3.1.2 - Berkeley against indirect realism
Scope: This lesson covers Berkeley's challenge to indirect realism in AQA 7172 3.1.2: mind-dependent ideas cannot be like mind-independent objects, so indirect realism cannot give knowledge of the nature of those objects. It stays within the perception topic by targeting indirect realism's representational bridge; it does not teach Berkeley's Master Argument or the wider objections to Berkeleyan idealism.
The Pressure On Indirect Realism
Indirect realism tries to avoid the problems faced by direct realism. It says that when you perceive a table, the immediate object of your awareness is not the mind-independent table itself. What you are immediately aware of is a mind-dependent item: a sense-datum, or in Berkeley's vocabulary, an idea. The mind-independent object is still supposed to exist, cause your experience, and be represented by the idea.
That creates a sharp question:
How can a mind-dependent idea tell us what a mind-independent object is like?
This is not yet Berkeley's full idealism. The issue here is narrower. Berkeley attacks the indirect realist claim that ideas can represent mind-independent objects in a way that gives us knowledge of their nature. If the immediate object of perception is always an idea, then the mind-independent object is always one step away. Berkeley asks whether the step from idea to object can be justified.
The problem matters because indirect realism needs more than this claim:
- I have ideas or sense-data.
- Something mind-independent causes them.
It also needs this stronger claim:
- My ideas represent the mind-independent object.
- So I can use those ideas to know at least some of what the object is like.
Berkeley's attack is aimed at the second claim. It is a challenge to representation, not merely a challenge to causation.
Key Distinctions
An idea, for Berkeley, is an immediate object of awareness: a colour as seen, a sound as heard, a shape as visually or tactually experienced, a pain as felt. In AQA's wording for indirect realism, the nearby term is sense-datum: a mind-dependent object of perception.
A mind-dependent idea exists only as perceived by a mind. This does not mean the learner is choosing it or imagining it at will. A painful heat sensation or a loud sound can be involuntary. The point is about dependence: its existence as that experienced item is tied to being experienced.
A mind-independent object is supposed to exist whether or not anyone perceives it. For the indirect realist, it is not the immediate object of perception. It is what causes and is represented by the mind-dependent idea.
The likeness principle is Berkeley's claim that an idea can be like only another idea. In the First Dialogue of Three Dialogues Between Hylas and Philonous, Berkeley has Philonous ask whether "anything be like a sensation or idea, but another sensation or idea?" The argument is not just that we happen not to have compared ideas with external objects. It is that the comparison itself seems blocked: one side of the comparison is mind-dependent and immediately perceived, while the other is supposed to be mind-independent and not immediately perceived.
| Distinction | What it means here | Why it matters |
|---|---|---|
| Idea/sense-datum vs object | The idea is immediately perceived; the object is supposed to be represented by it. | Berkeley presses the gap between what is perceived and what is inferred. |
| Mind-dependent vs mind-independent | The idea depends on being perceived; the object is supposed not to. | The two sides seem to belong to different categories. |
| Causation vs resemblance | An idea might be caused by something without resembling it. | Berkeley's attack is strongest when indirect realism needs resemblance to explain knowledge of nature. |
| Existence vs nature | There may be an external cause; the question is what it is like. | The AQA issue is specifically knowledge of the nature of mind-independent objects. |
Berkeley's Argument Reconstructed
Berkeley's challenge can be reconstructed as an argument against the indirect realist's representational claim.
Claim at issue: indirect realism cannot explain how mind-dependent ideas give us knowledge of the nature of mind-independent objects.
1. P1: Indirect realism says that the immediate objects of perception are mind-dependent ideas or sense-data.
2. P2: Indirect realism also says that these ideas represent mind-independent objects.
3. P3: To know the nature of a mind-independent object through an idea, the idea must represent what the object is like.
4. P4: If representation is by resemblance, then the idea must be like the mind-independent object.
5. P5: A mind-dependent idea can be like only another idea, not a mind-independent object.
6. Therefore, C: Indirect realism cannot show that our ideas give us knowledge of the nature of mind-independent objects.
Hidden assumption: knowledge of an object's nature through perception requires a representational link strong enough to license claims about what the object is like.
Inference type: conceptual and epistemological. It analyses what representation by likeness would require, then draws a sceptical conclusion about knowledge of nature.
Main pressure point: P4 and P5. Does representation require resemblance, and is cross-category resemblance really incoherent?
Notice the conclusion. Berkeley is not merely saying that we cannot stare directly at mind-independent objects. The indirect realist already accepts that. Berkeley is saying that, given indirect realism's own picture, the object behind the idea becomes unknowable in nature. The idea is all we have immediate access to, but the idea cannot be treated as a likeness of something non-ideal.
This is why the argument is dangerous for indirect realism. It turns the theory's protective move into a liability. Indirect realism protects us from illusion and hallucination by saying we directly perceive mental items. Berkeley replies: if all we directly perceive are mental items, how do those items get beyond themselves?
The Target Of The Attack
Berkeley's target is not every possible form of realism. Nor, in this lesson, is the target the whole existence of the external world. The precise target is the indirect realist bridge from ideas to mind-independent objects.
That bridge has two jobs:
- Causal job: explain why ideas occur.
- Representational job: explain why ideas tell us what objects are like.
Berkeley's likeness argument attacks the representational job. A cause need not resemble its effect. A fire can cause smoke without looking like smoke. So even if mind-independent objects caused our ideas, that alone would not show that the ideas reveal the objects' nature.
The attack is especially sharp against a Lockean form of indirect realism. Locke's primary/secondary quality distinction suggests that some qualities, such as shape, extension, motion, number, and solidity, belong to objects in a way that our ideas of them resemble. Berkeley presses: if extension and shape as perceived are still ideas, how can they resemble something mind-independent and insensible?
The inference can be mapped like this:
| Stage | Indirect realist claim | Berkeley's pressure |
|---|---|---|
| Immediate perception | We directly perceive ideas or sense-data. | Then ideas are the only directly available items. |
| External object | Mind-independent objects cause those ideas. | Causation alone does not reveal nature. |
| Representation | Ideas represent objects by being like them, at least in primary qualities. | Likeness between an idea and a non-idea is not intelligible or knowable. |
| Result | We can know the nature of mind-independent objects indirectly. | We are left with ideas, but not justified access to the object's nature. |
The argument is therefore a form of internal pressure. Berkeley uses indirect realism's own division between immediate ideas and external objects to ask whether indirect realism has cut the learner off from the very world it wanted to preserve.
Objections And Replies
The strongest evaluation does not simply say that Berkeley is right or wrong. It asks which part of the argument is under pressure.
| Objection | Target | Why it matters | Berkeleyan reply | Residual issue |
|---|---|---|---|---|
| Representation does not require resemblance. Words, maps, mathematical symbols, and signals can represent without looking like what they represent. | Hidden assumption and P4. | If representation can be non-resembling, Berkeley has not defeated representation itself. | Berkeley can reply that the issue is knowledge of nature. Symbols work when we can calibrate them against what they represent; the indirect realist lacks direct access to the mind-independent object for that comparison. | A causal or structural account of representation might still give probable knowledge, though less than direct resemblance would promise. |
| Ideas might resemble the structure of objects rather than their sensory qualities. For example, spatial relations in experience might track real relations outside the mind. | P5. | This avoids claiming that colour-like or sound-like qualities exist in matter while preserving some knowledge of the external world's structure. | Berkeley can press that the experienced structure is still an idea. If the object is not itself perceivable, how do we know the structure in experience matches its mind-independent structure? | The indirect realist may move toward a best-explanation argument: structural fit is justified by prediction and coherence, not direct comparison. |
| Causal origin is enough. An idea represents the object because the object reliably causes it. | P3 and P4. | If reliable causation is sufficient for representation, resemblance is unnecessary. | Berkeley can accept that causes may be inferred while denying that this reveals their intrinsic nature. Many different causes could produce the same pattern of ideas. | Scientific theory may strengthen the causal story, but the conclusion becomes abductive and fallible, not a direct grasp of nature. |
| Berkeley compares across categories too quickly. We compare unlike kinds of things all the time, such as a photograph with a person or a graph with a process. | P5. | If cross-category comparison is coherent, the likeness principle may be too strong. | Berkeley can reply that those examples involve two accessible sides: we can compare the photograph with the person or the graph with observed data. The mind-independent object, as indirect realism defines it, is not available except through ideas. | The objection works best if the indirect realist can explain independent checks on the external object. Without those checks, Berkeley's epistemic worry remains. |
These objections show why the debate is not a simple knockdown. Berkeley's argument is most forceful against indirect realism when representation is understood as resemblance. It is less decisive if indirect realism can replace resemblance with causal, structural, or abductive representation. But those replacements often concede Berkeley's main pressure: our knowledge of mind-independent objects is not secured by a simple likeness between idea and object.
Exam Synthesis
For AQA, the core move is to explain Berkeley's challenge and its target precisely. A strong answer keeps the debate inside indirect realism and does not drift into a full essay on Berkeley's idealism.
For a 3-mark definition or distinction, focus on compact accuracy:
| Term | Creditworthy point |
|---|---|
| Mind-dependent idea | An immediate object of awareness whose existence depends on being perceived by a mind. |
| Mind-independent object | An object supposed to exist independently of being perceived. |
| Likeness principle | Berkeley's claim that an idea can be like only another idea. |
For a 5-mark explanation, reconstruct the route:
- Indirect realism says we immediately perceive mind-dependent sense-data or ideas.
- It also claims those ideas represent mind-independent objects.
- Berkeley argues that ideas cannot resemble mind-independent objects because ideas are mind-dependent and objects are not ideas.
- If the ideas cannot resemble the objects, they cannot reveal what those objects are like.
- So indirect realism cannot secure knowledge of the nature of mind-independent objects.
For a 12-mark explanation, add target and inference analysis as part of accurate AO1 explanation:
- State the target: the representational bridge from sense-data to objects, not merely the existence of external causes.
- Reconstruct the likeness argument with premises and conclusion.
- Explain why the argument is plausible: indirect realism seems to put a veil of ideas between the mind and the world.
- Identify the hidden assumption: knowing an object's nature through an idea requires a resemblance or comparability relation.
- Note the main pressure point: an indirect realist may deny that representation requires resemblance.
A controlled synthesis could be:
Berkeley gives a powerful internal criticism of indirect realism because he uses its own distinction between mind-dependent ideas and mind-independent objects against it. The argument is strongest if indirect realism depends on resemblance to explain representation. However, if the indirect realist shifts to causal, structural, or best-explanation representation, Berkeley has not shown that all representation fails. He has shown that indirect realism owes a further account of how ideas can justify claims about the nature of mind-independent objects.