:up: Ordering that book today. Despite...

Where fans of synaesthesia allege "cross-talk" between folds of cortex (so what?) I prefer this kind of talk:

How our lookings at pictures and our listenings to music inform what we encounter later and elsewhere is integral to them as cognitive. Music can inform perception not only of other sounds but also of the rhythms and patterns of what we see. Such cross-transference of structural properties seems to me a basic and important aspect of learning, not merely a matter for novel experimentation by composers, dancers, and painters.
— Goodman: Languages of Art

(My emphasis.) — bongo fury

Boxing Day, 10.25: Just noticeably sharp.

13.10: Semitone sharp.

16.25: Good.

17.40: Good.

20.20: Good. Maybe sharp.

23.25: Bit sharp.



Day 36, 11.10: Good.

12.40: Good.

14.40: More than just noticeably sharp.

17.45: Good.

21.15: Slightly flat. (But quick.)



Day 37, 15.45: Semitone sharp.

18.55: Good.

22.20: Most of a semitone flat.



Day 38: 16.35: Good. And Quick.

20.25: Slightly flat.

21.15: Good.



Day 39: 10.05: Good. and Quick.

13.00: Good.

18.15: Slightly sharp.

21.30 Good.



Day 40, 10.15: Slightly sharp.

16.45: Good.

18.45: Semitone sharp.

21.35: Good.



Day 41, 10.35: Good.

12.50: Good.

16.05: Good.

18.55: Good.

20.55: Good.

23.05: Good.



Day 42, 10.05: Good.

11.40: Good.

15.30: Well flat.

16.50: Good.

21.20: Good.



Day 43, 08.40: Good.

15.50: Good.

19.00: Semitone sharp.

20.00: Sharp again.

22.50: Good.



Day 44, 12.55: Semitone flat.

19.35: Good.



Day 45, 08.20: Good.

17.45: Good.

20.40: Slightly sharp.

20.55: Good.



..day 46, 07.50: Good.

11.30: Good.

13.05: Good.

16.20: Good.

17.15: Most of a semitone sharp.

19.20: Good.



Day 47, 12.30: Good.

16.05: Good.

18.50: Good.

20.55: Semitone sharp. Ish.

22.40: Good.



Day 48, 12.55: Good.

16.05: A tad sharp.

22.00: Good.

23.40: Good.



Day 49, 10.30: Good.

12.25: Slightly flat.

13.00: Good.

16.25: Good.

18.40: Good.

21.40: Good.



Day 50, 10.10: Good.

12.10: Good.

16.05: Good.

19.35: Good.

20.43: Good.



Day 51, 17.35: Good.

23.00: Good.



Day 52, 17.40: Good.

22.10: Good.



.Day 53, 10.50: Good.

17.50: Good.

18.45: Good.

21.10: Good.

23.20: Good.



Day 54, 08.30: Slightly sharp.

14.15: Ditto. (Both hurried.)

18.00: Good. (And hurried.)

24.00: Good.



Day 55, 13.20: Good.

16.10: Good.

19.50: Good.

21.35: Good.



Day 56, 09.50: Good.

14.15: Good.

16.20: Good.

23.50: Semitone sharp.



Day 57, 09.45: Good.

14.25: Good.

20.05: Good.



Day 58, 14.05: Good.

15.45: Good.

17.45: Good.

00.15: Can't get YouTube but pretty sure it's good. In at least 20 per cent of trials I now have this feeling straight away, generally confirmed. Still have to wait a while usually.



Day 59, 09.50: Haha, forgot to hang on and check last night's dubious claim. Anyway, good right now, but only after a couple of minutes' struggle to obtain an "image" having the required feeling of certainty.

17.10: Good.

00.35: Good.



Day 60, 15.05: Good.

16.30: Good.

18.45: Slightly sharp.

21.00: Good.



Day 61, 08.30: Good.

15.00: Good.

21.20: Ouch, can't hurry yet. Semitone sharp.

23.00: Good.



Day 62, 10.00: Good.

19.45: A bit flat.

20.35: Good.

22.10: Good.



Day 63, 10.55: Slightly flat.

14.35: Good.

17.20: Good.

00.55: Good.



Day 64, 09.25: Good.

15.50: Good.

17.20: Good.

20.00: Good.

21.30: Good.



Day 65, 16.40: Good.

19.40: Sharp.



Day 66, 10.05: Good.

17.40: Good.

22.20: Semitone flat.



Day 67, 12.20: Slightly flat.

17.55: Good.

23.45: Good.



Day 68, 19.05: Good.

00.30: Good.



Day 69, 14.55: Good.

16.50: Good.

21.10: Good.

00.50: Slightly flat.



Day 70, 10.45: Slightly flat.

15.00: Slightly sharp.

17.45: Good.

21.00: Good.

00.45: Slightly flat.



Day 71, 09.45: Good.

12.05: Good.

18.55: Good.

20.30: Good.



Day 72, 19.50: Good.

21.35: Slightly sharp.



Day 73, 10.15: Slightly flat.

12.20: Good.

17.00: Good.

21.20: Good.

23.50: Sharp.



Day 74, 10.55: Slightly flat.

18.30: Good.

21.00: Aargh, semitone sharp. Still can't hurry.



Day 75, 14.30: Flat.

17.35: Good.

19.25: Good.

00.00: Good.



Day 76,10.30: Good.

15.15: Good.

18.40: Flat.

20.35: Good.

23.00: Good.



Day 77, 13.00: Good.

17.15: Good.

20.15: Flat.



Day 78, 11.45: Good.

14.40: Flat.

16.20: Good.

21.10: Good.



Day 79, 11.35: Good.

19.10: Good.

22.00: Slightly flat.



Day 80, 09.10: Good.

18.00: Good.

21.35: Good.

23.05: Good.



Day 81, 13.10: Good.

17.25: Good.

19.00: Good.

22.25: Good.



Day 82, 13.20: Good.

16.35: Good.

19.00: Slightly sharp.

20.50: Good.



Day 83, 09.00: Good.

11.55: Good.

15.55: Good.

18.15: Good.

20.30: Good.



Day 84, 08.35: Good.

13.50: Good.

18.05: Flat.

20.00: Good.

23.20: Good.



Day 85, 11.00: Good.

13.20: Good.

16.30: Good.

19.10: Slightly flat.

22.25: Good.

A week or two ago I thought it might be time to conclude almost complete failure. Except where allowed a minute's (or more) silence. Thereby excluding any prospect of being able, which I would very much like to be able, to confirm a hypothesis about a pitch relation (such as a modulation) on the basis of a judgment of a subsequent pitch, and not be restricted to the converse justification, deducing a later pitch from knowledge of earlier pitches and judgements of intermediate relations (intervals and modulations). Having (on the contrary) to go into a trance stops it from being, regardless of accuracy, even a party trick, let alone useful.

But I now kick myself for not questioning the reliance on the single "target stimulus" (the Ravel). I suppose the restriction resulted from the complete novelty and possible hopelessness of the task: of trying to consistently disqualify 99 per cent of pitch-wise perfectly good transpositions of the music as true instances of the music.

Anyhow, I have belatedly noticed that certain other frequently invoked intro sounds on youtube (namely those of Honky-tonk Women and the music from Would I Lie To You) were manifesting as "images" with a vividness that I have been learning, slowly, to trust as an indication of pitch-truth. (Slowly because I was fairly expert in producing vivid images of the music that were true only relatively i.e. relationally. Hence the vividness is of a particular and new kind. Though not a Mary's Room level of new (see OP) as yet. :grin:)

Conveniently, both of these stimuli were in more or less the same key as the Ravel, making it feasible to collage their images swiftly. A lot more easily, for example, than collaging alternative pitchings of any one of them. Judging the absolute (as opposed to relational) pitch-truth of the whole collage caused a surprising and promising feeling of confirmation, as though triangulating from information at relatively distant points in the cortex. (As though.)

Which makes the prospect of judging an image in the midst of a sounding context (without recourse to an undisturbed silent trance) slightly less daunting. And I feel I'm starting to be able to resist automatically parsing musical pitches relationally, or at least to begin to expect to discern their absolute aspect. Haha, we shall see.

Day 86, 13.50: Good.

16.30: Good.

19.05: Flat.

22.41: Flat.



Day 87, 11.35: Flat.

13.55: Good.

18.08: Good.

21.05: Good.

23.05: Good.



Day 88, 14.00: Flat.

14.50: Sharp.

15.50: Good.

17.20: Good.

21.15: Good.

23.40: Good.



Day 89, 09.55: Astounding inertia in acting on the foregoing insight, belated as it was, and bothering to check against a different clip: the Stones HTW. Good, anyway.

14.40: Good. (Ravel)

18.30: Yikes, 2 semitones sharp, wtf?

23.10: Again!! (Both the Ravel)

00.45: Good.



Day 90, 14.30: Slightly flat.

22.30: Definitely flat.

Both of the mooted new test clips cause a small opening in the can of worms that is (so called and alleged) octave equivalence. Also they draw attention to the elephants in the room that are: the differential sensitivity to absolute pitches (to some more than others), and the differential sensitivity to a particular absolute pitch in different musical contexts. Second elephant first:

E.g. I just noticed that a first image of HTW seemed vivid enough to warrant testing, to see if the vividness happened to be on target. But a mere image of the Ravel at the same pitch was enough to conclude that the HTW image was roughly half an octave out. This fits with the assumption (see OP) that vividness had been previously unfettered by choice of pitch, and only now (through training) correlates with accuracy (absolute, non-relational), and perhaps (why not) only or mostly in the case of images of the test clip. That the skill would be strongest with the examples used for training. So, strong with the Ravel intro but not the HTW intro, no matter (because the vividness generally unfettered) how often one had recently played the latter.

Day 91, 13.15: Good.

17.55: Good.

20.50: Decided to find a fourth stimulus to try and perceive non-relationally. Well, another Stones clip that the algorithms had lately found me receptive of is Gimme Shelter. I must have been exposed to it upwards of 10 recent times and, as with HTW, I wondered whether the vividness of an image of it, that had now launched (or rather, landed - see below) immediately upon consideration of the tune as an option, might be a sign of its veracity pitch-wise... perhaps disconfirming yesterday's no doubt rash conclusion that the number of exposures is "no matter". This time I didn't attempt an estimation straight away but planted the Ravel opening into the midst of it and found it (the putative g4) a comfortable fit on step 5 of the Stones' key. I don't mean I tried alternative placings of it for comfort; rather, it seemed to land there. (A third elephant. Or second can of worms.)

So, turns out the GS image was precisely a semitone flat. But then, no wonder the Ravel fit so nicely. Placing it on step flattened-fifth would surely have been awkward. I even wonder whether the Ravel image (true, as it turned out) could have dragged down the GS image. It probably kicked up enough dust upon "landing" to obscure the shift. (Had the original GS image been also true.)

23.50: Good.



Day 92, 13.35: Good. Tried to overlay a GS on a Ravel; not quite sure what happened (too much dust), but then bringing the Ravel up a semitone seemed to allow the GS to land in what I'm now aware is the right place relative to the semitone-sharp-Ravel, i.e. a fourth up from it. And the Ravel turned out to be where it ought, i.e. truly a semitone up.

16.25: GS rather immediate and vivid. And true.

Elephant 3: the "landing" of images... implying a flight, from a launch. But the launch and the flight are invisible (inaudible). It used to be (and still is in the main) that the image landed (and played out) without delay, at an arbitrary pitch. Now (for the 3 or 4 test clips) there is a delay while waiting for the system as a whole to find the right place. I suppose the sense of flight and of landing results from the time-limit imposed by such props as: imagining reaching to push with a finger on the g4 of a keyboard. In order to aid recall. And from the frequency of erroneous results. One isn't (yet) prepared to wait indefinitely for an image on the expectation of it being true when it arrives. One assumes that the image will need weighing up and then adjusting. And one feels that only a time limit will (as yet) stimulate the unconscious background search that would make the first image any more reliable than chance.

I'm constantly struck by the comparison between this process (if it isn't a completely empty fantasy) and a more familiar effort of recall: that of finding the right word. (wts)

19.35: Good, or just noticeably flat. But again, based on a first image of GS, whose vividness again presses the question whether some sound fragments are perceived non-relationally more easily than others. With GS (the intro) we are talking about a decorative display of glides and glissandos as much as clearly defined pitches, and I wonder if such a pattern, being relatively poorly captured in notation, is more easily perceived non-relationally than, say, a piano pattern? (There could well be research on that. See previous excuses for ignorance.)

22.40: Almost surprised to find the first GS image not confirmed upon pitching a Ravel image a tri-tone down. The Ravel protested and wanted to be two semitones higher: subsequently verified. I was thinking the foregoing theory must be true and I'd missed an obvious trick. But maybe not. Or the test was flawed. Try again tomorrow.

Day 93, 11.25: Not impressed with a particularly vivid GS image immediately upon its consideration. So tried to launch a search, by way of passive waiting. Next image vivid enough to urge testing. Verified. I suppose that passivity required losing track of the relation to the first image. So that relation unknown. But nvm.

15.15: I think the sequence here was: first GS fairly vivid; implied Ravel image far too low (2 semitones); "waited" for second GS image, hopefully uninfluenced by Ravel judgement; result good.

17.00: Waited (a second or two) and was pleased enough with the first image (GS); and my pleasure (with the vividness) was validated as an indication of truth.

19.35: Noticed a GS image already playing as ear-worm. Seemed authentic enough, but the corresponding Ravel image seemed badly flat, by a fourth, in fact (duly verified). So this trial (if significant at all and probably not) then a point against the hypothesis of notatability (of a musical stimulus) varying inversely with non-relational pitch sensitivity.

21.35: As per 11.25.



Day 94, 10.10: As per yesterday at 15.15.

12.45: As per 11.25 yesterday, but result a semitone sharp.

16.00: Good. As per 17.00.

21.40: And again.



Day 95, 13.05: As per day 93 at 15.15.

Time to choose target stimulus no. 5. WILTY the panel show reminds me of the Charles and Eddie... will try that. The Ravel fits nicely... truly it turns out.

What we expect is that the normal propensity to produce images at an arbitrary pitch, which may or may not be a result of skill in relative pitch, will reduce, as the ability to produce them at the correct absolute pitch increases. Will that be a shame?! Do we expect it, actually?

16.05: Reasonably vivid. The Ravel found the image to be a semitone too high, though. Which was indeed the case.

Pasting the Ravel into other mental images is proving fairly easy. Must try more often to paste it into sounding ones.

18.30: Fairly vivid Ravel/C&E mashup: not sure which appeared first; anyway, verified.

20.45: Hmm... mashups not rushable: semitone sharp.

Day 96, 11.05: Borderline (Madonna) after calibrating last night... Just checked the image that happened to be present as ear worm... didn't gel well with the Ravel, and turned out a semitone flat. Good that the Ravel is proving reliable (ish) for calibrating other images.

13.20: Trying to launch several at once, with equal influence. This one probably unequal in favour of the Ravel. Anyway, good.

15.10: The several being, GS, the Ravel and Borderline, in that order of priority and influence. Good. GS was convincing straight off.

22.00: Good. I think first to land was Madonna.



Day 97, 13.30: Madonna present as ear worm, but a semitone flat according to the Ravel, which checked out.

15.25: Delighted (if not deluded) to sense the GS image conflicting with the others (being in a distant key) even before landing. Anyway, confirmed thereafter by the Ravel, which then checked out.

18.40: Starting with GS, I think. Anyway, good. Oh yes, I was thinking of trying Claire de lune as company for GS.

22.05: CDL not hugely anchoring... Relied on GS. But good.



Day 98, 12.30: See last.

15.15: Considerable want of anchorage from the Ravel. Had to relaunch it. Which has become a rare necessity. Especially, I speculate, since collaging it with others.

19.55: Madonna present as ear worm, again. True this time.

00.10: Good.

I have next to no sense of time. I was blown away when I found out that other people do. — frank

Do you cook? I discovered that cooking turned out to be a very good practice for gauging time.
Although mostly in a practical sense, not in terms of testing myself with a timer.

For example, when I make pancakes or crepes, it's not necessary to stand by the stove all the time, so I often do small chores, such as fold laundry. I developed a sense of whether I have enough time to fold another shirt or other item before needing to flip the pancake or crepe.
And then, of course, cooking a multicourse meal and serving it at the exact time. This is primarily about good organization, even more so when cooking dishes that must be served within some 10 minutes or less after being cooked.

I think it must be really hard to teach oneself to gauge time just sitting there and trying to gauge how much time has passed.

When I started trying to teach myself to guage time, like just starting with 10 minutes, I felt an overwhelming aversion to doing it.

I can link that up with other aspects of my personality where I cant handle being pigeon holed or caged in any way. I wonder if personality can influence the skills you have access to.

A love of learning is something to be learned. :)

Do you cook? — baker

Most of my kitchen gadgets are self timing. When I say no sense of time, its pretty close to that.

Day 99, 09.25: Surprised to find a crystal clear Madonna ear worm to be a fourth flat, upon fitting the Ravel.

Very tempted to regret wasting several weeks not thinking to triangulate. Although, that notion not supported by the peculiarity of the reliability of the Ravel. However, the (as it were) conspiracy of the several targets does at least appear to keep the options confined to steps on the modern (A-440) scale.

17.20: Good. I'd better make sure I have a couple of weeks without upset before regarding the Ravel on a par with an actual check.

18.50: Good.

21.25: Well, didn't take long... Tried to use the Ravel to get bearings in Bowie's Pretty Things: wasn't sure, but went with what turned out to be a semitone flat, probably through an ill-advised effort to get a comfortable fit key-wise.

What I forgot in my disappointment, however, is that this was a (nearly) first try at superimposing the Ravel image onto an actual, sounding image. So, not such a bad.

Most of my kitchen gadgets are self timing. When I say no sense of time, its pretty close to that. — frank

:( The perils of technology.

I've burned a lot of food and destroyed one copper bottom heirloom pot, so, I like the timers.

Just more reasons to learn to cook. Without a timer. :p

Day 101, ...Trust restored (above) in the Ravel/GS-centred repertoire as a reliable test of other mental images if not sounding ones, I reneged on the intention to keep testing that repertoire. Day 100 spent instead trying to parse the Bowie, from the chorus out. The key changes are typical of the kind I lose track of, and would hope eventually to trace efficiently by means of non-relational sensitivity. Too ambitious in this case, partly (possibly) because not heavily exposed to the track. Still working on it.

Day 112. Ok so I guess I have entry level absolute pitch: I can now calibrate musical 'calculations' with respect to a core of mental images (mainly, the Ravel and Gimme Shelter), without feeling obliged to refer to a higher authority i.e. a physical image; and while it's a long way (see day 99) from battle-ready for parsing actual music, I can begin to theorise, albeit introspectively, about the nature and behaviour of the imagery.

I.e... theorise about the nature and behaviour of the introspection, which is soaked in folk-psychological theory about imagery and introspection.

What I gradually embraced and reinforced (if grudgingly, and assuming a growing debt of needed clarification) was talk of an image as an internal mental event, of the kind typically conceived as coinciding with a neural event conceived, in turn, as a physical trace or recording or representation of some external music not actually present. More specifically, I associated pitch-wise veracity of an image, increasingly closely, with an intuition of the vividness or clarity or realism or authenticity or solidity or immediacy or effortlessness of the image. Immediacy in the sense of directness of acquaintance, or absence of noise: not in the sense of rapidity of formation; on the contrary I got in the habit (see day 92) of either waiting patiently for an arrival or being prepared to "launch" a large number of (individually quicker) "flights".

An empirical question, I expect, is whether this kind of association, of perceived clarity with pitch-wise veracity, is well-founded. This might be the case if clarity resulted specifically from a forensic, causal connection between stimulus and image, as with (according to naive folk-semiotics at least) the clarity of a photograph or a sound recording. And if so, we might ask whether either or both of the clarity and the veracity are available with respect to relative (relational) as well as absolute (non-relational) pitch sensitivity; and which available kinds of clarity and veracity are associated. How, further, the abilities and their associations may vary among differently (from most to very least) gifted or trained musicians.

For example, my hypothesis of a conflict between development of absolute and relative sensitivity: perhaps clarity is more pertinent to the first. Perhaps clarity of an image is a requirement for correct recall of a pattern by untrained listeners. And perhaps an image of Gimme Shelter (intro) is clearest when truest because of its exhibiting of non-notate-able (and hence less readily transposed) patterns.

Whatever the empirical or theoretical merits of such a view, I've indulged it, and acquired a degree of skill in facilitating spasms of neural activity as though... well, partly as though recalling a stored image to a viewing area, and checking it for authenticity... but partly also as though conjuring or manufacturing such an image in situ (on stage in the viewing area), by the authority of one possessed of absolute pitch, and then subjecting it to a similar evaluation. The "as though" is effected by a rather thorough visual running commentary (of the recalling or the manufacturing) which matches the sound images to visual ones from e.g. the relevant youtube visuals, or my finger approaching the g4 on a piano etc. I.e., a folk-psychological narrative of phenomenal sound events is maintained by a (narrative of a) visual narrative.

Either way (imagined as recall or manufacture), clarity of an image upon "viewing" (or rather, "auditing") has become sought after as an indication of its veracity with respect to absolute pitch. This has created a variety of distinct navigational predicaments:

(A) Stage empty: thoughts have turned to music, but no ear worms are present. Free and able to call up any image, probably the Ravel, and to reject and re-order if not completely satisfied. The re-ordering may express a preference up or down, or it may not. I thought I noticed a drift (with increasing skill) towards not; but possibly that aspect of the successive improvement had merely become quicker and less conscious. While unsatisfied, also free to,

(B) call up an image of different music, but must then expect that the process (whether of manufacture or selection from pre-pitched alternatives) is influenced by the pitch of at least the last image from (A or C) however unreliable that pitch. (B) repeatable like (A). Satisfaction during (A) or (B) may lead to,

(C) consolidation-cum-testing: try an image of different music (possibly returning to that of A or D, if here from B). If tending to the view that the image is recalled whole from storage, one might hope to allow it to land according to its own 'gravity'. The landing place not being as expected relative to the previous image would in that case mean dropping or re-launching (C again) one of the two, probably the first. But on the contrary view i.e. assuming the image's manufacture in situ to be guided by the growing skill in absolute pitch, one must assume that its correctness depends on that of the previous image. (The skill in pitching the current image can hardly be uninfluenced by the approval of the previous one.) So the current and previous images can't be evaluated for reliability independently of each other. On the other hand, neither are they acting entirely in concert. One of them may present an unclarity or instability that undermines the other. ...Badly, and go to (B). Apparent fit, on either view, and repeat (C). The weight of influence of a (possibly wrong) consensus then increases.

(D) Thoughts have turned to music, and found an ear worm active. Increasingly often, tempted to evaluate it for veracity, or even for the indication of it in vividness. (If not tempted, discard and go to B). The latter is an option at least if the ear worm happens to be one of the core. (Occasionally, an ear worm is actually a rapid sequence-of-Ravel-starts-as-calibration-attempt, haha. Not sure if that's a good or bad.) If not core, the image might yet be suspected of being significantly vivid, if it is music likely heard only in one key. As with (A), go to (B or C) depending on satisfaction.

(E) Reminded of (e.g. from reading about) a piece of music. Lately (never previously) construction of the image may well be interrupted and restarted in an effort to position it right.

(F) ... Sub-species of (C), impressed by the vividness of the images of more than one core fragment, but aware of the possibility of deepening error: or rather, the possibility of deepening trust entrenching the same error, such that the vividness might be caused by the pitching being relative to each previous image as much as by veracity of the present one. (The "wrong consensus", above.) Have been on occasion inclined by this awareness to interrogate one or more of the images for signs of deviance, but gratified instead by a spontaneous correction: presentation of an image differently pitched and apparently uninfluenced by the prevailing consensus.

This inspires me to see what frequencies I imagine when I sing Give Me Shelter in my mind. I found I had to practice to even duplicate it with my voice, but I do seem to come back to the same starting frequency.

I think according to the Private language argument, u shouldn't be able to do this. Do you know what I mean?

I found I had to practice to even duplicate it with my voice, — frank

Well this inspires me to actually getting around to browsing empirical research, rather than simply saying, in this case: doesn't everyone experience musical "ear worms"?

...

Ok, the wiki page says yes, everyone (98%) does. So the next question is, do we have, in your case, a syndrome as rare and curious as that of aphantasia, or a simpler misunderstanding? Or, not simpler, but par for the course in phenomenological discussion! ... E.g. I wonder what degree of clarity (reality? hallucination?) of imagery you are supposing is involved in "duplication"?

I do seem to come back to the same starting frequency. — frank

Ok. You checked against a recording? A few times, not too close together? I'm tempted (as already mentioned) to trawl the research for reference to the obvious question of the variance of most people's ear worms (or spontaneous performances) from the usual pitch (e.g. the pitch of a recording where relevant). But there was no mention of this on the ear worms wiki page.

Btw @Metaphysician Undercover would be interested in your involving vocalising in the aiming for the pitch.

Or do you just mean you maintain whatever random starting pitch for the duration of the performance? (Which is a thing, that by no means always happens.)

I think according to the Private language argument, u shouldn't be able to do this. Do you know what I mean? — frank

Haha, is it about ear beetles? ... Not sure. Good question.

Far as I can tell, no... or, not so as the ear beetles would matter, or cause an unacceptable philosophical scene, by refusing to 'drop out of the equation', at least in principle, and give way to corresponding external stimulus types.

To be fair, the creatures haven't exactly been (as Witty's figure has it) 'cancelled out' from my 'arithmetic', here. I'm certainly talking the talk of internal sensations (phenomenal qualia, mental images). I guess this is largely because I'm talking about non-perceptual imagery. But Witty doesn't seem to regard either the presence or absence of an external stimulus as characteristic of the problematic scenario critiqued as "private language". The problem inherent in that scenario is (I think I gather) the lack of criteria for identifying and classifying the internal entities, should these be (as folk-psychology often leads them to be) conceived as the sole and sufficient basis for their own classification. Where such criteria are available, because the entities are conceived instead as part of a larger, pre-existing game (e.g. there is meaningful comparison with blood pressure, or actual train times), no immediate critique is offered.

So I don't think that your returning

back to the same starting frequency. — frank

is a problem for Witty, because you are judging the pitch-identity (vs. difference) of the earlier and later 'internals' according to a pre-existing system that (if it recognises them as internal) maps them to externals that are identified or distinguished according to (an ordering correlating roughly with) frequency. So he wouldn't see that judgement as a problem. He would doubt that the imagery-talk is true literally, but he wouldn't be sceptical about the viability of pitch comparisons among internal images...

I found I had to practice to even duplicate it with my voice, — frank

...nor between images and actual sounds. And neither would I. So I'm still curious as to the nature of the difficulty you describe.

@frank's thread.

Predicament (G): The spontaneous correction reported in (F) has, pleasingly, become voluntary; or has, at least, coincided with a marked improvement in the efficacy of (B), so that I can expect to recalibrate and regain certainty about my location sometimes as readily as by tooting or plucking on a nearby instrument... but more usually demonstrating the continued necessity of "going into a bit of a trance". Anyway, generally, a much quicker trance.

(H) Another pleasant surprise: although the trance still requires silence in order that the image should land without being thrown off course, landing in occupied territory is easier than expected. Quite probably off target, but with the feasibility of assessing the matter. Off target probably as a result of achieving a musical fit of the core/standard image (usually the Ravel) with the occupied territory. Identifying the g4 itself to a diatonic step in the occupied territory (the sounding music), for example. Probably also not just any diatonic step, but one that allows further agreement between other steps. But, as I say, the degree (or at least the fact) of deviation from the correct target feels (and seems upon verifying to be) discernible, despite all the noise. So I'm able to have a go at guessing the key of music on the radio, say. I should report on this more systematically, or more often, now that I'm not filing reports on (because rarely verifying) the acts of silent calibration.

(I) The faintest glimmer of hope of being able (in silent mode) to recalibrate (i.e. B) without losing entirely the previous image: being able therefore to re-land that previous image (at the same pitch) in the recalibrated context so that I can see what pitch it really (according to the new context) was. Then I may be able to monitor the (non-relative kind of) accuracy of random earworms (D).