Module 2. Elements of Structure: Entities, Relationships, Attributes
Course 3. Modeling Meaning: How We Structure Humanities Data
Estimated Time: 30 minutes
🧠Module Objectives
- Define and distinguish entities, relationships, and attributes.
- Describe how these components form the core of any data model.
- Recognize different kinds of relationships (one-to-one, one-to-many, many-to-many).
- Apply these concepts to real-world humanities examples.
- Explain how modeling choices express interpretation and priority.
The Grammar of Data
If data models are like languages, then entities, relationships, and attributes are their grammar.
- Entities → the nouns — the things that exist.
- Relationships → the verbs — the ways those things connect or act upon each other.
- Attributes → the adjectives/adverbs — the descriptive details that give each thing character.
Just as sentences have structure, so do models. A good model communicates clearly and consistently, showing both what is there and how it matters.
Entities: What Exists
An entity represents a type of thing or concept you want to describe. Each entity becomes a node (or table, depending on the database type).
| Example Entity | Description | Example Instances |
|---|---|---|
| Person | Individual creator, performer, or subject |
Jesse Welles, Dave Cobb, Charlie Kirk |
| Song | A musical work or recording |
"War Isn’t Murder", "Horses", "Charlie" |
| Album | A curated collection of songs |
Red Trees and White Trashes,* Devil's Den |
| Theme | Abstract ideas present in lyrics |
War, Forgiveness, Addiction |
| Event | Something that has happened, including performances |
Farm Aid 2025, Assassination of Charlie Kirk |
Entities are conceptual categories: they don't store the data yet, but define what kinds of data exist.
Attributes: What Makes Each Entity Unique
Attributes (or "properties") describe or qualify an entity. They add detail, allowing us to compare and filter.
| Entity | Example Attributes | Description |
|---|---|---|
| Song | title, releaseDate, duration, key, sentiment |
Basic properties and measurable aspects |
| Person | fullName, birthPlace, role, instrument, socialHandle |
Biographical or functional metadata |
| Theme | name, description, relatedConcepts |
Abstract descriptors |
| Event | date, location, audienceSize | Spatiotemporal context |
Attributes often come from interpretation, e.g., "sentiment = melancholy" or "theme = protest." Choosing which attributes to include expresses a model's values: what matters enough to record. It is worth noting that it is impossible to record every possible attribute about any given entity, so there must be a choice about which are significant enough to include.
Relationships: How Things Connect
Relationships link entities together: the lines between the dots. They define the logic of the model: who did what, when, and why.
| Example Relationship | From → To | Meaning |
|---|---|---|
| WROTE_SONG | Song → Person | Identifies the songwriter or composer |
| PERFORMED_AT | Song → Event | Connects a song to a specific performance |
| EXPRESSES | Song → Theme | Indicates that a work conveys a concept or emotion |
| INFLUENCED_BY | Person → Person | Captures creative or intellectual influence |
| BELONGS_TO | Song → Album | Groups works within a collection |
In humanities models, relationships often carry interpretive weight: "Expresses," "Reflects," "Challenges," or "References" are analytical as much as factual.
Cardinality: How Many?
Cardinality specifies how many of one thing may relate to another.
| Type | Example | Meaning |
|---|---|---|
| One-to-One (1:1) | A Person → a Birthplace | One person has one birthplace |
| One-to-Many (1:N) | An Album → multiple Songs | One album includes many songs |
| Many-to-Many (M:N) | A Song ↔ multiple Themes | Songs can express several themes, and themes appear in many songs |
Understanding cardinality ensures that your model accurately reflects complexity rather than oversimplifying it.
Putting It Together: A Mini-Model
Imagine we want to model a small part of the Wellespring world:
(Person)—[:WROTE_SONG]—>(Song)—[:RELEASED_ON]—>(Album)
(Song)—[:EXPRESSES]—>(Theme)
(Album)—[:EXPRESSES]—>(Theme)
Here's how the pieces fit:
- Entities: Person, Song, Album, Theme
- Attributes (examples):
Song.title,Song.releaseDate,Person.role - Relationships:
WROTE_,RELEASED_ON,EXPRESSES - Cardinality:
- One Person → Many Songs
- One Album → Many Songs
- Many Songs ↔ Many Themes
This compact structure already lets us ask meaningful questions, such as:
- Which themes appear most across Jesse Welles' songs?
- Which collaborators contribute to more than one album?
- Do certain themes cluster within specific albums or time periods?
Interpretation Through Structure
Choosing these entities and relationships is itself an interpretation. We could instead focus on audience reception, political context, or emotional tone: each choice yields a different version of the same cultural world.
A good model is not "objective"; it's transparent about what it represents and why.
Key Takeaways
- Entities, relationships, and attributes are the core elements of data structure.
- Together, they form a framework that makes connections explicit and discoverable.
- Cardinality ensures that quantities and directions of connection make sense.
- Every modeling decision reflects priorities, interpretations, and values.
Knowledge Check & Reflection
Suggested Readings & Resources
Practical Guides to Data Modeling
- Backadar, Jeff. "Introduction to MySQL with R." The Programming Historian 7 (2018).
- Specific to MySQL databases and the R programming language, but an excellent and concrete step-by-step tutorial for how to implement abstract concepts of entities, attributes, and relationships in a real database and humanities context.
- Simsion, Graeme, and Graham Witt. Data Modeling Essentials. Third edition. Elsevier Science, 2004.
- A foundational textbook that delves deeper into the art and science of identifying and defining the core components of a data model.
Entities, Relationships, and Attributes in the Humanities
- Drucker, Johanna. "Database Design." In The Digital Humanities Coursebook: An Introduction to Digital Methods for Research and Scholarship. Routledge, 2021.
- Skallerup Bessette, Lee, Katia Bowers, Maria Sachiko Cecire, et al. "The Data-Sitters Club." 2023.
- A lively and clear example of how humanities scholars decide on entities, attributes, and relationships for a much-loved (and, perhaps, despised by a different population segment) literary corpus.
Critical Perspectives on Categorization
- D’Ignazio, Catherine, and Lauren F. Klein. Data Feminism. The MIT Press, 2020.
- Chapter 4, "What Gets Counted Counts," offers a powerful analysis of how the choices that are made to model data is never neutral and always carries political and ethical weight.