Context:

You are analyzing whether one or more causal relationships are described in a {granularity}.

---

Causal:
One condition or factor directly influences, enables, or determines another.  
This includes actions, requirements, or conditions of necessity or sufficiency.

Common language cues:
- increase, decrease, lead to, prevent, can, requires, necessary, essential, depends on, results in

Examples:
- "A larger farm size is necessary to achieve a living income."
- "Access to irrigation leads to higher crop yields."
- "To achieve a living income, farmers must diversify crops."

---

Non-causal:  
The {granularity} does **not** describe a causal relationship.  
This includes unrelated statements, neutral descriptions, or associations without direct influence.

Examples:
- "Living income is associated with land size."
- "Yields vary with fertilizer use."
- "The farmer interviewed five neighbors."

---

Task:

Consider the following {granularity}(s), which occur before the {granularity} to be analyzed:  
{context_window}

Now, read the following {granularity}:  
{item}

Does the {granularity} describe any causal relationships?

---

Instructions:

1. Internally reason step by step through the logic of the {granularity}, but provide only the final label as a JSON object.
2. Use only one of these terms: `causal` or `non-causal`.
3. Respond in valid JSON using the following format:

{{
  "label": "causal"
}}

