“The Relation of Sense-Data to Physics,” pages 164-179
Recall that this second post covering Chapter VIII is devoted to the following sections:
IX. The Definition of Matter (p. 164)
X. Time (p. 167)
XI. The Persistence of Things and Matter (p. 169)
XII. Illusions, Hallucinations, and Dreams (p. 173).
We still have the issue of what constitutes “matter,” even after we define a thing as the class of its various appearances. Individual appearances are affected by this “matter.” Appearances provide more detail as they become closer; so, we can characterize a thing’s matter as the limit of its appearances as the closeness of the thing goes to zero (p. 165). (We cannot be sure that empirically, such a limit exists; but, we can infer (with error) a limit from the appearances that we do observe.)
“The appearance of a thing in a given perspective is a function of the matter composing the thing and of intervening matter [p. 165].” The intervening matter, for instance, might be a mist or some element of the receiving sense-organ. As we get closer to the thing, the intervening matter is less relevant – hence the thing itself is the limit of appearances as we approach the thing in question. Appearances are deceptive, but the closer they are, the more confidence we have in them. Matter is not “more real” than sense-data, but it is more reliable – more substantive? – than any one piece of sense data. Of course, as we approach an object, we see that it is not one object, but several, and objects seem to be infinitely divisible in this fashion: a single appearance can include many “things.”
For a single observer (not more generally, as relativity makes clear), two perspectives can be ordered in time, with one perspective before, after, or simultaneous with the other. We can extend this notion to sensibilia, so that a “biography” is “everything that is (directly) earlier or later than, or simultaneous with, a given ‘sensibile’ [p. 167].” The world’s history, then, is comprised of the union of “mutually exclusive biographies.” [Russell cites A. A. Robb as a source for his own views on time. In 1925, Russell went on to write ABC of Relativity, which was the first Russell book I [Bert] ever came across and read. An amazing audio version of ABC of Relativity, read by Derek Jacobi, is available for free here.]
How can the time in different biographies be synchronized? In the case of a sound, we can’t say that in every biography containing the sound, it occurs simultaneously, because some listeners (those closer to the source) do hear it earlier. Russell chooses to build a “velocity of sound” into his framework for such audible sensibilia. A similar approach, using a velocity of light, is taken for visual sensibilia. This solution has the (perhaps unfortunate?) implication that in trying to encapsulate a thing at a certain instance, we use appearances that themselves do not all occur at that same instant. “The” time at which the thing is in a certain state is the lower limit of all these instances.
So, we now have correlated appearances (in various perspectives) that give us one thing at a specific instant. But what about persistence, the connection of that same thing at different instances? Again, we have to construct persistence, just as we constructed the notion of a thing at a certain time with different observers – now the construction involves observers at different times.
If we look at just a single biography, how do we know that the same thing exists at different moments? The different appearances must be correlated, display a type of continuity. But we don’t observe anything continuously, so our assumption of continuity is only a hypothesis – though a hypothesis that we already have employed in developing our notions of particulars from sensibilia.
Continuity doesn’t imply a constant material source. A homogeneous fluid like a sea possesses continuity, but the motions of sea water “cannot be inferred from direct sensible observation together with the assumption of continuity [p. 171].” So, for those seemingly persistent objects that we take to be collections of sensibilia, we must also include adherence to the laws of dynamics.
Motion is the change over time in those assembled sensibilia that we take to be the same object. Because we have some discretion over what sensibilia constitute the same object over time, our notion of motion has some unavoidable ambiguity, if continuity is our only guide. Requiring adherence to the laws of dynamics resolves this ambiguity – and we will assume that there is a unique grouping of appearances into things that possesses this coherence (p. 173). By this method, we can identify how appearances at different times can be attributed to the same object.
Much of the luster of physics is due to its empirical successes, despite our inability to generate appropriate sense-data to test some of the hypotheses of physics. Nonetheless, we find that the sense-data we do have are not in contradiction with those hypotheses, and indeed, the hypotheses, combined with some sense-data, allow us to predict other sense-data.
Russell concludes section XI (“The Persistence of Things and Matter”) with a definition and a claim. The definition is: “Physical things are those series of appearances whose matter obeys the laws of physics [p. 173].” The claim is that we know empirically that such things exist, and this knowledge completes the program, verifies physics, deduces stuff from sense-data.
But what about “unreal” sense-data like hallucinations, which are not correlated in the usual way with actual stuff? What if we dream about an unreal thing? Recall that our definition of stuff like a table is that it is the collection of all sensibilia that contain it. Given that definition, then, “as well sleeping, as waking,” any observer doesn’t sense the table, but only one sensibilia.
Dream objects exist in the private space of the dreamer; they lose the correlation with other private spaces that real objects possess.
Such notions as “existence” and “non-existence” do not apply to sense-data, though they can apply to things described in terms of such data. The fact that a sense-datum exists, and that that datum is of a table, does not imply that a table exists. [Russell cites the fuller, symbol-based discussion of this point in Principia Mathematica, and also makes reference (p. 176) to the theory of descriptions.]
Russell goes on (pages 176-179) to indicate how his theoretical development answers four common arguments against realism:
(1) The same object can appear differently to different people. Russell allows different perspectives, which render these observational differences to be irrelevant for the question of object reality.
(2) The same object can give us seemingly incompatible sense-data. A stick in water can appear bent to the eye, though straight to the touch. But when a stick is a collection of sensibilia, and we recognize that the proper inference from one sensibile to another need not be fixed, incompatible sense-data are not a problem for a realist view of matter.
(3) Objects in dreams often are considered to be unreal—but the sense-data that lead to dreams are as real as any. As noted, dream sense-data lack the usual continuity and correlation with other sense-data. Nonetheless, they are physical, subject matter for physics to deal with.
(4) Hallucinations are like dreams, but it takes multiple observers for it to become clear that one person is suffering from hallucinations. (Dreams would have the same property if they were more closely connected to a person’s sense-data when awake.) The person suffering from hallucinations is in no position to know whether this is the case, or whether other people are conspiring against him.
“From the above instances it would appear that abnormal sense-data, of the kind which we regard as deceptive, have intrinsically just the same status as any others, but differ as regards their correlations or causal connections with other ‘sensibilia’ and with ‘things’ [p. 179].” The problem is not unreal data but rather, our unjustified expectations. As a result, abnormal sense-data do not present an impregnable obstacle for the science of physics. The approach provided above, though quite preliminary, especially as regards its role for time, permits physics to be empirically verifiable.
Recall that this second post covering Chapter VIII is devoted to the following sections:
IX. The Definition of Matter (p. 164)
X. Time (p. 167)
XI. The Persistence of Things and Matter (p. 169)
XII. Illusions, Hallucinations, and Dreams (p. 173).
We still have the issue of what constitutes “matter,” even after we define a thing as the class of its various appearances. Individual appearances are affected by this “matter.” Appearances provide more detail as they become closer; so, we can characterize a thing’s matter as the limit of its appearances as the closeness of the thing goes to zero (p. 165). (We cannot be sure that empirically, such a limit exists; but, we can infer (with error) a limit from the appearances that we do observe.)
“The appearance of a thing in a given perspective is a function of the matter composing the thing and of intervening matter [p. 165].” The intervening matter, for instance, might be a mist or some element of the receiving sense-organ. As we get closer to the thing, the intervening matter is less relevant – hence the thing itself is the limit of appearances as we approach the thing in question. Appearances are deceptive, but the closer they are, the more confidence we have in them. Matter is not “more real” than sense-data, but it is more reliable – more substantive? – than any one piece of sense data. Of course, as we approach an object, we see that it is not one object, but several, and objects seem to be infinitely divisible in this fashion: a single appearance can include many “things.”
For a single observer (not more generally, as relativity makes clear), two perspectives can be ordered in time, with one perspective before, after, or simultaneous with the other. We can extend this notion to sensibilia, so that a “biography” is “everything that is (directly) earlier or later than, or simultaneous with, a given ‘sensibile’ [p. 167].” The world’s history, then, is comprised of the union of “mutually exclusive biographies.” [Russell cites A. A. Robb as a source for his own views on time. In 1925, Russell went on to write ABC of Relativity, which was the first Russell book I [Bert] ever came across and read. An amazing audio version of ABC of Relativity, read by Derek Jacobi, is available for free here.]
How can the time in different biographies be synchronized? In the case of a sound, we can’t say that in every biography containing the sound, it occurs simultaneously, because some listeners (those closer to the source) do hear it earlier. Russell chooses to build a “velocity of sound” into his framework for such audible sensibilia. A similar approach, using a velocity of light, is taken for visual sensibilia. This solution has the (perhaps unfortunate?) implication that in trying to encapsulate a thing at a certain instance, we use appearances that themselves do not all occur at that same instant. “The” time at which the thing is in a certain state is the lower limit of all these instances.
So, we now have correlated appearances (in various perspectives) that give us one thing at a specific instant. But what about persistence, the connection of that same thing at different instances? Again, we have to construct persistence, just as we constructed the notion of a thing at a certain time with different observers – now the construction involves observers at different times.
If we look at just a single biography, how do we know that the same thing exists at different moments? The different appearances must be correlated, display a type of continuity. But we don’t observe anything continuously, so our assumption of continuity is only a hypothesis – though a hypothesis that we already have employed in developing our notions of particulars from sensibilia.
Continuity doesn’t imply a constant material source. A homogeneous fluid like a sea possesses continuity, but the motions of sea water “cannot be inferred from direct sensible observation together with the assumption of continuity [p. 171].” So, for those seemingly persistent objects that we take to be collections of sensibilia, we must also include adherence to the laws of dynamics.
Motion is the change over time in those assembled sensibilia that we take to be the same object. Because we have some discretion over what sensibilia constitute the same object over time, our notion of motion has some unavoidable ambiguity, if continuity is our only guide. Requiring adherence to the laws of dynamics resolves this ambiguity – and we will assume that there is a unique grouping of appearances into things that possesses this coherence (p. 173). By this method, we can identify how appearances at different times can be attributed to the same object.
Much of the luster of physics is due to its empirical successes, despite our inability to generate appropriate sense-data to test some of the hypotheses of physics. Nonetheless, we find that the sense-data we do have are not in contradiction with those hypotheses, and indeed, the hypotheses, combined with some sense-data, allow us to predict other sense-data.
Russell concludes section XI (“The Persistence of Things and Matter”) with a definition and a claim. The definition is: “Physical things are those series of appearances whose matter obeys the laws of physics [p. 173].” The claim is that we know empirically that such things exist, and this knowledge completes the program, verifies physics, deduces stuff from sense-data.
But what about “unreal” sense-data like hallucinations, which are not correlated in the usual way with actual stuff? What if we dream about an unreal thing? Recall that our definition of stuff like a table is that it is the collection of all sensibilia that contain it. Given that definition, then, “as well sleeping, as waking,” any observer doesn’t sense the table, but only one sensibilia.
Dream objects exist in the private space of the dreamer; they lose the correlation with other private spaces that real objects possess.
Such notions as “existence” and “non-existence” do not apply to sense-data, though they can apply to things described in terms of such data. The fact that a sense-datum exists, and that that datum is of a table, does not imply that a table exists. [Russell cites the fuller, symbol-based discussion of this point in Principia Mathematica, and also makes reference (p. 176) to the theory of descriptions.]
Russell goes on (pages 176-179) to indicate how his theoretical development answers four common arguments against realism:
(1) The same object can appear differently to different people. Russell allows different perspectives, which render these observational differences to be irrelevant for the question of object reality.
(2) The same object can give us seemingly incompatible sense-data. A stick in water can appear bent to the eye, though straight to the touch. But when a stick is a collection of sensibilia, and we recognize that the proper inference from one sensibile to another need not be fixed, incompatible sense-data are not a problem for a realist view of matter.
(3) Objects in dreams often are considered to be unreal—but the sense-data that lead to dreams are as real as any. As noted, dream sense-data lack the usual continuity and correlation with other sense-data. Nonetheless, they are physical, subject matter for physics to deal with.
(4) Hallucinations are like dreams, but it takes multiple observers for it to become clear that one person is suffering from hallucinations. (Dreams would have the same property if they were more closely connected to a person’s sense-data when awake.) The person suffering from hallucinations is in no position to know whether this is the case, or whether other people are conspiring against him.
“From the above instances it would appear that abnormal sense-data, of the kind which we regard as deceptive, have intrinsically just the same status as any others, but differ as regards their correlations or causal connections with other ‘sensibilia’ and with ‘things’ [p. 179].” The problem is not unreal data but rather, our unjustified expectations. As a result, abnormal sense-data do not present an impregnable obstacle for the science of physics. The approach provided above, though quite preliminary, especially as regards its role for time, permits physics to be empirically verifiable.
