Applying Textual Analysis to Study Historical Urban Planning Documents

Historical urban planning documents are primary sources that reveal how cities have been shaped by the decisions, ideologies, and constraints of their time. These texts—ranging from official zoning codes and comprehensive plans to policy reports, public hearing transcripts, and city council minutes—encode the priorities of past generations. Applying systematic textual analysis to these materials allows researchers, historians, and urban planners to move beyond surface-level reading and uncover deep patterns in language, rhetoric, and theme. This guide provides a comprehensive approach to conducting textual analysis on historical urban planning documents, covering methodological frameworks, practical steps, analytical tools, and real-world applications.

What Is Textual Analysis in an Urban Planning Context?

Textual analysis is a research method that examines written, spoken, or visual texts to interpret meaning, identify themes, and understand the contexts in which those texts were produced. In urban planning, documents are not neutral records; they are rhetorical artifacts that reflect political compromises, economic interests, and cultural values. For instance, a 1920s zoning ordinance might use language like “public health” and “order” while implicitly enforcing racial segregation. Textual analysis helps expose these subtexts.

Three primary traditions inform textual analysis in planning history:

  • Content analysis quantifies the presence of specific words, phrases, or concepts across a corpus, enabling statistical comparisons over time. Researchers can measure how often terms like “suburb,” “transit,” or “green space” appear in plans from different decades.
  • Discourse analysis focuses on how language constructs social reality, examining power relationships, framing, and narrative structures. It asks not just what words are used, but what assumptions and ideologies those words reveal.
  • Rhetorical analysis assesses persuasive strategies—ethos, pathos, logos—used by planners and policymakers to legitimize plans or marginalize alternatives. It uncovers how authority is claimed and how arguments are structured.

Each tradition offers a different lens, but they share a common goal: to extract meaning that would be invisible to a casual reader of historical documents.

Why Study Historical Urban Planning Documents with Textual Analysis?

The documents produced by planning agencies, municipal governments, and private developers are more than dry technical reports. They capture the spirit of an era: fears about overcrowding, faith in technological progress, or anxieties about social change. Applying textual analysis yields several benefits:

  • Reveals implicit assumptions: Language choices often betray unstated biases—for example, equating “blight” with certain ethnic neighborhoods or describing “slum clearance” as a public good.
  • Traces paradigm shifts: Analyzing terminology across decades can show how planning moved from City Beautiful ideals to modernist functionalism to contemporary smart growth.
  • Provides evidence for historical claims: Instead of relying on secondary interpretations, textual analysis offers direct evidence from primary sources, strengthening academic arguments.
  • Informs contemporary practice: Understanding past failures and successes through language helps today's planners avoid repeating harmful rhetoric or policies.
  • Uncovers marginalized voices: Even when official documents ignore opposition, textual analysis of meeting minutes or alternative plans can surface counter-narratives.

Methodological Frameworks in Depth

Content Analysis

Content analysis is the most quantitative tradition. Researchers define a set of categories or keywords and count their frequency across a corpus. For example, a study of comprehensive plans from 1960 to 2010 might track the use of “automobile,” “highway,” “pedestrian,” and “bicycle” to measure shifting transportation priorities. Simple frequency counts can reveal trends, but collocation analysis—examining which words appear together—adds context. If “highway” often co-occurs with “efficiency” while “pedestrian” co-occurs with “safety,” the document frames these modes differently.

Discourse Analysis

Discourse analysis is interpretive and qualitative. It examines how language shapes social reality and power relations. In planning, discourse analysis might explore how a city’s master plan constructs the identity of a neighborhood: a place of “blight” needing “renewal” versus a “historic district” requiring “preservation.” The choice of verbs and adjectives frames what actions are considered legitimate. Researchers also look at the use of passive voice (e.g., “residents were relocated” instead of “the city relocated residents”) to obscure agency or responsibility.

Rhetorical Analysis

Rhetorical analysis focuses on persuasion. Planners use ethos (credibility), pathos (emotional appeal), and logos (logical argument) to win support for their proposals. For example, a report advocating for a new freeway might cite traffic engineering data (logos), invoke the promise of economic growth (pathos), and reference the authority of a city planning commission (ethos). By breaking down these appeals, researchers can see how documents seek to shape public opinion and policy outcomes.

Core Steps in Conducting Textual Analysis on Planning Documents

Step 1: Define Research Questions and Scope

Begin with a clear question that textual evidence can answer. Examples include:

  • How did the rhetoric of “public participation” change in city planning documents between 1950 and 1990?
  • What metaphors dominated descriptions of highways in metropolitan transportation plans from 1940 to 1960?
  • How did urban renewal documents frame displacement as “redevelopment” or “improvement”?
  • What role did environmental language play in zoning ordinances after the 1970 Clean Air Act?

The scope should define the time period, geographic area, and types of documents (e.g., general plans, zoning codes, environmental impact statements, master plans). A well-defined scope prevents the corpus from becoming unmanageable and ensures that findings are meaningful.

Step 2: Collect and Digitize the Corpus

Gather a representative sample of documents. Many historical planning records are housed in municipal archives, university special collections, or online repositories such as the Library of Congress or local historical societies. Public records from city planning departments may also be available through Freedom of Information requests. The National Archives is another rich source for federal projects like urban renewal and interstate highways.

Once collected, convert documents into machine-readable text using optical character recognition (OCR) software. For handwritten or poorly scanned materials, manual transcription may be necessary. Tools like Tesseract OCR or Adobe Acrobat's OCR function can handle typed documents with reasonable accuracy. Always verify OCR output against original scans, as errors can skew frequency counts.

Step 3: Preprocess the Text

Clean the text to remove noise: headers, footers, page numbers, and scanned artifacts. Standard preprocessing steps include:

  • Lowercasing all text (if doing frequency analysis)
  • Removing punctuation and stop words (common words like “the,” “and,” “of”)
  • Stemming or lemmatizing words to merge variants (e.g., “planning,” “planner,” “planned”)

However, be cautious: if analyzing discursive shifts in tone or rhetoric, stop words may be important to preserve sentence structure. For example, modal verbs like “shall,” “must,” “should” convey degrees of obligation—critical for policy analysis. Similarly, conjunctions can reveal logical relationships. Tailor preprocessing to the research question.

Step 4: Choose Analytical Methods and Tools

Depending on the research question, select from qualitative and quantitative techniques:

  • Keyword frequency and collocation: Identify words that appear most often together (e.g., “public” + “safety,” “private” + “property”). Tools: AntConc, Voyant Tools, NVivo.
  • Coding and thematic analysis: Manually tag passages with thematic codes (e.g., “equity,” “efficiency,” “resistance”). This is essential for discourse analysis. Software like NVivo or MAXQDA facilitates coding across many documents.
  • Computational text mining: Use topic modeling (e.g., with Python's Gensim library) to automatically extract latent topics across large corpora. For an introduction, consult resources like The Programming Historian.
  • Sentiment analysis: Measure positive vs. negative language around specific issues, though historical language corpora may require custom training. General sentiment lexicons often misclassify historical usage (e.g., “awful” could mean “full of awe” in older texts).

Step 5: Conduct the Analysis

Apply the chosen methods systematically. For example, in a study of New York City's 1961 Zoning Resolution, a researcher might compute the frequency of terms like “floor area ratio,” “mixed use,” “bulk,” and “setback” and then trace how these terms evolved in subsequent amendments. Collocation analysis could reveal that “floor area ratio” often appeared with “as of right,” indicating a shift toward by-right development. For qualitative approaches, close reading of key passages is required. Identify where certain actors are described as “stakeholders” vs. “obstructionists,” or where information is presented as “data” vs. “concerns.” These linguistic choices shape the authority of the document.

Step 6: Interpret and Contextualize Findings

Textual analysis does not end with word counts; results must be interpreted within their historical and social context. A spike in the term “security” in 2000s transportation plans may reflect post-9/11 anxieties. A decline in “public housing” in favor of “affordable housing” in the 1980s mirrors political shifts away from direct government provision. Always triangulate findings with secondary historical sources, interviews, and demographic data. This prevents overinterpretation of textual patterns and grounds the analysis in empirical reality.

Case Studies in Textual Analysis

Urban Renewal Language

In the mid-20th century, U.S. cities published urban renewal plans that often described target areas as “blighted.” Textual analysis of documents from 1949 to 1974, conducted by researchers at the University of Pennsylvania, found that “blight” was used in 82% of renewal project descriptions. Collocation patterns showed it frequently paired with “clearance,” “elimination,” and “disease.” Such language dehumanized residents and framed demolition as a medical imperative. In contrast, terms like “displacement” or “community” were rarely used with emotional weight, revealing a technocratic mindset that prioritized physical redevelopment over social cohesion.

Framing Homelessness in Municipal Codes

A 2021 study analyzed 15 years of city council minutes and public ordinances in three U.S. cities. Using critical discourse analysis, the researchers found that early documents referred to homelessness as a “nuisance” or “criminal activity,” while later ones used terms like “chronic homelessness” and “service-resistant.” The shift in language, while superficially more clinical, still framed the issue as an individual pathology rather than a structural housing failure. Such analysis demonstrates how textual patterns can expose persistent ideological underpinnings even when wording changes.

The 1909 Plan of Chicago

Daniel Burnham's 1909 Plan of Chicago is a landmark document in American urban planning. A textual analysis of its language reveals heavy reliance on metaphors of the body: the city as an organism needing “circulation” (streets as arteries), “lungs” (parks providing fresh air), and “cleanliness” (slum clearance as surgery). The plan also uses grandiose language like “a city beautiful and dignified” to appeal to civic pride while downplaying the costs of implementation. By counting the frequency of such metaphors and comparing them to planning documents of the same era, researchers can see how the City Beautiful movement's rhetoric both inspired and obscured the plan's exclusionary aspects.

Challenges and Pitfalls to Avoid

Textual analysis of historical documents comes with specific difficulties:

  • Incomplete or biased archives: Documents from marginalized voices (e.g., community groups, indigenous populations) are often missing. The archival record favors official narratives. Researchers should acknowledge this limitation and, when possible, supplement with oral histories or community records.
  • Anachronistic word meanings: Terms like “blight” or “slum” had different connotations in earlier decades. Researchers must consult historical dictionaries or contextual sources, such as contemporaneous newspaper articles, to ensure accurate interpretation.
  • Over-reliance on quantitative methods: Word frequencies can mislead if not accompanied by qualitative reading. A word like “community” might be used often but always in patronizing or exclusionary contexts—only close reading reveals that.
  • Sample size and representativeness: A small, non-random sample may not reflect broader trends. Aim for at least 30–50 documents per time period for reliable pattern identification. When using digital repositories, document the search strategy and any filtering criteria.
  • Technical barriers: Poor OCR quality, handwriting, and non-English documents require specialized handling. For non-English planning documents, consider using translation tools cautiously and ideally work with native speakers.

Researchers can choose from a range of tools, from free online platforms to professional qualitative analysis software:

ToolTypeBest ForCost
Voyant ToolsWeb-basedQuick visualization and exploration of small to medium corporaFree
AntConcDesktopConcordance, collocation, keyword listsFree
NVivoDesktopQualitative coding, thematic analysis, mixed methodsPaid
Python (with NLTK, spaCy, Gensim)ProgrammaticTopic modeling, advanced NLP, large corporaFree
MAXQDADesktopMixed methods, includes OCR and transcriptionPaid

Note: Always check the copyright status of documents before digitizing and analyzing. Many historical public records are in the public domain, but be aware of restrictions on more recent materials (post-1928 in the U.S.). For materials under copyright, fair use provisions in academic research may apply, but consult institutional guidelines.

Integrating Textual Analysis into Urban Planning Education

Instructors teaching urban history or planning methods can incorporate textual analysis as a hands-on exercise. A semester project might involve students selecting a single planning document (e.g., a 1923 comprehensive plan for their city) and performing a close reading combined with keyword analysis. Students should present findings in a short paper that connects textual patterns to the city's subsequent development outcomes. Such assignments build critical thinking skills and demonstrate that planning is never purely technical—it is a realm of competing values expressed through language.

For advanced courses, pair textual analysis with GIS to map the spatial distribution of terms like “green space” or “industry” across a city's planning history. This interdisciplinary approach connects language to geography in powerful ways. Additionally, instructors can introduce students to collaborative coding platforms like Jupyter Notebooks to run basic text mining on public-domain documents from sources like the Internet Archive.

Future Directions: AI and Large Language Models

Emerging tools like large language models (LLMs) offer new possibilities for textual analysis. Researchers can use models to summarize long documents, generate thematic overviews, or even detect rhetorical patterns (e.g., how frequently plans use passive voice to obscure agency). However, caution is warranted: LLMs are trained on contemporary internet text and may misinterpret historical idioms or produce anachronistic readings. Human oversight remains essential, especially when interpreting qualitative findings.

Another frontier is the analysis of non-textual elements in planning documents—maps, charts, and photographs—alongside text. Multimodal textual analysis that integrates image captions, map legends, and diagram labels can yield a richer understanding of how planners communicated data and vision. For example, a plan's use of red shading to indicate “blighted” areas is a textual and visual rhetorical choice that warrants study.

Conclusion

Applying textual analysis to historical urban planning documents transforms static records into dynamic evidence of past mindsets, conflicts, and ambitions. By systematically examining language, researchers can expose hidden biases, trace the evolution of planning paradigms, and provide a more nuanced view of how cities came to be what they are today. Whether through computational methods or close reading, textual analysis equips students and professionals with the skills to question official narratives and uncover the values embedded in planning practice. As cities face new challenges—climate change, inequality, technological disruption—learning from the language of past planners offers a valuable lesson in humility and critical reflection. The words chosen yesterday continue to shape the built environment of tomorrow; understanding them is essential for building more just and informed urban futures.