, double-triple-agent talk, is incomprehensible." [p4]; "1216 Get explanation of how AFR traced samizdat to Antitoi into earlier scene?" with "Emphasize more the new plan, to look for those associated with the maker." [p5]
  • [draft]+reading: red boxed margin note beside 786 (the Antitoi-murder entry asking to clarify "what the Antitois unknowingly have"): "Mention cartridge" — moderately confident reading, unruled — the minimal-disclosure fix in Wallace's working hand.
  • [draft]+reading: blue margin note beside 1216/1222 (how'd the engineer know Joelle was in the hospital): "Engineer clue." — reading uncertain (see transcription); print's engineer-capture scene (pp. 620–626) is the executed mechanism.
• Notes: For macguffin-design (proposed, TODO inventory, ijbc-ep04): the editor-side record shows the MacGuffin's dissemination chain was never fully specified even at MS-1994, and that Wallace's fixes were minimal mentions rather than explanations — the hole is load-bearing design, pressure-tested. Also documents an editor-entertained dissemination reading (JOI-intentional / delayed revenge on the attaché) that print never resolves.

THEORY: The moved-objects (wraith) thread registered as a systematic pattern to a reader with no wraith key — and stayed undecodable to him through two reads

• Crux: wraith-reality
• Stance: mentions
• Evidence:
  • [draft] "the whole weird things-move-around aspect of the story that I still don't really understand" [p1].
  • [draft] "1118 Boys in tunnel. Can this come out whole? There are lots of references to moved objects, and this ends up being just one more. I know how you feel about tunnels, though..." [p5] — the editor counting moved-objects references as a set.
  • [text] The tunnel scene survives whole (print pp. 666–673); the moved-objects set survives.
• Notes: Pairs with diff-stice-wraith: MS-1993 named the Wraith outright in Stice's nickname and the rename moved the mechanism into subtext — this letter shows the subtext holding (the editor sees the pattern, not the agent). "I know how you feel about tunnels" is also a small [author]-adjacent datum: Wallace's tunnel-attachment was known to his editor.

THEORY: The manuscript's late stretch had a first-person "I" its editor couldn't place — "who's the I who speaks at the end?"

• Crux: narrator-proximity (the Hal-as-narrator strand)
• Stance: entertains
• Evidence:
  • [draft] "1149 Matty Pemulis's story. It's hard to take in another new character so far in the book. Could this come sooner? Also, who's the I who speaks at the end?" [p5]
• Notes: Second dated Pietsch-witness of unassigned first-person surfacings in the MS (his 1993 letter's "an 'I' who may be the one trying to put everything together" is already in diff-hal-narrator-ms-witness). Which "end" (of the Matty Pemulis chapter, most plausibly — print pp. 682–686) is ambiguous in the entry; do not over-assign.

THEORY: Notes-at-the-end was settled in this letter — with the bookmark idea attached

• Crux: annular-structure (apparatus context; diff-endnote-system)
• Stance: endorses
• Evidence:
  • [draft] "You're right that notes should go at the end. A novel with lots of long notes at the bottoms of pages looks much more academic and daunting than one with a section at the end. Maybe we'll package it with a bookmark to help readers keep their place." [p2]
  • [draft] "I still hope you'll look for many, many notes you can cut. I've suggested xing more of them but will go into that another time." [p2]
• Notes: Primary verification of dtmax-bio ch.6's December-1994 endnote-capitulation account, bookmark and all. The "many, many notes you can cut" pressure is the environment in which roughly half the endnotes were nonetheless added (diff-endnotes-half-added) — the refuge held.

Themes

Composition history (the December 1994 editing round at scene grain) · the cut/refusal ledger as design evidence (what survived pressure is stable-by-design) · engineered obscurity witnessed pre-publication (samizdat provenance, Luria, moved objects, Hal's breakdown) · the Orin/Avril question as a 1994 editor-reading · the endnote apparatus settlement · Wallace's four-ink working method and Jan–Feb 1995 revision sessions.

Cross-references

• [the manuscript-differences survey](../the manuscript-differences survey) — this letter is the primary document behind the record's negotiation account.
• debraski-ij-cuts — the survey that pointed at these scans (its circled-Yes-yet-remaining description is refined by the mark-level record here); its open acquisition follow-up — archive the six Pietsch-letter pages — is now DONE.
• infinite-summer-pietsch-editing-ij — Pietsch's 2009 retrospective of the same round (the "my canines are bared" marginalia, ~250 MS pages); this is the contemporaneous artifact.
• moore-first-draft-ij — MS-1993 collation; its cut inventory (wise old bee, catatonic eavesdropper) matches this list's executed items.
• dtmax-every-love-story — ch.6 quotes this letter's Hal's-mom's-affairs sentence and reports the endnote/bookmark capitulation; both now primary-verified.
• speak-1996 — the author-origin "cut by 400 pages or so" rider on ij-cuts; this document is the scene-grain reality that figure compresses.

Notable passages

All typed-layer, byte-gated against the transcription. [pN] = scan page.

• "I finished rereading Infinite Jest last night and wanted to tell you again how thrilled I am to be working with you to publish this." [p1]
• "err on the side of elaborating or stressing some points that are now made clearly but briefly" [p1]
• "Their whole supporting cast--Fourier, Broullime, Plourd, Luria--remains indistinct." [p1]
• "withdrawal, DMZ, or maybe something connected to the whole weird things-move-around aspect of the story that I still don't really understand" [p1]
• "The revelation that Hal's known all along of his mom's many affairs seems like a key to it all, a part of the puzzle that includes his father's alcoholism and suicide, that Hal's been avoiding with the help of pot." [p1]
• "Orin is calling because he's beginning to try to deal with it." [p2]
• "The ending of the novel, the horror of Gately's hitting bottom, is gorgeous and very very powerfully sad." [p2]
• "But I still don't get how the tape appeared in the first place, and why in Canadian hands." [p2]
• "Have I missed something or is it meant to be obscure?" [p2]
• "This aspect of the novel remains shaggy, which is mostly okay but any clarity you can add will help." [p2]
• "I'm also still confused by the placement of the smiley logo and the meaning of the Gaudeamus Igitur headings (and nobody I've polled here recognizes it, except distantly)." [p2]
• "On the attached pages I've suggested chapters and scenes that maybe can come out without killing the patient." [p2]
• "You're right that notes should go at the end." [p2]
• "Sorry to be dumping this back just at your Christmas break." [p2]
• "643 Lyle's story of wise old bee: cut." [p3]
• "1028 The draining of the duck pond is for me the hardest scene in the novel to keep reading." [p4]
• "1180 Who is Avril calling? Orin?" [p5]
• "1321 Molly Notkin's interview: Orin's among those who slept with Avril?" [p6]
• "1520 make concluding dream--digging w/Joelle in grave--more explicit?" [p6]
• "1553 Joelle interviewed by Steeply. If movie had nothing to do with James's suicide, what does she think did?" [p6]
• Margin (red, p6, interpretive): "J46 - 50 Eliminate all jamais vu" — executed to zero print occurrences.