Editorial Note
This citations document supports The Architect and the Agent (folio Nº IV, three parts). It is organized to surface the primary sources the paper’s accuracy depends on, rather than to survey adjacent literature exhaustively.
The three categories that carry most of the paper’s weight:
- The Realization Engine series: the Multiplier-Mirror framework (The Multiplier and the Mirror, folio Nº II, the parent work) and the first companion memorandum (Enquiry, folio Nº III). This paper depends on both; every equation cited in the paper is cited by its parent-paper number.
- The IASA Business Technology Architecture Body of Knowledge (BTABoK) primary sources, including the Manifesto and the specific engagement-model, competency-model, and career-path pages this paper instantiates the framework against. The paper’s accuracy audit in development traced each architectural claim to a specific file path in the BTABoK repository; those file paths are listed here.
- The engineering artifacts that operationalize the paper’s derivation: the ASAP platform design document, SpecChat (as upstream dependency), and the Global Corp exemplar (as bench and substance-preserved corpus).
The remaining categories (formal methods, mentoring and apprenticeship transmission, automation and deskilling, commons governance) provide targeted support for specific claims in the paper’s argument. Many of the adjacent literature threads the parent framework draws on (Cobb-Douglas production, labor-market polarization, Matthew effect, stochastic-parrots / mirror, jagged-frontier studies) are inherited through the parent paper and are catalogued in its citations companion. They are not repeated here unless this paper makes a specific load-bearing use of them distinct from the parent paper’s use.
1. The Realization Engine series
[1.0] Landi, D. A. Mirror as Cognitive Feedback
Loop. Folio Nº I. Realization Engine, 2026.
https://github.com/Realization-Engine/realization-engine.github.io.
Digest: The foundational inquiry the program
presupposes. Names the Mirror as a cognitive feedback loop: a coupled
system in which articulated cognition is reflected in inspectable
external form, the practitioner responds to the reflection, and the
iteration continues. Establishes the substance and presentation
channels, the asymmetry between them, and the studio-mirror /
Narcissus’s-pool regime distinction that The Multiplier and the
Mirror (Folio Nº II) formalizes downstream. This paper inherits the
Mirror’s foundational naming from Folio Nº I and instantiates the
framework Folio Nº II builds on top of it.
[1.1] Landi, D. A. The Multiplier, Mirror and
The Tipping Point. Folio Nº II. F-Star project, 2026.
https://github.com/Realization-Engine/fstar.
Digest: The parent framework. Establishes the general
model of human capability under LLM amplification: output as multiplier
times composite FORCE, the Mirror as a structured object with substance
and presentation channels, the layered structure of FORCE, the atrophy
equation, the tipping point and its hysteresis, the cohort
discontinuity, the F→M transfer, the coupled phase portrait, the
seven-loop cascade, the decision bottleneck, the erosion of competitive
moats, the sovereignty question, the four futures. This paper cites
parent-paper equations by number throughout; every equation reference
(Eq. 1, 2, 3, 4, 4a, 6, 7, 7a, 9, 10, 11, 11a-c, 12, 12a, 12b, 13, 14,
14a, 15a/b, 16, 16a, 17, 18, 19, 20, 21, 22, 23, 24, 24a, 25, 26, 26a,
27, 28, 29, 30, 31, 32, 34, 36) is sourced from this document.
[1.2] Landi, D. A. Enquiry Into Specification as
Meaningful Struggle. Folio Nº III. Spec Chat project, 2026.
https://github.com/Realization-Engine/spec-chat.
Digest: The first companion memorandum. Narrows inside
the Engagement Model of BTABoK to the specification medium, names
SpecChat as the authored-commitment artifact the medium requires,
identifies the developmental, medium, and practice questions, warns
against cosmetic relocation, and introduces the
amplification-architecture boundary this paper zooms out from. This
paper takes Enquiry’s “realization engine” framing as given and extends
the reach of the framework to Value, People, and Competency work.
[1.3] Landi, D. A. The Multiplier and the Mirror, Citations. Folio Nº II · a. F-Star project, 2026. Digest: The adjacent-literature catalogue for the parent framework. This paper inherits these citations for the framework-general claims (Cobb-Douglas production, labor-market polarization, Matthew effect, stochastic-parrots / mirror, jagged-frontier studies, tacit-knowledge research, Ostrom commons). They are not duplicated here.
[1.4] Landi, D. A. Notes for The Architect and
the Agent. Preparatory notes, 2026.
https://github.com/Realization-Engine/asap/blob/main/docs/notes/Notes-for-The-Architect-and-the-Agent.md.
Digest: The analytical groundwork for this paper.
Organizes the architecture-practice corollaries of the framework section
by section, derives the theoretical foundation for ASAP’s design, and
traces every load-bearing parent-framework concept into its operational
form for enterprise architecture under LLM amplification. Superseded as
the primary argumentative artifact by this paper; retained as
preparatory reference.
2. IASA BTABoK primary sources
BTABoK, the body of knowledge this paper instantiates the framework against. Citations give the repository file path where the paper’s primary-source claims can be verified.
[2.1] IASA Global. Business Technology
Architecture Body of Knowledge (BTABoK).
https://iasa-global.github.io/btabok/.
Digest: The authoritative source for all BTABoK claims
in the paper. Repository mirror referenced during development:
E:/Archive/GitHub/dlandi/btabok.
[2.2] IASA Global. BTABoK Manifesto.
pages/manifesto/manifesto.md. Digest:
Names the five competency pillars every architect shares (Business
Technology Strategy, Human Dynamics, Design, Quality Attributes, IT
Environment) and the five specializations BIISS (Business, Information,
Infrastructure, Software, Solution) layered atop them. This paper cites
the Manifesto as the primary authority on the pillar/specialization
distinction.
[2.3] IASA Global. Get Started with the
BTABoK. pages/top_menu/get_started_m.md. By Paul
Preiss. Digest: Enumerates the BTABoK’s working models:
Outcome, Operating, Value, People, Engagement, Competency, Maturity,
surrounded by the Architecture Practice article, the Structured Canvas
Approach, and the Topic Areas. This paper cites this page as the primary
source for the reach-note that ASAP Studio’s four-working-model coverage
is a selection from a larger set.
[2.4] IASA Global. Architecture Practice.
pages/engagement_model/architecture_practice.md.
Digest: Establishes the practice model for BT
architecture, grounded in professional-body framing parallel to
medicine, law, and accounting. Primary source for the
practice-architecture concept the paper builds on.
[2.5] IASA Global. Architecture Lifecycle.
pages/engagement_model/architecture_lifecycle.md.
Digest: Primary source for the Architecture Development
Life Cycle (ADLC) and its six stages: Innovation Cycle, Strategy,
Planning, Transformation, Utilize and Measure, Decommissioning. This
paper uses the primary-source stage names throughout.
[2.6] IASA Global. Governance (Engagement
Model). pages/engagement_model/governance_em.md.
Digest: Primary source for the governance-spectrum
framing, the directive that architects should be governed rather than do
the governing, and the “avoid the police state” caution. Load-bearing
for the authority-gradient derivation.
[2.7] IASA Global. Decisions.
pages/engagement_model/decisions.md.
Digest: Primary source for the architecturally
significant decision (ASD) concept, framed as “the stuff that matters”
per Martin Fowler. Supports the Decision Bottleneck argument and the
decision scribe subagent derivation.
[2.8] IASA Global. Principles.
pages/engagement_model/principles.md.
Digest: Primary source for the eight-to-twelve
memorable-principle rule and the properties of good principles
(constructive, compelling, memorable, testable, decisive, useful).
[2.9] IASA Global. Benefits Realization.
pages/engagement_model/benefits_realization.md.
Digest: Primary source for Rapid Value Management (RVM)
and its three indicator tiers: early validation indicators (within the
first 90 days, ideally 30), leading indicators (91 days to one year),
lagging indicators (end-success). Also the biweekly Value Map update
minimum and the monthly investment validation minimum. Load-bearing for
the cadence section of Part One and for the RVM subagent derivation.
[2.10] IASA Global. Investment Planning.
pages/engagement_model/investment_planning.md.
Digest: Primary source for the NABC (Needs, Approach,
Benefits, Considerations) business case format and for the “attack the
PMO” demand-shaping framing.
[2.11] IASA Global. Technical Debt.
pages/engagement_model/technical_debt.md.
Digest: Primary source for the Technical Debt Ratio
formula (Remediation plus Maintenance over Development, times 100) and
for the typology of technical debt (code, structural, QA, value, legacy,
organization). Note: tagged operating_model rather than
value_model in the primary source; ASAP’s operational
placement of tech-debt under Value is a practice-architecture choice
documented in Part Three.
[2.12] IASA Global. Community.
pages/engagement_model/community.md.
Digest: Primary source for the “Community Eats
Framework for Lunch” tenet and the “Specialization-Based Conflict
Destroys Practices” warning. Tagged people_model.
[2.13] IASA Global. Competency.
pages/engagement_model/competency.md.
Digest: Primary source for the five proficiency levels
(Awareness, Basic, Delivery, Experienced, Shaping), the Bloom-hierarchy
basis, the five pillars as the shared competency core, the BIISS
specializations as lateral depth, and the 42 total competency count
across pillars. This paper cites the level-label table (lines 46-52) as
authoritative for proficiency-level labels.
[2.14] IASA Global. Career.
pages/engagement_model/career.md. Digest:
Primary source for the six-level managed career path: Aspiring
(Unmeasured), Foundational, Associate/Advanced, Professional,
Distinguished, Chief. Establishes that the career path is distinct from
the four-tier CITA certification ladder and that Aspiring is
pre-certification while Chief is post-Distinguished, recognized through
demonstrated practice. Load-bearing for the Part One §5 treatment of
multi-form FORCE and the CITA-ladder asymmetry.
[2.15] IASA Global. Organization.
pages/engagement_model/organization.md.
Digest: Primary source for the claim that rotation is
“a critical component of architecture skill acquisition and shared value
delivery” (line 186). Supports the Loop-3 and Loop-7 treatments of
architect rotation.
[2.16] IASA Global. Mentoring (BTABoK
Overview). pages/btabok_overview/mentoring.md.
Digest: Primary source for the three mentoring stages
and their durations: Early Stage mentoring for Foundation and Associate
architects at three-to-six-month engagements within two-to-three-year
trajectories; Professional Mentoring for Professional and Distinguished
at one-to-three-year relationships (or longer); Topic Area mentoring at
one-to-three-month individual engagements. This paper cites the
primary-source durations in Part One §2 and §6 and in Part Two Loop
3.
[2.17] IASA Global. Adopting the Competency
Model.
pages/btabok_overview/adopting_the_competency_model.md. By
Paul Preiss. Digest: Primary source for the
five-levels-from-Bloom derivation, the four CITA certifications
(Foundation, Associate, Professional, Distinguished), the four-method
360-degree measurement (self, peer, mentor, certification, per line
139), and the five operative rules of the IASA Mentoring Method (lines
155-166), most load-bearing of which is that architects cannot progress
beyond critical milestones without reviewed outputs from one or more
qualified mentors.
[2.18] IASA Global. Culture.
pages/engagement_model/culture.md. Digest:
Primary source for the Westrum Culture Diagnostic and the Culture Map as
canvases used in architecture practice. Supports the
culture-diagnostic-runner subagent.
[2.19] IASA Global. Engagement.
pages/engagement_model/engagement.md.
Digest: Primary source for two principles this paper
cites: “the outcome of the work is more important than the method of
achieving that outcome” (line 54) and “Stop Doing Architecture, Start
Digitally Enhancing Your Customer” (principle heading, line 52).
[2.20] IASA Global. Structured Canvases.
pages/structured_canvases/structured_canvases_m.md.
Digest: Primary source for the over-75 canvas count and
the Structured Canvas Approach. Supports the canvas-library skill and
the renderer in Preview/Rendering.
[2.21] IASA Global. Topic Areas.
pages/topics/, including cloud.md,
ai_ml.md, security.md,
integration.md, dev_ops.md,
systems.md, agile.md,
sustainability/. Digest: Primary source
for the topic-areas-as-awareness-level framing: “starting points where
topic areas provide context for IT architects to be aware of” (e.g.,
topics/cloud.md line 14). Supports the topic-area skills
and the discipline that keeps them brief.
[2.22] IASA Global. Competency Model Index.
pages/competency_model/competency_model_m.html.
Digest: The index page that navigates the 42
competencies across the five pillars plus the additional competencies
per BIISS specialization. Note: the page’s navigational grouping
(listing pillar headers and specialization headers together in a flat
list) is a common source of confusion between pillars and
specializations; the Manifesto ([2.2]) is the authoritative
taxonomy.
[2.23] IASA Global. Design Decisions
(Overview).
pages/btabok_overview/design_decisions_3.md.
Digest: Primary source for the Decision Bias Calibrator
concept (“The BTABoK includes a set of decision bias calibrators to help
unearth biases found”, line 138). Supports the decision-scribe
subagent’s invocation of the calibrator.
3. The engineering artifacts
[3.1] Landi, D. A. Architect Support Agentic
Platform (ASAP) Design Document.
https://github.com/Realization-Engine/asap/blob/main/WIP/Architect-Support-Agentic-Platform.md.
Digest: The amplification architecture this paper
derives and the ASAP Studio implementation of it. Enumerates the ten
primitives, the authority gradient, the AI-suitability bands and seven
safeguard patterns, the asymmetric coverage across the four working
models ASAP Studio covers, the ten platform components, the six-wave
delivery plan with the product fork at end of Wave 4, the stop-lines,
and the evaluation criteria. Where this paper states “ASAP Studio does
X,” the specific mechanism is described in this document.
[3.2] Landi, D. A. SpecChat and SpecLang.
https://github.com/Realization-Engine/spec-chat.
Digest: The specification medium ASAP consumes as an
upstream dependency. SpecLang is the specification language; SpecChat is
the implementation including parser, validator, MCP server, canvas
rendering, and the BTABOK profile. This paper refers to SpecChat
throughout as the Engagement-Model inside view, and to ASAP as the
zoomed-out view that surrounds it.
[3.3] Landi, D. A. Global Corp. Enterprise
Architecture Exemplar (narrative).
https://github.com/Realization-Engine/asap/blob/main/docs/examples/global-corp/Global-Corp-Exemplar.md.
Digest: The worked enterprise against which ASAP
components are designed and tested, maintained as a long-form narrative
document in the ASAP repository. A fictional global supply-chain company
specified completely in BTABoK terms: strategic thesis, portfolio of
capabilities and value streams, decision register, waiver process, view
gallery, personas, ASRs, and governance bodies. Functions in this paper
as fixed-point bench, substance-channel test bench, and, over the long
arc, as part of the substance-preserved corpus.
[3.3a] Global Corp. SpecChat spec collection.
https://github.com/Realization-Engine/spec-chat/tree/main/docs/examples/global-corp/.
Digest: The machine-processable companion to [3.3]. A
full BTABOK-profile spec collection (architecture spec, application
specs, gate specs, manifest) that SpecChat validators exercise
end-to-end. The narrative at [3.3] is authoritative for prose; the spec
files are authoritative for cross-references, counts, and IDs.
[3.4] Landi, D. A. ASAP Acronym and Term
Glossary.
https://github.com/Realization-Engine/asap/blob/main/WIP/ASAP-Acronym-and-Term-Glossary.md.
Digest: Working glossary for ASAP-internal terms,
BTABoK-practice terms used in ASAP, and the agentic-platform primitive
vocabulary. The glossary is a working reference; the Manifesto
([2.2]) and competency.md
([2.13]) remain authoritative for BTABoK taxonomy
claims.
4. Formal specification and specification-driven realization
Supporting Future 4 (the return to specification) as the architectural form of the framework’s fourth future. The Enquiry catalogues this tradition more fully; here only the sources this paper cites directly are listed.
[4.1] Hoare, C. A. R. “An Axiomatic Basis for Computer Programming.” Communications of the ACM, 12(10), 1969, pp. 576-580, 583. DOI: 10.1145/363235.363259. Digest: Foundational statement that programming is precise reasoning about behavior, not merely manual construction. Supports this paper’s Future 4 framing of architect-as-author-of-commitments.
[4.2] Lamport, L. Specifying Systems: The TLA+ Language and Tools for Hardware and Software Engineers. Addison-Wesley, 2002. Digest: Treating specification as the primary engineering artifact; tying high-level intent to machine-checkable behavior. The architectural analog is the authored-commitment boundary this paper preserves at the spec layer and surrounds with agentic practice.
[4.3] Dijkstra, E. W. A Discipline of Programming. Prentice-Hall, 1976. Digest: Specification and reasoning as the discipline’s substance. Cited as precedent for the transition this paper describes from implementation-derived FORCE to specification-derived FORCE, with the warning that Dijkstra’s generation built specification FORCE on top of implementation FORCE, a sequencing whose post-LLM substitute is untested.
[4.4] Abrial, J.-R. Modeling in Event-B: System and Software Engineering. Cambridge University Press, 2010. Digest: Refinement as an explicit relation with proof obligations. Supports this paper’s treatment of ADLC stage transitions as refinement events with proof-of-progression obligations.
5. Architecture practice, mentoring, and apprenticeship transmission
Supporting Loop 3 (tacit-knowledge transmission collapse) and Loop 7 (cohort discontinuity) as architecture-specific dynamics, and the six-level career path as the profession’s developmental trajectory.
[5.1] American Institute of Architects (AIA).
Making Mentoring Work. 2019. Via Catalyst.org.
Digest: Cited by the IASA Mentoring Method
([2.17]) as the structural precedent for separating
managers from mentors. Supports this paper’s analysis of the mentoring
channel as the profession’s primary transmission mechanism for tacit
FORCE.
[5.2] Preiss, P. BTABoK Career Path and Managed
Profession Framing. In [2.14].
Digest: The managed career path as the profession’s
alternative to role-based employment, grounded in research from Iasa’s
focus groups, 2000-plus surveys, 1000-plus interviews, 200-plus
corporate assessments, and 4500-plus individual competency assessments.
Load-bearing for this paper’s claim that the CITA ladder’s calibration
depends on a development trajectory the LLM is absorbing.
[5.3] Preiss, P. “Better Is Better: Why Architects
Need to Stop Chasing Enterprise and Start Serving the World Around
Them.” LinkedIn post, 22 April 2026.
https://www.linkedin.com/posts/ppreiss_better-is-better-why-architects-need-to-activity-7452655010259181568-0oVr/.
Digest: Demand-side essay by BTABoK’s primary author on
the small-and-medium-business expertise gap and its implications for the
architecture profession’s growth segment. Cited in Part One’s
cohort-discontinuity section and in Part Two’s “The SMB-architect”
section as the empirical source for three figures: approximately 42% of
small businesses lacking the resources and expertise to successfully
deploy the technology they are already using (Goldman Sachs survey), 96%
planning new-technology adoption including AI (US Chamber of Commerce),
and approximately 98% of companies in markets like Canada qualifying as
small and medium. Also cited for the call to redirect the profession
toward the SMB demand side, where the architectural expertise gap is
open and broadening as those organizations adopt AI without
architectural support, and for the framing that the
small-architecture-firm and solo-contract-architect practice structures
are the structures organized to serve that demand at volume.
6. Automation, deskilling, and out-of-the-loop fragility
Supporting the atrophy equation’s architectural form and ASAP Studio’s categorical exclusions on developmental work.
[6.1] Bainbridge, L. “Ironies of Automation.” Automatica, 19(6), 1983, pp. 775-779. Digest: The foundational statement that automation can create new vulnerabilities by removing the conditions under which operator skill forms. Direct theoretical support for the atrophy dynamic The Multiplier and the Mirror formalizes in Eq. 11 and this paper instantiates for architecture.
[6.2] Endsley, M. R. “From Here to Autonomy: Lessons Learned from Human-Automation Research.” Human Factors, 59(1), 2017, pp. 5-27. Digest: Synthesizes decades of automation research on out-of-the-loop performance problems, loss of situation awareness, and the specific conditions under which automation degrades operator capability. Supports the audit-trail proposition as a structural response to out-of-the-loop fragility in architecture practice.
[6.3] Skitka, L. J., Mosier, K. L., & Burdick, M. “Does Automation Bias Decision-Making?” International Journal of Human-Computer Studies, 51(5), 1999, pp. 991-1006. Digest: Classical evidence for automation bias: users accept incorrect automated recommendations even when contrary information is available. Supports the Mirror’s failure dimensions (automation-bias risk specifically) as a structural property the AI-suitability bands must calibrate against.
7. Commons governance and the substance-preserved corpus
Supporting Part Two’s sovereignty section and Part Three’s Global Corp framing as demand-side commons intervention.
[7.1] Ostrom, E. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge University Press, 1990. Digest: The canonical framework for institutional design around common-pool resources. This paper’s demand-side FORCE-commons argument is an application of Ostrom’s analysis to the architecture profession’s training-signal substrate. The institutional-design question this paper leaves open (“what monitoring, norm, and incentive structures could preserve the architecture-practice FORCE commons across firms that compete on the surface but share the underlying training-signal substrate?”) is in Ostrom’s terms.
[7.2] Ostrom, E., Gardner, R., & Walker, J. Rules, Games, and Common-Pool Resources. University of Michigan Press, 1994. Digest: Extends [7.1] with formal game-theoretic analysis. Cited here because this paper’s supply-side / demand-side distinction in the FORCE-as-commons analysis maps cleanly onto the rule-classes Ostrom, Gardner, and Walker identify.
8. Professional models and the managed profession
Supporting this paper’s treatment of architecture as a profession with structural defenses that LLM amplification challenges.
[8.1] Abbott, A. The System of Professions: An Essay on the Division of Expert Labor. University of Chicago Press, 1988. Digest: Foundational sociology-of-professions analysis. Cited here because this paper’s claim that architecture’s institutional review channels (CITA boards, EARBs, mentor sign-off) are the profession’s structural defenses against the legibility crisis is an application of Abbott’s framework to LLM amplification specifically.
[8.2] Larson, M. S. The Rise of Professionalism: A Sociological Analysis. University of California Press, 1977. Digest: Foundational analysis of how professions establish jurisdictional authority. Supports this paper’s reading of BTABoK’s managed-career-path commitment as the architecture profession’s jurisdictional project.
9. Empirical-research framing
Supporting the research agenda in Part Three §21. The jagged-frontier and variance-amplification empirical sources are inherited from the parent-paper citations ([1.3] above, §4 of the parent-paper companion) and are not duplicated here.
[9.1] Dell’Acqua, F., McFowland, E., Mollick, E., et al. “Navigating the Jagged Technological Frontier.” Harvard Business School Working Paper No. 24-013, 2023. Digest: The single most important empirical source for the variable-multiplier and variance-amplification dynamics the research agenda proposes to test in the architecture domain. The jagged-frontier finding translates cleanly to architecture practice through the paper’s working-model decomposition.
[9.2] Peng, S., Kalliamvakou, E., Cihon, P., & Demirer, M. “The Impact of AI on Developer Productivity: Evidence from GitHub Copilot.” arXiv:2302.06590, 2023. Digest: The RCT methodology this paper uses (and which The Multiplier and the Mirror cites) is the prototype for the architecture-practice studies the research agenda proposes. Extending it to architectural tasks (decision authoring, governance review, principle-set revision) is an open empirical project ASAP Studio makes tractable.
10. Ancillary and background
[10.1] Iasa Global. CITA Certification Program
Information. https://iasaglobal.org/certification/.
Digest: The four CITA levels (Foundation, Associate,
Professional, Distinguished) and the board-review process. This paper
cites CITA as the profession’s certification structure and names it as
the empirical instrument through which cohort-level predictions become
testable.
[10.2] IASA Global. SpecChat-BTABOK
Implementation Plan.
https://github.com/Realization-Engine/spec-chat/blob/main/docs/design-notes/btabok/SpecChat-BTABOK-Implementation-Plan.md.
Digest: The staged implementation plan for the BTABoK
profile in SpecChat. Load-bearing for this paper’s six-wave delivery
plan, which places SpecChat Waves 1-4 before the ASAP Wave 5-6
zoom-out.
[10.3] IASA Global. Concept Definition Language (CoDL) and Canvas Definition Language (CaDL). BTABoK 3.2 education portal, April 2026. Digest: The CoDL/CaDL publications referenced in the Global Corp Exemplar as the canonical vocabulary for BTABoK concept instances and canvas definitions. This paper treats CoDL/CaDL as the upstream vocabulary ASAP consumes through SpecChat without further modification.
Verification note
All BTABoK primary-source citations (§2) were verified against the
repository mirror at E:/Archive/GitHub/dlandi/btabok during
the paper’s accuracy pass. Each file-path reference above can be checked
directly in the mirror or at the corresponding URL on
iasa-global.github.io/btabok. The Manifesto
([2.2]) is authoritative where the navigational grouping on
the Competency Model index ([2.22]) or on
specialization-area pages might suggest a different count of pillars or
specializations.