A.01TRANSMITTED / PARTIAL
Textual witness
Evidence manuscript · annotation · variant · layout
Method comparison · transcription · source criticism
What can the surviving witness support?
SOURCE · MODEL · INTERPRETATION
Evidence becomes meaningful through the method used to connect it.
Structures of Inference explores how manuscript scholars, numerical mathematicians, and historians build defensible conclusions from sources, models, approximations, context, and explicit limits.
○ Independent educational resource
AΔ
B.02APPROXIMATE / TESTABLE
Numerical model
Evidence equation · mesh · residual · error estimate
Method discretization · refinement · verification
How close is the approximation?
C.03CONTEXTUAL / CONTESTABLE
Historical argument
Evidence text · concept · context · reception
Method comparison · interpretation · historical criticism
Which conclusion fits the sources and their context?
Cross-disciplinary comparison of research practices — not a claim that textual interpretation, numerical approximation, and intellectual history use equivalent mechanisms.
EVIDENCE DOES NOT INTERPRET ITSELF
A source, a numerical result, or a historical argument becomes meaningful only when its method, assumptions, and limitations are made visible.
Sources have histories.
Approximations have errors.
Comparisons need context.
Conclusions need limits.
FOUR STUDY FIELDS
Different evidence requires different forms of control.
Manuscripts & Transmission
Explore medieval manuscripts, book production, glosses, palaeography, codicology, textual transmission, biblical exegesis, scholarly networks, and historical source criticism.
- Manuscripts
- Palaeography
- Textual transmission
- Medieval history
Digital Editing & Text
Study critical editions, textual variants, digital editing, automated text recognition, intertextuality, manuscript layout, text reuse, digital medieval studies, and computational humanities.
- Digital editions
- Text variants
- ATR
- Digital humanities
Numerical Approximation
Explore partial differential equations, finite elements, discontinuous Galerkin methods, mesh refinement, error estimation, stability, convergence, and scientific computing.
- PDEs
- Finite elements
- Error analysis
- Scientific computing
Comparative Thought
Examine Christian-Muslim intellectual encounters, Islamic theology, historical theology, philosophical argument, religious pluralism, textual traditions, reception, and historical comparison.
- Intellectual history
- Theology
- Christian-Muslim relations
- Interpretation
INFERENCE CASES
Three disciplines, three different routes from evidence to conclusion.
CASE 01 A
WITNESS → VARIANT → EDITION
Manuscript scholarship
How should a scholar reconstruct a text when surviving manuscripts do not agree?
Evidence may include
Variant readings, scribal corrections, annotations, manuscript dates, layout, linguistic features, transmission history.
Methodological controls
Source comparison, palaeography, codicology, textual criticism, historical context.
Limit A critical edition is a scholarly reconstruction and should not be presented as direct access to an unattested original text.
CASE 02 ≈
MODEL → DISCRETIZATION → ERROR
Numerical mathematics
How can a continuous mathematical problem be approximated reliably on a finite computational mesh?
Evidence may include
Residuals, mesh size, computed solutions, convergence behaviour, error indicators, stability properties.
Methodological controls
Finite-element formulation, a posteriori estimation, adaptive refinement, verification, convergence analysis.
Limit A numerical approximation is not automatically reliable merely because a computer produced a result.
CASE 03 ?
TEXT → CONTEXT → COMPARISON
Intellectual and religious history
How can texts from different religious traditions be compared without removing them from their historical contexts?
Evidence may include
Primary texts, theological arguments, historical vocabulary, translations, polemics, reception, manuscripts.
Methodological controls
Historical context, source criticism, language, intellectual history, careful comparison.
Limit Comparison should clarify difference as well as similarity and should not collapse distinct traditions into one system.
THE INFERENCE PROTOCOL
Seven checks between evidence and conclusion.
Define the question
State exactly what is being reconstructed, approximated, interpreted, or compared.
Identify the evidence
List the manuscripts, equations, computational outputs, historical texts, or contextual sources actually available.
Locate the gaps
Identify missing witnesses, unresolved variants, modelling assumptions, discretization limits, unavailable sources, or uncertain historical context.
State the method
Explain how the evidence is being transformed into an edition, approximation, historical interpretation, or comparative argument.
Check error and alternatives
Ask which other reconstruction, mesh, model, reading, or historical interpretation could also fit the evidence.
Verify the result
Use independent witnesses, convergence tests, error estimates, additional sources, or alternative interpretations where appropriate.
Limit the claim
State what the evidence supports and what remains uncertain, approximate, hypothetical, or context-dependent.
EDUCATIONAL REFERENCE POINTS
Six scholars across manuscripts, numerical mathematics, and comparative intellectual history.
These profiles are presented as educational reference points for exploring public academic work. They are not presented as members, employees, partners, collaborators, representatives, endorsers, or affiliates of Structures of Inference.
Platform contact note
The first three email addresses are platform contact addresses supplied for this site and are not presented as verified university or institutional email accounts.
CGTEXT / A
Cinzia Grifoni
Postdoctoral University Assistant · Austria
University of Vienna
Faculty of Historical and Cultural Studies
Department of History
Institute of Austrian Historical Research
Academic research in European history in late antiquity and the early Middle Ages, medieval Latin, manuscript studies, codicology, palaeography, scholarly networks, Carolingian intellectual culture, biblical exegesis, textual criticism, critical editions, digital editing techniques, and automated text recognition of medieval sources.
Medieval manuscripts · Textual criticism · Carolingian scholarship · Digital editing
ORCID 0000-0002-0560-6031
Platform contactcinzia.grifoni@cititrends.org
NUMERICS / ΔSC
Scott Congreve
Assistant Professor · Czech Republic
Charles University
Faculty of Mathematics and Physics
Department of Numerical Mathematics
Academic research in numerical methods for partial differential equations, finite-element and discontinuous Galerkin methods, nonlinear PDEs, the Helmholtz equation, two-grid techniques, a posteriori error analysis, hp-adaptive mesh refinement, polytopic meshes, virtual element methods, domain decomposition, and scientific computing.
Finite elements · PDEs · Error estimation · Adaptive methods
ORCID 0000-0002-7314-3120
Platform contactscott.congreve@cititrends.org
RLHISTORY / ?
Ralph Leo
Assistant Professor of History · United Arab Emirates
United Arab Emirates University
Department of Tourism and Heritage
Academic research at the historical, theological, and philosophical interface of Christian and Islamic thought from late antiquity through the early modern period, including comparative religious thought, doctrinal development, theological argument, intellectual history, and historical encounters between Christian and Muslim traditions.
Christian-Muslim relations · Religious thought · Historical theology · Intellectual history
ORCID 0000-0001-5299-1828
Platform contactralph.leo@cititrends.org
TEXT / BGV
Gabriel Viehhauser
Professor of Digital Editing · Austria
University of Vienna
Faculty of Philological and Cultural Studies
Department of European and Comparative Literature and Language Studies
Academic research in digital editing and digital humanities, including critical electronic editions, medieval literary traditions, textual variance, text reuse, intertextuality, manuscript transmission, digital narratology, computational text analysis, digital medieval studies, and methods for representing complex textual traditions in digital form.
Digital editing · Textual variance · Digital medieval studies · Intertextuality
ORCID 0000-0001-6372-0337
Educational reference pointECNUMERICS / R
Erin Carson
Associate Professor · Czech Republic
Charles University
Faculty of Mathematics and Physics
Department of Numerical Mathematics
Academic research at the intersection of numerical linear algebra, high-performance computing, and parallel algorithms, including Krylov subspace methods, iterative refinement, mixed-precision computation, numerical stability, communication-avoiding algorithms, least-squares problems, Gram-Schmidt methods, and reliable large-scale numerical computation.
Numerical linear algebra · Mixed precision · Iterative methods · High-performance computing
ORCID 0000-0001-9469-7467
Educational reference pointHISTORY / CLD
Lejla Demiri
Professor of Islamic Doctrine · Germany
University of Tübingen
Center for Islamic Theology
Chair of Islamic Doctrine
Academic research in systematic Islamic theology, kalam, religious pluralism, Christian-Muslim theological interactions, Arabic and Ottoman Turkish textual sources, Ottoman intellectual history, Islamic manuscripts, scriptural interpretation, and the history of theological exchange between religious traditions.
Islamic theology · Christian-Muslim relations · Religious pluralism · Intellectual history
ORCID 0000-0001-8217-9953
Educational reference pointREFERENCE STATUS
Academic reference does not imply participation.
Structures of Inference is an independent educational prototype. Academic names and institutional references are included solely to help readers discover relevant areas of public scholarship.
The first three platform contact addresses were supplied specifically for this site. They are not presented as verified personal, university, institutional, or employer-provided email accounts.
The remaining profiles are educational reference points only and are not presented as participants in, contributors to, endorsers of, or affiliates of this resource.
RESEARCH FILES
Open a file and inspect the route from evidence to conclusion.
Explore concise educational notes across manuscript scholarship, digital editing, numerical approximation, scientific computing, religious history, and critical inference.
10 files
Manuscript Studies / A.01Why can two manuscripts preserve different versions of the same text?
Explore copying, correction, transmission, and variation in manuscript cultures.
Consider scribal copying, corrections, omissions, additions, glosses, exemplars, contamination, manuscript families, physical damage, linguistic change, local scholarly practice, transmission history, palaeography, and codicology. Textual variation is historical evidence rather than simply an error to remove.
Textual Criticism / A.02What does a critical edition reconstruct?
Explore the difference between surviving witnesses and an editor's scholarly reconstruction.
Consider manuscript witnesses, variant readings, stemmatic relationships, editorial judgement, emendation, apparatus criticus, uncertainty, historical context, language, scribal intervention, competing readings, digital editions, and why a critical text should make editorial decisions visible.
Digital Editing / B.03What changes when a manuscript tradition becomes a digital edition?
Explore encoding, representation, interfaces, and the choices behind digital scholarly editions.
Consider transcription, markup, textual variants, facsimiles, metadata, search, automated text recognition, manuscript layout, uncertainty, linked data, text reuse, user interfaces, editorial models, interoperability, and why digitization does not eliminate the interpretive decisions of editors.
Finite Elements / Δ.04Why do numerical methods replace a continuous problem with a finite one?
Explore discretization as the foundation of finite-element approximation.
Consider partial differential equations, function spaces, meshes, basis functions, weak formulations, degrees of freedom, discretization, approximation spaces, boundary conditions, computational solutions, convergence, and why a finite-dimensional solution must be interpreted in relation to the original continuous problem.
Error Estimation / Δ.05How can a numerical method know where its approximation is weak?
Explore residuals, error indicators, and adaptive refinement.
Consider a posteriori error estimation, residuals, local error indicators, mesh refinement, adaptive algorithms, reliability, efficiency, computational cost, hp-refinement, solution features, convergence, and why adaptive methods concentrate computational effort where the approximation requires it most.
Numerical Stability / R.06Why can mathematically equivalent algorithms behave differently on a computer?
Explore finite precision, rounding, stability, and computational implementation.
Consider floating-point arithmetic, rounding error, conditioning, backward error, numerical stability, iterative refinement, mixed-precision computation, Krylov methods, communication costs, algorithmic variants, loss of orthogonality, and why mathematical equivalence does not guarantee identical finite-precision behaviour.
Religious History / C.07How should historians read theological disagreement?
Explore doctrine, polemic, language, context, and historical purpose.
Consider theological arguments, polemical texts, doctrinal vocabulary, intended audiences, intellectual networks, translation, historical circumstances, authority, philosophical concepts, religious identity, manuscript transmission, and why historical disagreement should be reconstructed before it is evaluated from a modern perspective.
Christian-Muslim Relations / C.08What makes historical comparison between religious traditions difficult?
Explore similarity, difference, terminology, and contextual asymmetry.
Consider comparative categories, Christian and Muslim theological vocabulary, scriptural reasoning, polemics, translation, historical periods, intellectual exchange, internal diversity, religious pluralism, manuscript sources, reception, and why comparison should avoid treating either tradition as internally uniform.
Source Criticism / ?.09When is silence in a source meaningful?
Explore the risks of drawing conclusions from what historical evidence does not mention.
Consider preservation bias, genre, authorial purpose, audience, missing manuscripts, archival survival, censorship, conventions, implicit knowledge, argument from silence, corroborating sources, historical context, and why absence of evidence and evidence of absence are not automatically equivalent.
Inference / V.10What makes an approximation or interpretation defensible?
Explore transparency, verification, alternatives, error, and explicit limits.
Consider evidence quality, assumptions, methodological transparency, uncertainty, alternative interpretations, numerical error, historical context, model validation, source comparison, reproducibility, verification, disciplinary standards, and why a strong conclusion should explain not only what is claimed but also what could make that claim fail.
No research files match your search.
ABOUT STRUCTURES OF INFERENCE
Reliable conclusions depend on making the route from evidence visible.
Structures of Inference is an independent educational prototype connecting manuscript scholarship, numerical mathematics, and comparative intellectual history.
It does not suggest that textual reconstruction, computational approximation, and historical interpretation are equivalent procedures.
Instead, it explores a shared scholarly responsibility: identifying evidence, locating gaps, stating methods, checking errors and alternatives, verifying results where possible, and limiting conclusions to what the available evidence can support.
Structures of Inference is not a university, religious institution, research laboratory, mathematical institute, archive, publisher, professional association, consultancy, technology company, or commercial service.
Evidence has a history
A manuscript, text, dataset, or computational result reaches the researcher through a specific process of production and transmission.
Methods transform evidence
Editing, discretization, modelling, and historical interpretation all introduce explicit methodological choices.
Error should remain visible
Uncertainty, textual variation, approximation error, missing sources, and competing interpretations should not be hidden.
Claims need boundaries
Responsible scholarship distinguishes what is supported from what remains approximate, hypothetical, or context-dependent.
TRACE THE REASONING
Take one conclusion and work backward to the evidence beneath it.
Browse research files, compare inference cases, and use the Inference Protocol to inspect sources, assumptions, alternatives, verification, and limits.