Unconventional methods
Walking, drift, fiction and practice, and how they earned legitimacy.
Each tradition covers part. Hartman's critical fabulation, and Rosner's use of it in design, tie fiction to a thin archive and say what is invented. Smith's mythogeography and the dérive line (Debord, Sinclair, Careri) tie walking to a place and to openly limited myth, but offer no procedure for being wrong. Perec's three-day list is a Trace that stays flat. On rigour and legitimacy the closest comparator is autoethnography: the 2006 exchange between Anderson and Ellis and Bochner already contains the argument a conventional reviewer will put to Holagram. Law, Lury and Wakeford, and Back and Puwar give the licence for mess and invention. Tracy, Pierce and Lawhon, and Borgdorff give the vocabulary for reporting it. The threats are specific. Bleecker treats fact and fiction as indistinguishable; Kirby shows that labelled fiction still persuades; Flavell shows that fictocriticism guarantees no rigour; decolonial critics of speculative design ask whose epistemologies the fiction runs on; and a 2026 article argues that AI-written stories only simulate fiction, which bears on the synthetic Alter. Coverage of 2020 to 2026 is thinner than intended because the shared web-search budget ran out part-way.
Sources
- Guy Debord (English translation by Ken Knabb) (1956). Theory of the Dérive. Les Lèvres Nues, no. 9 (Nov 1956); reprinted in Internationale Situationniste, no. 2 (1958)not re-checkedopen access
The founding protocol of walking as research: the walker sets aside the usual motives for moving, reads the terrain for what Debord calls psychogeographical effects, and reports. It is the ancestor of Holagram's fixed fieldwork block and of Sellars' narrator.
- Phil Smith (writing as CrabMan) (2010). Mythogeography: A Guide to Walking Sideways. Triarchy Presscheckedmust read
The nearest practitioner precedent for Reading and Fiction as Holagram wants them. It joins a walk narrative to a handbook of drifting, holds several readings of one place at once, and uses 'limited myth': accounts that can show patterns of power or culture while being openly questionable or fictional. A reviewer in the Journal of Cultural Geography called it at once a handbook, a manifesto and a parody of both. Its group walk is also a model for the outside reader. Unconventional item.
Read first: The second half, the 'handbook of drifting', as the publisher page describes it.
- Iain Sinclair (2002). London Orbital. Granta Books (later Penguin)not re-checked
A book-length walk around the M25 motorway written as literary non-fiction, where the walked ground and written sources run together and the narrator is the instrument. For Holagram it is a planning object (a ring road) read on foot as a sign system, the same terrain as Ballard's Crash, and a case of authorship as persona. Unconventional item.
- Francesco Careri (2002). Walkscapes: Walking as an Aesthetic Practice (Italian edition: Walkscapes. Camminare come pratica estetica). Gustavo Gili; Italian edition Einaudinot found
An architect's history of walking as an aesthetic and spatial practice, written for designers and urbanists. It is the route into the dérive tradition from inside architecture and planning, and it exists in Italian, which a planning-school reviewer may already know.
Audit: Check the imprint pages: Gustavo Gili 2002 (Spanish/English) versus Einaudi (Italian, mid-2000s, from memory). Do not give one year for both.
- Joseph Pierce and Mary Lawhon (2015). Walking as Method: Toward Methodological Forthrightness and Comparability in Urban Geographical Research. The Professional Geographer 67(4): 655-662checked
Argues that urban researchers walk constantly but rarely report it, and that reporting the walking in the methods section adds rigour and lets studies be compared. This is the conventional-reviewer language for Holagram's fixed fieldwork blocks: say where, when, how long, what was looked at and what was recorded. The keywords include qualitative validity. It gives no help with fiction. One walkable place over three weeks is thin for the comparability they ask for, so the first study should claim a case, not a pattern.
Read first: The closing proposal for comparable walking practices (per the abstract).
- Georges Perec (translated by Marc Lowenthal) (2010). An Attempt at Exhausting a Place in Paris. Wakefield Press (English translation)checked
A three-day sitting in Place Saint-Sulpice in October 1974, logging what passes: people, buses, pigeons, a wedding, a funeral, signs. It is the model for a Trace register that lists and does not interpret, and for the infraordinary as a research object. The publisher notes that the flat cataloguing gradually takes on rhythm and a slightly eerie quality, which is a warning that selection is already a reading. Seated, fixed-time observation is also a lower-risk alternative to long drifting. Unconventional item.
- Shannon Mattern (2015). Deep Mapping the Media City. University of Minnesota Press (Forerunners: Ideas First)not re-checked
Holagram's linked entries plus failure log fit that description of a deep map, written by an urbanist for planners and designers.
- John Law (2004). After Method: Mess in Social Science Research. Routledgecheckedmust read
The licence for mess. Law argues that methods do not only describe social realities but help make them, so they are political, and that many realities are vague or fleeting and need methods that can cope. This backs Holagram's claim that spatial facts resist totalising description and gives a reason for registers that disagree with each other. It also attacks Holagram: a tidy set of nested levels and four ordered registers is the kind of clarity Law warns about, and the book's own table of contents has chapters on multiple worlds and ontological politics that the first study should answer.
Read first: Chapter 1 (Introduction) and chapter 6 (Non-Conventional Forms), from the table of contents on the Routledge page.
- Celia Lury and Nina Wakeford (eds.) (2012). Inventive Methods: The Happening of the Social. Routledgechecked
Treats methods as devices (the list, the pattern, the anecdote, the photograph, the probe, the tape recorder) that are inventive when they let the social happen instead of fixing it. A Holagram Entry is a device in this sense, and the chapters on lists, anecdotes, probes and speculation map onto Trace, Reading, Fiction and Test. The volume is also a worked example of how a mainstream series presents non-standard methods to social scientists. It does not give criteria for when an inventive method has failed.
Read first: The introduction, 'A Perpetual Inventory', then the chapters 'Anecdote' (Mike Michael) and 'Speculation' (Luciana Parisi).
- Les Back and Nirmal Puwar (eds.) (2012). Live Methods (The Sociological Review Monograph 60, supplement 1); see also 'Live Methods Revisited', The Sociological Review 73(5), 2025. Wiley-Blackwell / The Sociological Reviewchecked
A sociological manifesto for an artful, crafty practice, in eleven provocations; a library summary lists experimentation, attentiveness, collaboration, scavenging, mapping, topologies and 'idiotic methods' among the volume's themes. It fits Trace-as-scavenging and a method that is remade each time. The 2025 section (introduction by Paton and Jackson, open access) asks what became of the programme and concentrates on the politics of methods under institutional pressure, which is a useful check on how long a style like Holagram survives in a department.
Read first: The opening 'manifesto for live methods' (eleven provocations); for the 2025 revisit, the open-access introduction at https://eprints.lse.ac.uk/128280/
- Henk Borgdorff (2012). The Conflict of the Faculties: Perspectives on Artistic Research and Academia. Leiden University Press (LUP Dissertations)checkedopen access
An open-access doctoral thesis on what artistic research is and where it sits in academia, with the argument that what is said about artistic research helps make the practice. For a practice-based PhD at EKA this is the standard reference for justifying non-standard work to disciplines with other criteria. It does not supply tests for a particular method; it supplies the case that the academy has to make room.
Read first: The Introduction (chapter titles are listed by number only on the repository page).
- Natalie Loveless (2019). How to Make Art at the End of the World: A Manifesto for Research-Creation. Duke University Presschecked
A manifesto for research-creation, the Canadian name for practice-led research in the humanities. Loveless argues that it can produce scholarship in its own right and change teaching and research, and she gives the question of disciplines a chapter. It is a manifesto, so it argues for legitimacy more than it sets criteria.
Read first: Introduction ('Art in the Expanded Field') and chapter 2 ('Discipline(s)').
- Sarah E. Truman (2021). Feminist Speculations and the Practice of Research-Creation: Writing Pedagogies and Intertextual Affects. Routledgechecked
What to read it for: how writing that moves between theory and fiction is taught, assessed and attributed.
- Saidiya Hartman (2008). Venus in Two Acts. Small Axe 12(2)checkedmust read
The clearest model of Trace plus Fiction. Hartman starts from a thin archival record, the 1792 trial of a slave-ship captain, says what the record cannot give, and builds narrative from the gaps while keeping her own act of narrating in view and trying not to repeat the violence of the record. That matches Holagram's rule that Fiction answers to a sourced Trace and is labelled, and adds a discipline: the fiction states what the document leaves out. The ethical frame belongs to archives of slavery; carrying it to a planning document is an analogy, not a transfer, and it should be used with that said.
- Daniela K. Rosner (2018). Critical Fabulations: Reworking the Methods and Margins of Design. MIT Presschecked
Takes critical fabulation into the design of computing: telling stories that bring up alternative histories of technology, with a case on the women who wove memory wires for Apollo-era computers, where craft work on the ground and documents about computing meet. It shows how a scholar in a conventional design field presents fabulation to colleagues who expect evidence. For Holagram it is the bridge from Hartman's archive to computational systems, and to synthetic agents in the loop. The review linked here (Tecnoscienza, 2020) is the open way in.
- Julian Bleecker (2009). Design Fiction: A Short Essay on Design, Science, Fact and Fiction. Near Future Laboratory (self-published essay)not re-checkedopen access
Defines design fiction as making things that tell stories about near futures, and says that fact and fiction are 'productively indistinguishable'. That is the opposite of Holagram's separation of Fiction from evidence, so it is the first thing a design-fiction reader will say to you.
- David A. Kirby (2010). The Future Is Now: Diegetic Prototypes and the Role of Popular Films in Generating Real-World Technological Development. Social Studies of Science 40(1): 41-70checked
Shows how fictional technologies in film act as 'diegetic prototypes': they persuade audiences that a technology is needed and workable and so help produce it. This is the strongest argument against the rule that Fiction is never evidence: a labelled fiction still circulates and persuades. It supports a circulation rule for Alter's Fiction (see ideas). It is about films and the technology industry, not about field research, so the transfer to a planning study is by analogy.
- A Parede (collective; credited on the page as 'by A Parede') (2017). Cheat Sheet for a Non- (or Less-) Colonialist Speculative Design. Exhibist (web)checkedopen access
A practitioner critique, not an academic article. It says speculative and critical design has neglected race, class and gender privilege, and that adding non-Western perspectives is not enough: designers have to examine the power structures that keep some ways of knowing at the centre. It bears on Alter, whose Fiction comes from a narrow corpus, and on who walks the place and whose reading counts. A companion open-access paper, Turner and Taboada (Pivot 2021, QUT ePrints), argues that alternatives are often made from the same epistemologies as the dominant path and tries role-playing and improvised world-building as a remedy. Only the introduction of the cheat sheet came through; read the rest before citing. Unconventional item.
- Ruth Levitas (2013). Utopia as Method: The Imaginary Reconstitution of Society. Palgrave Macmillanchecked
A sociologist's case for treating imaginative reconstruction as a method in a conventional discipline, written for social scientists. Holagram's Fiction is not utopian, but the legitimacy move is the same: the imaginary is a way of doing analysis, and the book shows how that is argued to sceptics. From memory the book separates archaeological, architectural and ontological modes; check this before citing.
- Bartosz Lutostański (2026). Poza fikcjonalnością. Symulacja i narracja w generowanych przez AI opowiadaniach o polskich wykładowcach akademickich. Forum Poetyki 1(43): 40-55 (DOI 10.14746/fp.2026.43.3)checkedopen access
In English: 'Beyond fictionality: simulation and narrative in AI-generated stories about Polish academic lecturers'. It argues, from a corpus of short stories written by Claude, that such texts imitate the formal features of fiction without the rhetorical function that makes fiction fiction, and asks for a hermeneutics of simulated narrative. Holagram lets synthetic agents write the Fiction register. If this reading holds, a 'fiction' label on machine-written text is not enough: the register holds simulations, and the method must say what the human hand does to them. The corpus is one model and one genre, so it is an argument to meet, not a finding.
- Jack (Judith) Halberstam (2011). The Queer Art of Failure. Duke University Pressnot re-checked
Treats failure as a way of refusing success-defined measures of knowledge, and works from 'low theory' and eccentric archives that include popular culture. It supports keeping a failure log as an object worth writing from, and gives a vocabulary that is not apologetic. Caution: Halberstam's failure is political and cultural, a refusal of norms, while Holagram's is epistemic, a reading that did not survive its Test. A reviewer may take 'failure log' as rhetoric unless the log reports counts and criteria.
- Leon Anderson; with commentaries by Paul Atkinson, Kathy Charmaz, DeLysa Burnier, Norman Denzin, and Carolyn Ellis and Arthur Bochner; and Anderson's reply (2006). Analytic Autoethnography (with commentaries and reply, including Ellis and Bochner, 'Analyzing Analytic Autoethnography: An Autopsy'). Journal of Contemporary Ethnography 35(4)checkedmust read
The nearest conventional argument about a researcher who is also the instrument. Anderson asks autoethnography to be analytic and warns against self-absorption; Ellis and Bochner reply that this puts an unruly form under the control of analysis. Holagram is the same quarrel with a split answer: Trace and Test are analytic, Reading and Fiction are evocative, and the rule that no reading is tested by its author is the guard Anderson says evocative work lacks. A reviewer will ask which camp an Entry belongs to; the answer has to be given register by register. The exchange is also a worked example of how a field defends a first-person method.
Read first: Anderson's article (pp. 373-395), then Ellis and Bochner's 'Autopsy' (pp. 429-449). For an open-access overview, Montagud Mayor (2016), FQS 17(3): https://qualitative-research.net/index.php/fqs/article/view/2432
- Sarah J. Tracy (2010). Qualitative Quality: Eight 'Big-Tent' Criteria for Excellent Qualitative Research. Qualitative Inquiry 16(10): 837-851checkedmust read
A reviewer's checklist for qualitative work: worthy topic, rich rigour, sincerity, credibility, resonance, significant contribution, ethics and meaningful coherence. It can carry the conventional write-up: Trace maps to rich rigour, Test to credibility, the failure log to sincerity, the care rules to ethics, the diagram to meaningful coherence. A reviewer then finds the usual headings even though the method is unusual. It is written for qualitative inquiry and does not mention fiction.
- Helen Flavell (2009). Who Killed Jeanne Randolph? King, Muecke or 'ficto-criticism'. Outskirts 20 (May 2009), University of Western Australiachecked
The tradition Sellars sits near. Flavell traces the term 'ficto-criticism': popularised in Australia by Muecke and King in 1991, but already used in Canada by Jeanne Randolph and earlier in an interview with Jameson. Her argument is that the form is not automatically ethical or anti-authoritarian, and that each text has to be judged on whether it gives up the writer's mastery. For Holagram the lesson is that a licence to write fiction inside criticism does not itself produce rigour; the safeguards have to be built. The page also records that the form has gained some institutional recognition (courses, postgraduate research).
Ideas to take
Give every Fiction a gap line: what the Trace leaves out, and what the Fiction supplies.
Changes: The Fiction register gains a mandatory first sentence naming the gap in the Trace, and a closing line naming what is invented. It borrows Hartman's open statement of what the record cannot give and Smith's 'limited myth', and it makes the 150 words auditable by the outside reader.
Risk: Hartman's method answers the violence of archives of slavery. Used on a planning document it can read as borrowed gravity, and a gap line can become a ritual sentence. It may also push the Fiction toward explanation and away from the strangeness that makes it useful.
Write the conventional account under Tracy's eight headings, and report each walking block in the way Pierce and Lawhon ask.
Changes: The log produces a methods card per block (place, hours, who walked, what was recorded) and an appendix mapped to the eight criteria. The conventional write-up is then a translation job with a fixed target, and the thesis can say in its own method chapter that it does this on purpose.
Risk: Translation into criteria can make the unconventional parts vanish, or become box-ticking. Tracy's 'sincerity' also invites performed confession. Borgdorff's argument is that the academy should adapt, so keep one place where the method is explained in its own terms.
Design the fieldwork block from three precedents: list-only Trace sittings (Perec), a fixed drift with a companion (Debord, Smith), and a written reading held back until afterwards.
Changes: Splits the Trace stage into two modes. Seated, timed listing in one place gathers material with no reading allowed. Walked blocks are done with a second person and followed by a break before any Reading is written. Debord's actual rules should be checked in the text and borrowed where they fit the care rules.
Risk: Listing is not neutral: Perec's text turns eerie under its own flatness, so selection is already a reading.
State, register by register, which parts are analytic and which are evocative, and cite the Anderson and Ellis-Bochner exchange as the argument you are answering.
Changes: Adds a short table to the concept text: Trace and Test follow analytic standards, Reading and Fiction follow evocative ones, and authorship rules differ accordingly. It turns the hardest reviewer question (is this research or literature?) into a structural answer.
Risk: The method then has two audiences and a reviewer may judge it by its weaker half. Flavell's point that unconventional form guarantees nothing means the evocative half needs its own accountability, not just a label.
Treat circulation of Fiction as an intervention, and label machine-written Fiction as simulated narrative that a human has accepted.
Changes: Adds two rules: a Fiction never leaves the diagram (no workshop, no public document) without a log entry saying who saw it, because Kirby shows fiction acts; and every Alter text carries a marker naming the human who read it and what they changed, because Lutostański questions whether a machine's story is fiction at all.
Risk: This restricts uses of Alter that a platform may want, and the wording of the marker depends on a disputed claim from a single 2026 article. Bleecker's view that fact and fiction are indistinguishable means some readers will reject the separation however it is labelled.
Describe the diagram and the failure log as a deep map: provisional, not authoritative, and report the counts (held, broke, unchecked) with the criteria used.
Changes: Publishing counts with criteria stops 'failure log' from reading as rhetoric.
Risk: Counts invite gaming and performance metrics, which is the opposite of Halberstam's refusal of success measures. A provisional deep map is also an easy target for a reviewer who wants a stable finding.
Counter-arguments
- Holagram's answer is the Test register, but the Fiction register itself is exposed to this objection and the first study should show a Fiction that changed a Test.
- Anderson's complaint about self-absorption maps onto the Reading register: strong, even paranoid readings by the same person who walked and wrote are the failure Sellars' narrator shows. The only guard in the concept (no reading tested by its author) is untried; the literature in this thread gives no empirical case in which such a guard worked.
- Bleecker holds that fact and fiction are productively indistinguishable, and Kirby shows labelled fiction persuades and shapes real development. If either is right, a rule that Fiction is never evidence governs the page but not its effects.
- Debord conceded that psychogeography was vague, and the commentary on Sinclair (Hay, reported on Wikipedia) says the tradition became a literary mode without its politics. Holagram risks the same drift: a style that is admired and a method that is not used.
- Flavell argues that a fictocritical form is not automatically ethical or anti-authoritarian. Borrowing the licence of the Australian tradition does not supply rigour.
- Law's After Method challenges tidy structure: nested levels from body to territory and four ordered registers may be the totalising description that the concept says spatial facts resist.
- Decolonial critics of speculative design (A Parede 2017; Turner and Taboada 2021) say that alternatives made inside the dominant epistemology reproduce it. A synthetic Alter trained on a narrow corpus, and a single researcher choosing the walkable place, are open to that charge.
- Lutostański argues that AI-written stories simulate fiction without its rhetorical function. If so, the synthetic hand in the Fiction register produces something that is not fiction in the sense the concept relies on.
- Hartman's method is ethically bound to the people the archive silences. Writing fiction about real, named places and public servants, even from public documents, needs an ethics argument of its own; 'public material only' is not that argument.
- Pierce and Lawhon show that walking is under-reported and so hard to compare. One place and three weeks yields a case; a reviewer will not accept it as evidence of a pattern.
Gaps
- The shared web-search budget ran out after 22 searches, so the 2020 to 2026 literature is under-covered. Recent items found: Lutostański (2026), the Live Methods Revisited section (2025), Truman (2021) and Turner and Taboada (2021).
- Unverified leads left out of the 25, to be checked by someone with search budget: Haraway, 'SF: Science Fiction, Speculative Fabulation, String Figures, So Far' (Ada, 2013); Manning and Massumi, Thought in the Act (Minnesota, 2014); Ingold and Vergunst, Ways of Walking (Ashgate, 2008, seen at catalogue level only); Springgay and Truman, Walking Methodologies in a More-than-Human World (Routledge, 2018, seen at catalogue level only); Muecke's fictocritical essays; Back and Puwar's volume in full.
- Nothing found on care rules for walking research: stop-calls, an outside reader, switching reading modes on and off. This part of Holagram may be original in this thread.
- No source combines a formal Test with fiction and walking. The nearest are Anderson (analytic standards for first-person work) and Pierce and Lawhon (reporting walking); none asks what would show a reading wrong.
- Failure as method beyond Halberstam: nothing verified. Lisa Le Feuvre's Failure (Whitechapel, Documents of Contemporary Art) is recalled only and not listed.
- The Journal for Artistic Research and its Research Catalogue, which publish peer-reviewed practice work, and the EKA doctoral requirements were not checked; both matter for how the thesis will be examined.
- Content of Sinclair, Careri, Smith, Rosner, Loveless, Truman and Levitas was not read. Their relevance notes rest on publisher copy, tables of contents and reviews, and should be checked against the texts.