Argumentari 🗣️

Argumentari is a lightweight, cross-platform Python package for constructing, serializing, manipulating, and visualizing structured argument data.
Build argument trees, apply formal reasoning schemes, and interactively analyze debate structures.

Features

• Define argument nodes (premise, conclusion) and build trees
• Manipulate arguments: add, move, clone, edit, prune and collapse nodes
• Visualize argument graphs interactively
• Register and use formal argumentation schemes (e.g., Authority, Causal)
• Find and replace nodes by text or scheme
• Serialize and deserialize argument trees
• Cross-platform: Linux, macOS, Windows
• Simple API for integration into your projects

Usage

Create and manipulate an argument tree

from argumentari.nodes import premise, conclusion
from argumentari.tree import argumentTree
from argumentari.manipulation import (
    addNode, removeNode, replaceNode, moveSubtree,
    editNode, cloneSubtree, findNodes, pruneNode, collapseSubtree
)

root = premise("Electric cars reduce emissions.")
child1 = premise("EVs produce no exhaust.")
child2 = premise("Fossil fuels are the main source of carbon emissions.")
concl = conclusion("Adopting EVs will decrease fossil fuel use.")

addNode(root, child1)
addNode(child1, child2)
addNode(child2, concl)

tree = argumentTree(root)

Manipulate argument nodes

# Add a premise
addNode(root, premise("Countries with high EV adoption show lower emissions growth."))

# Move a subtree
moveSubtree(tree, child1, concl)

# Edit a node's text
editNode(concl, new_text="Governments should invest in EVs.")

# Clone a subtree
clone = cloneSubtree(child1)
addNode(root, clone)

# Prune a node
pruneNode(child1)

# Collapse a subtree
collapseSubtree(child2)

Visualize your argument

from argumentari.visualize import treeToGraph, visualize

graph = treeToGraph(tree.root)
visualize(graph)

Apply argument schemes

from argumentari.schemes import getScheme, argumentScheme

root.setScheme("Argument from Authority", {"Authority": "Dr. Smith", "Statement": "EVs help climate"})

Serialize and deserialize argument trees

from argumentari.io import argumentSerializer

# Serialize to JSON
json_str = argumentSerializer.toJson(tree.root)

# Deserialize from JSON
new_root = argumentSerializer.fromJson(json_str)

Find nodes

matches = findNodes(tree, pattern="EVs")
for node in matches:
    print(node.text)

Contact

• GitHub Issues: Argumentari Issues
• Author: literal-gargoyle