3.2 Theories of reference: descriptive and causal-historical

Theories of Reference

Theories of reference ask a deceptively simple question: how do words latch onto things in the world? When you say "Aristotle," what makes that name pick out one specific person who lived in ancient Greece? Two major answers have emerged in the philosophy of language: descriptive theories and causal-historical theories. They disagree sharply on what fixes reference, and each has trade-offs worth understanding.

Descriptive vs. Causal-Historical Theories

Descriptive theories say that a term refers to whatever satisfies the descriptions a speaker associates with it. The reference of a name or phrase is determined by its descriptive content.

• "The tallest mountain in the world" refers to Mount Everest because Everest is the object that fits that description.
• On this view, the name "Aristotle" refers to whoever satisfies descriptions like the student of Plato, the teacher of Alexander, the author of the Nicomachean Ethics, and so on.
• Meaning is tied to the descriptions, not directly to the object itself.

Causal-historical theories (developed most famously by Saul Kripke) say that reference is fixed by an initial baptism or naming event, and then preserved through a causal chain of communication from speaker to speaker across time.

• "Aristotle" refers to a specific person because that person was given the name at some point, and each subsequent use of the name traces back through a chain of speakers to that original event.
• The descriptions people associate with the name don't determine who it picks out.

The core contrast comes down to this:

This leads to different predictions. If Everest somehow lost height and K2 became taller, a descriptive theory predicts "the tallest mountain" would shift to refer to K2. A causal-historical theory, by contrast, predicts that "Aristotle" still refers to the original person even if every description we associate with him turned out to be wrong.

Strengths and Weaknesses

Descriptive theories have some clear appeal:

• They capture the intuition that meaning is connected to what we know about a referent. When you use a name, it feels like the descriptions in your head are doing some work.
• They explain how reference can shift over time. "The fastest man alive" referred to Usain Bolt for years, and now it refers to someone else. Descriptive theories handle this naturally.

But they run into trouble:

• The problem of ignorance: Someone can successfully refer to Einstein without knowing any accurate descriptions beyond "some famous scientist." Descriptive theories predict this should fail, but it clearly doesn't.
• The problem of error: If all the descriptions a community associates with a name turn out to be false, the name still seems to refer to the original person. Descriptive theories have difficulty explaining why.

Causal-historical theories handle those problems well:

• Reference succeeds even when speakers have incomplete or inaccurate beliefs, because what matters is the causal chain back to the naming event, not the speaker's knowledge.
• They explain the stability of reference across centuries. "Plato" picks out the same person today as it did in antiquity, regardless of how our descriptions of him have evolved.

But they face their own challenges:

• They struggle to account for the intuition that descriptive content matters to meaning. If "Aristotle" means nothing more than "the guy originally given that name," it's hard to explain why "Aristotle was a philosopher" feels more meaningful than a bare identity claim.
• Explaining how entirely new terms get introduced, or how reference changes, is less straightforward on this model.

Application to Real-World Examples

These theories generate different predictions when applied to concrete cases.

Using a descriptive theory: "The first person to walk on the moon" refers to Neil Armstrong because he satisfies that description. If new evidence somehow revealed that Buzz Aldrin secretly stepped onto the surface first, the referent of that phrase would shift to Aldrin. The description determines who gets picked out.

Using a causal-historical theory: "Christopher Columbus" refers to the person originally given that name, full stop. Even if historians discovered that a secret twin actually performed the voyages attributed to Columbus, the name "Columbus" would still refer to the person baptized with it. The descriptions we associate with the name don't control who it picks out.

These aren't just thought experiments. Whenever historical facts get revised, the two theories make genuinely different predictions about whether our terms still refer to the same individuals.

Implications for Meaning and Truth

The two theories lead to different accounts of how meaning and truth work.

Under descriptive theories:

• Meaning is determined by descriptive content, so "The tallest mountain is Mount Everest" is true if and only if Everest actually satisfies the description tallest mountain.
• Meaning and truth values can shift as the world changes or as our descriptions get updated.

Under causal-historical theories:

• Truth depends on the properties of the actual referent, the individual fixed by the original naming event. "Aristotle was a philosopher" is true if and only if the person originally named "Aristotle" was in fact a philosopher, even if many other things we believe about him are wrong.
• Meaning and truth are more stable because they don't fluctuate with changes in what speakers happen to believe.

Neither theory has won outright. Many contemporary semanticists draw on elements of both, recognizing that descriptive content and causal history each play a role in how reference works. For an intro course, the key takeaway is understanding what each theory claims and where it runs into difficulty.