Truth and Semantics

How to read this page: This article maps the topic from beginner to expert across six levels � Remembering, Understanding, Applying, Analyzing, Evaluating, and Creating. Scan the headings to see the full scope, then read from wherever your knowledge starts to feel uncertain. Learn more about how BloomWiki works ?

Truth and Semantics is the study of "How sentences match the world"—the investigation of what it "Means" for a statement to be "True." While we use the word "Truth" every day, defining it mathematically and philosophically is a massive challenge. From **Alfred Tarski's** "Convention T" (which defined truth for formal languages) to **Donald Davidson's** "Truth-Conditional Semantics" (which argued that 'Knowing the Meaning' of a sentence is 'Knowing the Conditions' under which it would be True), this field explores the "Structure of Fact." It is the science of "Verification," explaining how "Words" become "Knowledge" and why the "Liar Paradox" (e.g., 'This sentence is false') almost destroyed the foundations of logic.

Remembering

• Semantics — The branch of linguistics and philosophy concerned with "Meaning."
• Truth-Condition — The "State of the world" that must exist for a sentence to be true (e.g., the truth-condition for 'It is raining' is 'That water is currently falling from the sky').
• T-Schema (Convention T) — Alfred Tarski's formula: **"P" is true if and only if P** (e.g., "The snow is white" is true iff the snow is white).
• Correspondence Theory of Truth — The view that truth is "Correspondence to Reality" (A statement is true if it matches a 'Fact').
• Coherence Theory of Truth — The view that truth is "Consistency" with other beliefs (A statement is true if it 'Fits' into our web of knowledge).
• Pragmatic Theory of Truth — The view that truth is "What Works" (A statement is true if it is 'Useful' for solving problems).
• Compositionality — The principle that the meaning of a "Whole Sentence" is built from the "Meanings of its Parts" (Words) and its "Structure" (Grammar).
• Metalanguage vs. Object Language — Tarski's solution to the Liar Paradox: you must use a "Higher Language" (Metalanguage) to talk about the "Truth" of a "Lower Language" (Object Language).
• Logical Form — The "Skeleton" of a sentence that shows its "Truth-Logic" (e.g., 'All men are mortal' becomes 'For all x, if x is a man, then x is mortal').
• Analytical vs. Synthetic — The distinction between "True by Definition" (e.g., 'Bachelors are unmarried') and "True by Observation" (e.g., 'The cat is on the mat').

Understanding

Truth and semantics are understood through Conditions and Levels.

1. Meaning as Truth (Davidson): How do you "Understand" a sentence you have never heard before?

• Donald Davidson argued that to "Understand" is to "Know when it would be true."
• If I say: "There is a blue elephant in your kitchen," you know what I mean **because** you know "Exactly what would have to happen in the world" for that to be true (you would see a blue elephant).
• Meaning is a "Bridge" between "Sound" and "World-Conditions."

2. The "T-Schema" (Tarski): Is truth just "Repeating the sentence"?

• Tarski wanted to "Define" truth without using the word "Truth."
• He said: The sentence "Snow is white" (The Object) is true if, in the world, **snow is white** (The Fact).
• This "Redundancy" is the foundation of "Model Theory"—the math used to check if a computer program is "Correct."

3. The "Liar" Problem (Levels): "This sentence is False."

• If it's True, then it's False. If it's False, then it's True.
• Tarski solved this by saying a language "Cannot talk about its own truth."
• To say "Sentence A is true," you must be standing "Outside" the language of A. This "Hierarchy of Languages" prevents the "Loop" of the Liar Paradox from breaking the system.

The 'Snow is White' Formula': Tarski's most famous example. It sounds simple, but it "Anchored" the abstract world of logic to the physical world of facts. It proved that "Truth" is not a "Mysterious Quality," but a "Relationship" between "Symbols" and "Objects."

Applying

Modeling 'The Truth Condition' (Mapping a sentence to its requirements): <syntaxhighlight lang="python"> def verify_truth(sentence, world_state):

   """
   Simulates Truth-Conditional Semantics.
   """
   # Sentence: "The light is ON"
   # Logic: True if state['light'] == 1
   if sentence == "Light is ON":
       condition = (world_state["light"] == 1)
   elif sentence == "Door is CLOSED":
       condition = (world_state["door"] == 0)
   else:
       condition = False
       
   if condition:
       return f"Sentence '{sentence}' is TRUE (Matches the World)"
   else:
       return f"Sentence '{sentence}' is FALSE (World does not match)"

1. World: Light is on, Door is open.

current_world = {"light": 1, "door": 1} print(verify_truth("Light is ON", current_world)) </syntaxhighlight>

Truth Landmarks

Tarski’s 'The Concept of Truth in Formalized Languages' (1933) → The most important paper in the history of semantics.

Wittgenstein’s 'Tractatus' (1921) → The "Picture Theory of Language": the idea that a sentence is a "Map" or "Picture" of a fact in the world.

Quine’s 'Two Dogmas of Empiricism' (1951) → Challenged the "Analytical/Synthetic" distinction, arguing that our "Whole Web of Belief" is tested against the world together.

Model Theory → The branch of math that uses Tarski's logic to "Build worlds" (Models) where certain "Sentences" are true, used in AI and Software Engineering.

Analyzing

Correspondence vs. Coherence

Feature	Correspondence (Tarski/Russell)	Coherence (Spinoza/Blanshard)
Test	"Does it match the World?"	"Does it fit my other beliefs?"
Strength	Realistic / Scientific	Logical / Integrated
Weakness	Hard to "Compare" words to things	Could be a "Consistent Lie" (e.g. a dream)
Analogy	A 'Photograph'	A 'Puzzle' (Every piece must fit)

The Concept of "Compositionality": Analyzing "Infinite Meaning." We only have 26 letters and a few thousand words, but we can make "Infinite Sentences." This is because meaning is "Additive." If you know "Red" and "Car," you automatically know "Red Car." Semantics is the "Math of Addition" for concepts.

Evaluating

Evaluating truth and semantics:

1. The "Post-Truth" Crisis: If "Truth" is just "Agreement" (Coherence), can we ever have an "Objective Fact" that everyone must accept?
2. Vagueness: What if a sentence is "Half-True"? (e.g., 'He is bald'). How many hairs make it true? (The "Sorites Paradox").
3. AI and Truth: Can a "Large Language Model" understand "Truth" if it has "No Eyes" to see the world? (Is it just 'Coherence' without 'Correspondence'?).
4. Relativism: Is truth "Relative" to a language? (Does "Truth" change if I change the "Rules" of my words?).

Creating

Future Frontiers:

1. The 'Truth-Verifier' AI: An AI that "Reads a Scientific Paper" and "Checks it against a Model of the World" to see if it is "Semantically Sound."
2. Formalized Global Law: Writing "Law" in a "Metalanguage" that is "Mathematically True," ending the "Ambiguity" of legal "Object Languages."
3. Semantic Web 3.0: An internet where "Data" has "Meaning" (Truth-Conditions) attached to it, allowing computers to "Understand" facts rather than just "Displaying" them.
4. Multi-Valued Logic: A new "Semantics" for a world where things can be "True," "False," "Unknown," or "Partially True" (Fuzzy Logic).