---
name: Pro-Con Tree
status: draft
description: DECISION family. The two-sided argument tree — the proposal at the root, branching into weighted arguments for and against, the visual form of Franklin's moral algebra for a single yes/no or this-vs-that choice.
sources:
  - title: "Janis, Irving L. & Mann, Leon (1977), Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment, Free Press"
    url: https://openlibrary.org/works/OL1282010W
  - title: Keeney, Ralph L. & Raiffa, Howard (1976), Decisions with Multiple Objectives, Wiley
    url: https://openlibrary.org/works/OL4107460W
---

# Pro-Con Tree

## Why it matters

A pro-con tree is the oldest decision aid there is, drawn as a tree: the proposal sits at the root, and two sets of branches grow out of it — the **pros**, the arguments for, and the **cons**, the arguments against — each of which can branch again into sub-considerations and rebuttals. Its whole job is to force the case *both* ways onto the page at once, so a decision isn't made by whichever side argued last or loudest, but by laying the full argument for beside the full argument against and seeing which few considerations actually carry the weight.

For example: a small team is deciding whether to take on a big new client. The room is excited — the revenue is real, the logo is impressive — and the conversation is all upside. Drawn as a pro-con tree, the cons branch fills in too: the client wants exclusivity, which kills three smaller accounts; the timeline collides with the product launch; the contact has a reputation for scope creep. None of those came up while everyone was talking about the money. Weighted, the picture flips — two large cons against one large pro and a cluster of small ones — and the "obvious yes" turns into "yes, but only if we can renegotiate the exclusivity clause." The tree didn't decide; it made the *whole* case visible, which the excitement had been hiding.

- **What it shows.** The case for and the case against a single proposal, side by side, each argument branching into its sub-arguments and the rebuttals that answer it — so the decision is examined from both directions, not just defended from one.
- **When to reach for it.** A single yes/no or this-versus-that choice where you want the trade-off laid open — few alternatives, considerations that are mostly qualitative, and a decision-maker who will live with the result.
- **How to read it.** Read the pro branches for the case in favor, the con branches for the case against, and the weights for which arguments actually matter — the strength of the case lives in a few heavy branches, not in how long either list is.
- **What you'd miss without it.** The side nobody was arguing. A room excited about a proposal under-populates the cons; a room sick of one over-populates them. The tree's discipline is that both branches must be filled, which is exactly where the decisive overlooked argument turns up.
- **Where it misleads.** Counting instead of weighing — seven trivial pros can visually outweigh two decisive cons unless the items carry weights (the "additive fallacy"). And the same consideration can sneak in on both sides in different words, double-counting itself, unless the tree is cross-checked for it.

## How to read it

Picture a tree drawn top to bottom. At the top, the **root**: the proposal under evaluation, stated as one decidable thing ("adopt a four-day work week at current pay," not "work-life balance"). Two thick branches grow from it. One gathers the **pros** — the arguments for. The other gathers the **cons** — the arguments against. Off each pro and each con grow thinner branches: a **sub-argument** that makes the point more specific, or a **rebuttal** that answers the argument on the *other* side. Read the left branch and you have the case for; read the right branch and you have the case against; the tree's value is that you can't see one without seeing the other.

The thing that turns this from a flat list into a judgment is **weight**. Each item carries a sense of how much it matters — large, medium, small — assigned by *importance, not by count*. This is the whole point, and it is what naive pro-con lists get wrong: the eye treats a long column as a strong case, but five small conveniences do not outweigh one large irreversible risk. Weighting lets you do what Benjamin Franklin called canceling — set a heavy pro against a heavy con and, if they balance, strike them both out. Do that across the tree and the balanced considerations disappear, leaving a *residual* on one side: the few arguments that the others don't answer. That residual is the decision.

Two disciplines keep the tree honest. First, **weigh, don't count** — mark each branch by importance and read the heavy ones, because the length of a list is not the strength of a case. Second, **cross-check for the same argument twice** — "it's faster" can appear as a pro of option A and, in different words, as a con of option B; flag it so one consideration isn't counted as two. Read back, a good pro-con tree shows not a tally but a balance: the full case each way, and the handful of considerations that survive the canceling and actually decide it.

## When to use it

The pro-con tree belongs to the **DECISION family** of diagrams — the ones built to lay a choice open so it can be made well — and within it this is the *two-sided argument* tool: you reach for it on one proposal to set the case for beside the case against and see which way the balance falls. That places it next to a few relatives, and the boundaries are how you pick the right one:

- A **Comparison Chart** is for *several options against several criteria* — a grid, options across the columns and attributes down the rows. Reach for it when you're choosing among three or more alternatives on many named dimensions; reach for a pro-con tree when it's one proposal, or two, and you want the *arguments* for and against rather than scores in a matrix.
- A **Decision Tree** handles *sequence and uncertainty* — a choice that unfolds over several stages with chance events and probabilities between them. The pro-con tree has no time axis and no probabilities; it weighs the standing arguments on a single decision, not a branching gamble.
- The **Balanced Critique** mode is even-handed evaluation of a *thing* — the strengths and weaknesses of an essay, a design, a plan — not a *decision* between courses of action. The pro-con tree's branches argue *toward a choice*; balanced critique appraises something on its merits without a yes/no to resolve.

Reach for a pro-con tree when the choice is a single yes/no or this-versus-that, the considerations are mostly qualitative, and you want the trade-off made visible and weighable. Skip it when you have many options on many criteria (use a comparison chart), when the decision is sequential and shot through with uncertainty (use a decision tree), or when you're evaluating something rather than deciding between courses (use balanced critique). The pro-con tree is the honest first pass — when it surfaces a decisive consideration the analysis is over, and when it fails to break a tie it tells you to escalate.

## How Ora builds it

Ora produces a pro-con tree from a **semantic spec** — a structured description of the proposal at the root, the pro nodes and con nodes that branch from it, the sub-arguments and rebuttals that branch from those, and an optional weight on each item (large, medium, small, by importance rather than count). That spec is rendered as a two-sided tree: the layout puts the proposal at the top and grows the pros and cons as labelled branches, with green styling for the case in favor and red for the case against, weights shown as brackets prefixing each item, and an outline view plus alt-text and keyboard navigation so the structure survives without relying on color.

The diagram is the visual face of Ora's **decision-support** work: when you ask "lay out the case for and against this — should we do it or not," that context gathers the arguments on each side, weighs them, and cross-checks for the same consideration counted twice, and this artifact is how it shows the result — both cases at once rather than a paragraph that quietly argues for one. It stops at the *qualitative* weighing on purpose; when a decision needs probabilities and stages it escalates to a **Decision Tree**, and when it needs many options scored on many weighted criteria it escalates to **Multi-Criteria Decision Analysis**.

The technique is the work of **Benjamin Franklin**, who set it out in a 1772 letter to the chemist Joseph Priestley. Asked for advice on a hard decision, Franklin declined to give an answer and instead described a method he called "moral or prudential algebra": list the considerations on each side, judge their weights, and strike out a pro against a con of equal weight — "if I find a reason *pro* equal to some two reasons *con*, I strike out the three" — until the residual on one side shows where the balance lies. That procedure is what every modern pro-con list still rests on. The discipline was carried into decision science as the *decision balance sheet* by **Irving Janis and Leon Mann** in *Decision Making* (1977), and the weighted, multi-criteria descendants run through **Keeney and Raiffa**'s *Decisions with Multiple Objectives* (1976) — the tree being the pre-formal sketch those heavier methods grow out of.

## Related

- **Comparison Chart** — the DECISION-family tool for several options against several named criteria in a grid, where this tree weighs the arguments for and against a single proposal.
- **Decision Tree** — the family member for sequence and uncertainty: a branching choice with chance events and probabilities, where this tree weighs standing arguments with no time axis.
- **IBIS Argument Map** — maps multi-party deliberation as questions, positions, and the arguments for and against each, where the pro-con tree is one decision-maker's two-sided weighing toward a single choice.
- **Balanced Critique** and **Decision Architecture** (modes) — the analytical operations this diagram serves: even-handed evaluation of a thing, and the structured laying-open of a decision that this two-sided tree renders.

## Sources

- [Janis, Irving L. & Mann, Leon (1977), Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment, Free Press](https://openlibrary.org/works/OL1282010W)
- [Keeney, Ralph L. & Raiffa, Howard (1976), Decisions with Multiple Objectives, Wiley](https://openlibrary.org/works/OL4107460W)