Generalization in measurements of plausibility for objects other than paradoxes

As many texts contain no paradoxes, our idea is to widen the type of measurement presented hitherto so as to be able to appraise certain interpretations of them numerically. To achieve this nearly all the conceptions previously put forward will have to be generalized and a re-examination carried out of notions initially set aside for fear of the presence of some irrationality. The notion of support was provisionally abandoned because of the risk of attributing to the author a meaning invented by the interpreter but now we may say that a trace constitutes a term, or the meaning of a very special support: if a support suggests it has a definite meaning, a meaning known to exist without its exact nature being clear, it is said to possess a pivot. Terms and pivots together form traces. We will still use the word “term” for any idea that could be part of an arbitration, even though, when used, the idea is part of a link other than an arbitration. When it is understood that in effect a support has a pivot since the creator obviously wanted to indicate a particular meaning, this meaning often still remains to be found. On the other hand if we are not sure whether a support has a meaning, it is not capable of representing a pivot.

We will extend the use of roman capitals to symbolize any element or elements taken from the text. A gloss links a brief commentary and a trace or consists of a connection, without repetition, between traces that the critics cannot reject absolutely since it never comes into conflict with any buffer. Often, a gloss can be represented as b(A~E) or d(A~E) where A and E are traces, (b) marks affirmation while (d) is the symbol of negation. For b(A~E) the significance of the gloss is “to understand the text as intended by its creator at the origin A and E should be united rather than opposed”. With d(A~E) the meaning becomes “to understand the text as originally intended by its creator A and E should be opposed rather than united”. There are three types of gloss: the problem gloss, the attenuation gloss and the neutral gloss. The gradient is the numerically situated plausibility of a neutral or problem gloss, as well as that of a pair of attenuation glosses which may be composed of the same one twice.

A problem gloss has to provide the description of an obstacle which hinders the comprehension of a text, in a serious way in the case of a collision, but also in a benign way in other cases; however indistinctness taken to extremes is never in itself acceptable, so that b(vast~unity) cannot be admitted. On the other hand the expression «…the day of swift feet…» used by Pindar to speak ambiguously of the day of the race when the runners match their strengths in Olympia can be described by b(day~feet) as a difficulty but not a serious one because it is obvious from the context that praise of an athlete is the purpose of the poem [720]. In the same way glosses have the advantage over collisions of applying both to the slightest difficulties and to the most serious. The idea that a “god-day” is advancing with a long stride, using a body larger than man’s, remains unobtrusive so that the stylistic effect is given as such and can be represented by a gloss which is not a collision.

An attenuation gloss must comment on a problem gloss, even if it sometimes unfortunately verges on the irrational in so doing; it cannot therefore oppose the text nor become so vague as to cast no further light on the text. If two attenuating glosses are found in the same interpretation of the work being commented on, they cannot be contradictory in the long run; however it is possible for one to prevent the other from being unclear as it might have been on its own. Besides this it can be seen that there is a certain rivalry in content between an attenuation and the problem it deals with. Yet the attenuating gloss never resolves anything definitively, being limited to explaining one aspect of the difficulty faced in the case of problems which are not collisions, and to avoiding the impression of complete absurdity for these. The barrier is the model for attenuation but an attenuation gloss does not necessarily form one. Nevertheless, as the join is the type for the clarification of problems in the case of texts where imagination dominates, the plausibilities of the barriers are added together to find the gradient, or the plausibility of an isolated attenuation is doubled to find its gradient. The measurement of the plausibility of the whole meaning clarification operation is shown in the product of numerical values: by multiplying the problem gradient by the attenuation gradient. Indeed a good idea plus its application to a barely perceptible difficulty together are of doubtful interest.

A neutral gloss obtains a good gradient when the ideas in it do not enable a problem or attenuation gloss to be formed easily. In this way b(long~echoes) or b(flesh~children) constitute sound neutral glosses, approaching what are commonly called statements.

In order to facilitate the identification of gloss formulae we can place an (r) in front of problem ones, a (v) in front of attenuation ones and an (o) in front of neutral ones. Glosses can only be considered to reinforce each other if they belong to the same type, the (r) type, the (v) type or the (o) type. By means of a device, the rivet, no gradient will be counted if, without being subsequently multiplied by any other number, it has a value of less than 0.062 or 1/(2)(2)(2)(2). In spite of this precaution, the present analysis can in no way be considered as a calculation of probabilities since the measurements may exceed 1. For this reason this value of 1 must be restored in practice by what we will call the leveller in the case of any result exceeding that number. The calculation of probabilities will nevertheless remain in theory considerably greater than all the present calculation of plausibility, even using such expedients.

If a gloss proves to be a broad tension, its gradient is not exactly its channel, except in the case that this gloss is a collision. The gradient for a non-collision gloss enables any weaknesses in plausibility to be felt more strongly than the channel can do with collisions whose clarity provides greater guarantees of the existence of a relationship in the mind of the author. All gradients contain seven quantities q, e, p, f, z, g, j, the components; the last four serve to generalize the ideas used for factors t, s, m, w of the vacillation; the first two (q), (e), enable confidence in any possible pivots to be measured; and the third is responsible for replacing the selection effected from among the tensions by the strict definition of the collisions.

The circuit of a gloss consists of its gradient plus all the values resulting from its reinforcement. The carousel can be defined as any part of the circuit due to a gradient and a single reinforcement value. However in both cases the calculation cannot be taken to its natural conclusion if it is likely to give a result greater than 1 because of the necessary levelling. The carousel for gradients (h) and (h’), for two glosses, one reinforced and the other reinforcing, amounts to h+(h’/k) with the quantity (k), the exterior spacing, being the means of evaluating the level of separation between the traces. For those attenuation glosses that are taken in pairs, circuits and carousels follow what happens to the gradients and so pairs of sums must be considered. Instead of h+(h’/k) we must count h+((i/k)+(i’/k’)).

When a gloss is deprived of affirmation or negation, the result is a spit, which is reduced to a framework if the gloss consists of an arbitration. The collisions form glosses but in order to insist on the fact that the calculations to be carried out concerning their plausibility are not the same as those for glosses, the sign (–) can still be used for these shocks of meaning while calling them glosses for convenience.

For a spit (A~E) the shelving consists of the six glosses that can be formed on it or the ((2)(3)) elements obtained with the combination of the two options, (b) and (d), and three types of gloss (r), (v) and (o): rb(A~E), rd(A~E), vb(A~E), vd(A~E), ob(A~E), and od(A~E) have to be counted. The sum of the gradients for a shelving should be 1, but the imperfections of the calculation often result in a lower or higher figure. The leveller rectifies this in the latter case, but the former situation remains without reparation.

A plateau is an interpretation formed solely of glosses, and the ell is the value of its plausibility, calculated using the gradients relative to its constituent elements. If any of its glosses are attenuation ones, it must be accompanied by the gloss of the problem thus being dealt with to reach a more comprehensive perspective than that of the gradient. Sometimes it is appropriate to add certain values, as with h+(h’/k): a carousel such as this can measure a plateau on its own if the glosses remain problem or neutral ones. Nevertheless apart from a situation of reinforcement, the values will be multiplied rather, for instance when there are two glosses, a problem one and an attenuation one, or two glosses without any special relationship. Applying the levelling principle, the ell will be limited to 1 as regards the appreciation of its greatest strength, but the rivet does not apply to it as it concerns only gradients.

When we have a problem gloss with the gradient 1/qepfzgj and a attenuation gloss tested out on it, with a gradient 2(1/q’e’p’f’z’g’j’), the exterior spacing (k) measures the distance from one to the other in 2(1/q’e’p’f’z’kg’j’). For the plateau comprising just the two glosses concerned, the ell is (1/qepfzgj) (2(1/q’e’p’f’z’kg’j’)) and with two attenuation glosses having a sum gradient, the latter amounts to (1/q’e’p’f’z’kg’j’)+(1/q’’e’’p’’f’’z’’k’g’’j’’) and the ell is (1/qepfzgj)((1/q’e’p’f’z’kg’j’)+(1/q’’e’’p’’f’’z’’k’g’’j’’)).

The fact that one collision may be capable of reinforcing the effect of another, does not in any way prevent the ideas contained in it from attenuating this same neighbouring collision. Notches for example serve to clarify certain clashes in meaning, but such power does not stop them from drawing to these shocks enough attention to strengthen them. To understand this, it is essential to avoid confusing ideas concerning superficial influences with the realms of implication and insinuation. One relationship may only occupy the mind in a secondary way while another, contrasting slightly with it, is immediately clear. An unexpected path may thus be taken at times resulting from the complicated arrangement of meaning: first only the shocking aspects of the collisions are thought of and then the possibility of one clarifying the other is envisaged.

Using a device that we will call a cut, instead of counting the same gradient twice in a plateau, because of the double use of an attenuation, for example once with one problem and then with another, it will be counted only once when calculating an ell, if the reasons for fixing the constituents remain the same.

The exterior spacing (k) is used with the components, in particular to measure either a reinforcement (1/qepfzgj)+(1/q’e’p’f’z’kg’j’), abbreviated to h+(h’/k),or a plateau containing just a problem gloss and an attenuating one (1/qepfzgj)(2(1/q’e’p’f’z’kg’j’)) giving h(2(i/k)). Finally for these two usages together we have h((2i/k)+2(i’/k’k’’)), or (1/qepfzgj)((2(1/q’e’p’f’z’kg’j’)+2(1/q’’e’’p’’f’’z’’k’k’’g’’j’’))). In the latter situation (k) is the exterior spacing between an attenuating and a problem gloss, while k’ plays a similar role but for the second attenuating gloss, and finally k’’ gives the distance this time between the two attenuating glosses, the one reinforcing the other.

As a contradiction within a plateau is unacceptable, this leads us to rule out of the study any texts with a blatant inconsistency, which in any case seem very rare whenever imagination predominates since the relationships between significances are too ambiguous for mistakes in logic to be apparent. What is more, all opposition to the text, such as in b(cool~others), will be avoided. Equally we will reject the near circle that could occur for example if, in order to compensate for any feeling of absurdity arising from rb(living– pillars) we were to have recourse to vb(forests–pillars) vb(living–forests) to then try to obtain an attenuation of rb(forests–pillars) using vb(living–pillars) vb(forests–living). Once (~) has been adopted in the various formulae which can be employed here, the series becomes rb(living~pillars) vb(forests~pillars) vb(living~ forests), then rb(forests~pillars) vb(living~pillars) vb(forests~living), whereas it is very awkward that the problem of living pillars should be explained by the notion of "forest-pillars" which is itself clarified by the idea of living pillars. However it is only within a same plateau that such a device, which can be summed up as rb(A~E) vb(F~E) rb(F~E) vb(A~E) or more generally as rb(A~E) vb(F~H) rb(F~H) vb(A~E), is excluded.

Not only can we reject the idea that a plateau can contain at the same time rb(A~E) and rd(A~E), or vb(A~E) and vd(A~E), or ob(A~E) and od(A~E), but it must not include rb(A~E) and vd(A~E) for example, nor generally b(A~E) with d(A~E) whatever the type of gloss concerned, in r, v, or o. Furthermore, if ob(A~E) belongs to a plateau, the two glosses rb(A~E) and vb(A~E) are excluded from it. In the same way, if od(A~E) is in the plateau, the glosses rd(A~E) and vd(A~E) remain outside it.

The supposition that intuition and calculation of plausibility proceed along the same lines must be extended to the field of glosses in the widest sense, having been applicable only to collisions up to paragraph 80. This leads to the acceptance of the idea that the larger the gradients or ells, the more intuition will favour the gloss or plateau whose measurement of plausibility has thus been made. In order to test the accuracy of such a view without muddling the ideas, the scope of the vocabulary for the next study of networks and collisions will be widened by adding further words. With this in mind, we can say that a modification of the text constitutes a porterage if it is restricted to that which is to be measured. The text created by the author with all its related measurements forms the hoard, while the porterage, in comparison with this, plus the related degrees of plausibility, forms the dispenser. To write formulae equally valid in both cases, square brackets will be used, reverting to normal brackets in the case of either a dispenser or a hoard on their own. Moreover it will be allowable in certain cases to contrast two modified texts, with their calculations, each group being called a cover. The formulae for these will be written in square brackets.

The components (q) and (e), the carvers, (q) for the left-hand side of the gloss and e for the right, are 1 when their reference trace is a term. In a broad tension with two terms and therefore no pivot, q=e=1 in all cases. When a meaning of the text which has never been in a collision, carries out the function of a pivot in accordance with the requirements of the context, the carver dealing with it also has the value 1. This, for example, does not concern b(N~Nature) since «temple» which would be a very proper meaning to explain this sort of absolute evoked by the N, belongs to the collision b(Nature~temple). When a trace is neither a term, nor possesses any equivalent free of any collision within the text itself, the carver attributable to it equals 2.

An attenuation gloss depends to a great extent on the problem gloss whose difficulty it is intended to deal with; for this reason, we see the quantities p, k, g, j related to the same attenuation gloss vary according to the different problems it can tackle. In order to clarify such a situation, the use of an attenuation target, which occurs when only the best effect of the attenuation, or the best effects of it available at any one time, are examined, must be distinguished from the use of the sieve, which depends on a specific problem gloss. Both usages coincide when referring to the best, or one of the best problems dealt with by the attenuation gloss for which the measurement of plausibility is made. The difficulty sometimes encountered is that using a target, one problem gives k=1 but g=2, while another gives the opposite, so that overall the value of the products qepfzkgj is the same but it has not been reached in the same way. It must then be looked into in detail.

The function of the component (p) or pole is to test whether a buffer exists for a non-collision gloss regarding its particular characteristic, that is problem, attenuation or neutral. On the contrary, if a gloss constitutes a collision (p) is equal to 1. As for any attempt to describe a problem which is too slight to have the status of a collision but which a buffer leads us to consider as a difficulty, we also concede that (p) remains equal to 1. The text itself may note an obstacle to thought, in particular by a question, directing in this way the interpreter to p=1 for a problem gloss. Thus “Can various echoes become entirely confused?” allows p=1 with rb(confused~echoes). For an attenuation gloss, 1 is only obtained for (p) if the text clearly specifies that a relationship of attenuation exists between the passages giving it its basic meaning and the problem it deals with. In this way “Can various echoes become entirely confused? From afar yes…” would give p=1 with vb(afar~confused) for the problem rb(confused~echoes). Finally, if the text clearly states that an idea concerns neither a problem nor a clarification, p=1 for the neutral gloss describing this idea. The difficulty in what a text states arises from the irony that at times accompanies the simplest meaning and so, as for problem glosses which are not collisions, a buffer must be present to achieve p=1 for neutral and attenuation glosses.

For broad tensions the peg (f) is fixed according to the rank (t): if t=1 then f=1; if t=2 then f=2. As for the various glosses that have no channel, their formulae being b(A~E) or d(A~E), the peg is 1 for (b) if an internal buffer exists in the text which means the same spit with a (d) cannot be written, and for (d) if there is a buffer prohibiting the spit from having a (b) before it. In any other case f=2.

Slides (g) and (j) are equal for a neutral gloss. They have a value of 1 only when the passage of the text serving as a basis for the gloss does not allow any countable problem or attenuation gradient to be shown. Otherwise g=j=2. For broad tensions (g) and (j) are the same as (m) and (w) in the vacillation: the oscillations. In the case of the other types of gloss describing a problem, let us see what occurs regarding rb(A~E) or rd(A~E): if, in spite of the name of the gloss, the first level of meaning suggests that there is no obstacle to understanding within the contents, this means that no serious problem is raised there and as a result the two slides are equal to 2; in any other situation g=1 and j=2 or g=2 and j=1, but always one of the slides is equal to 2 because it can be seen that no broad tension is formed and this must be described by the calculation.

For an attenuation gloss, the slides (g) and (j) are equal, both together being l, or both 2. They are 1 if the attenuation has some aid, however vague, within the text, and this will be called a stratum. This favourable indication may attain the level of a buffer but not in the case of the attenuation of a broad tension where it is not possible for a good stratum to exist since any oddity would then be excluded. The absence of stratum gives g=j=2, which implies a great risk of inventing the hoped-for attenuation when interpreting the problem itself.

The interior spacing (z) in the case of broad tensions follows the interior remoteness (s), one of the factors of the vacillation, since tsmw is strictly identical to fzgj applied to these problems. As far as the other glosses are concerned, the situation is somewhat similar, with merely a few counting details to be added to the principles already known. When a buffer implies justification of an interpretation which supposes that a creator has established within his text a clear link between A and E, letting us think b(A~E) or d(A~E), it does not matter which, then z=1 for all the (A~E) spit glosses; otherwise (z) varies by 1.

When (z) varies by 1 for a spit with terms, it is 2+(1(n/10)) with (n) being the numbers of fronts in the text from one term to the other. If the traces do not have the form of terms the calculation changes very little even though one or other appears to be a pivot attached to part of a word. Any compartment participating in any way in a support with a pivot is called a fold, which is taken as a front. It is therefore sufficient to count meticulously the remaining (n) fronts between the traces to get 2+(1(n/10)). In the case of “Nature Is a temple. Pillars let forth at times confused words…” and b(I~P) “Is” and “Pillars” must be avoided, which gives 2+(1(1/10)) exactly like 2+(1(1/10)) with b(is~Pillars) for “Nature is a temple. Pillars let forth at times confused words…” As regards the interaction of tones “La Nature tord mes tortures en m'affligeant de maux” (Nature twists my tortures by afflicting me with evils) and the gloss b(ture tord~tortures), (z) will give 2+(1(1/10)) since “Nature”, “tord”, and “tortures” will not be concerned. Counting will be different in the case of a pivot which is separate from a word. Regarding b(colour of paper~verts) (colour of paper-green) the use of green paper for the whole sonnet would give 2+(1(74/10)) since there are 74 fronts in the text. On the other hand, with a green mark only on “verts” (green) and the new gloss b(colour of paper~verts) (colour of paper-green), z, instead of being 2+(1(74/10))=9.4, will be more modestly 2+(1(0/10))=2.

The exterior spacing (k) is used to calculate the attenuation and reinforcement plausibilities. For broad tensions, (k) is reduced to (c), and for the other glosses a similar procedure is used: firstly, with two glosses of any kind k=1 when an interior spacing (z) of 1 exists for each, and when, from among their traces, some being possibly the same ones, A, E, F , H forming b(A~E) with b(F~H) or d(A~E) with d(F~H), there is a continuity of intention guaranteed by a buffer. In this way rb(N~Nature) has a relationship with vb(temple~man) which allows k=1. The basis for this in the poem may be understood afterwards in the following way: man, as a poet, or in general, would honour the supreme temple.

When the exterior spacing varies by 1 and the traces are terms, it is equal to 2+(1(n/10)) with (n) being the number of fronts between the terms furthest from each other in the glosses in question. For reinforcements as well as anything concerned with attenuations, the counting procedure therefore proves to be very similar to that used for (c) the exterior remoteness.

An exterior spacing differing by 1, if there are one, two, three or four pivots in the glosses examined, is sometimes found when the calculation is limited by two folds or by one fold accompanied by another front. The device used with (z) is reused for the distance between the traces concerning (A~E) and (F~H), that is for A and F, or A and H, or E and F, or E and H.

If a notch is used in an attenuation with another gloss, this gloss could sometimes also attenuate the notch and so run the risk of causing a mistake to be made in the various exterior spacings. Therefore concerning the clarification of rb(corrupt~temple), the relation vb(corrupt~incense) which was previously called rvb(corrupt–incense) as a notch, is used. In the device rb(corrupt~temple) vb(incense~temple) rvb(corrupt~incense) vb(senses~incense) vb(corrupt~senses), it is best to keep rb(corrupt~temple) vb(corrupt~incense) vb(incense~senses) to measure an attenuation plateau which is certainly less perfect than the collision with the canvas and the pincer of the notch, but gives the substance of it with two sum gradient attenuations and no attenuation reinforcement at all. The possibility of confusing the problem gloss with the attenuation one would mean that if the calculations were made too hastily the spacing relative to vb(corrupt~incense) vb(incense~senses) might be taken for the spacing of rb(corrupt~temple) vb(incense~ senses). It should be noted that the two exterior spacings in rb(corrupt~temple) vb(corrupt~incense) vb(incense~senses) proceed from the attenuations towards the single problem without taking into account the relationship between the attenuations.

Let us calculate the problem gradient of rb(temple~Correspondences) It is not a tension but a slight problem of interpretation; the author appears to have put together the notions of sounds, colours and perfumes and the idea given by the title, but he might not have related this latter principally to «temple», so let us consider the degree of plausibility in the idea that he has thought of this relationship as being awkward. With 1/qepfzgj, we have 1/(1)(1)(2)(2)(2.2)(2)(1)=0.056 which is insignificant. Since here the traces are terms, q=e=1 and as the sonnet is silent on the difficulty mentioned, p=2. The title is isolated from the first line so f=2. The distance is 2 fronts, «Nature» and «is», giving z=2+(1(2/10))=2.2 and finally, since nothing in the poem goes against the principle that an obstacle exists, certainly not «answer», «symbols» or «words», ((g)(j))=2 with ((g)(j))=(2)(1)=2 if we choose to give 1 for the isolated term «Correspondences».

Let us examine the gradient obtained by a problem gloss with pivot: rb(N~Nature). As «Nature» is a term, e=1, and since N has no clear meaning represented by any term impossible to find in a collision, the other carver (q) is equal to 2. There is no explicit indication in the text of any difficulty described by rb(N~ Nature), which gives a value of 2 to the pole (p). The component (f) is 1 because there can be no objection to the relationship “N-Nature”. As for (z), it also is 1 due to this very strong link. Then ((g)(j))=2 as admittedly the problem is likely to be posed, but does not have the strength of a collision, and furthermore with N being a difficult enough point for (g) to be 1, the value j=2 becomes inevitable for ((g)(j))=2. Therefore the gradient is 1/(2)(1)(2)(1)(1)(1)(2)=⅛ which is double 1/16, the limit below which it can be disregarded.

The gloss vb(incense~senses) attenuating rb(corrupt~incense) has a gradient with q=e=1 owing to the two terms; p=2 since no explicit clarification of the meaning of the corruption in question is given in the text; f=2 because the clear relationship which is established between «incense» and «transports» has no equivalent between «incense» and «senses»; z=2.4 from the fronts «which», «sing», «transports» and «mind»; g=j=1 as «transports» serves as a stratum, supplying the notion of intoxication and so also that of sensuality. Altogether 2(1/qepfzgj)=2(1/(1)(1)(2)(2)(2.4)(1)(1)=1/4.8=0.208, a result which seems tenable.

Let us measure the gradient obtained by vb(N~Nature) which will be perceived in the relationship with rb(answer~Nature). The initial components have already been seen: q=2 and e=1. In the same way f=z=1; (p), (g) and (j) remain to be defined. The notion of a temple provides a convenient stratum to justify the values g=j=1. Finally as nowhere does the text claim to clarify itself, p=2. The result is 2(1/qepfzgj)= 2(1/ (2)(1)(2)(1)(1)(1)(1))=2(¼)=½=0.5.

The gradient of ob(homme~parfums¹) (man-Perfumes), with «parfums» taken from the eighth line remains negligible. The value of the carvers is 1; since no declaration is made in the text on any attempt to present a pure statement, p=2; the relationship between the traces is minimal so f=2; z=4.6 because there are 26 fronts between the traces. As the principal meaning of the poem concerns the natural world, there is a problem in linking «parfums» to «homme» rather than to «Nature», so that g=j=2; in order to secure g=j=1, the gloss would have had to be totally lacking in any difficult or enlightening nature: this would really have justified its neutrality. The gradient is thus 1/qepfzgj=1/(1)(1)(2)(2)(4.6)(2)(2) which is smaller than 1/ (2)(2)(2)(2) and so considered impossible to be taken seriously.

Let us find the gradient of ob(N~Nature) since we already know that rb(N~Nature) has a gradient of ⅛ and we suspect it might be helpful to establish vb(N~Nature) in view of several problems. For ob(N~Nature) the result is 1/qepfzgj=1/(2)(1)(2)(1)(1)(2)(2) as N gives us q=2; «Nature» e=1; the text’s silence p=2; the clearness of the link f=z=1; the possibility of problems with N, g=j=2. In this way the threshold of insignificance 1/16 is in no way surmounted here so that vb(N~Nature) seems to reserve some interesting aspect.

The reinforcement of a neutral or problem gloss by another often presupposes a device which we will call the crosspiece. For this the glosses must be of the same type, with the symbol (r) or (o), must have formulae with the same signs for the judgement, (b) or (d), have a common trace and finally have no contradictory significances.

Two attenuation glosses with gradients obtained through multiplication by two reinforce each other when they deal in a calculable way with any identical problem, have the same sign, (b) or (d), possess a common trace and contain no conflict of meaning between each other or between them and the text.

For the reinforcement of glosses with sum gradients no repetition of spits is admissible from one pair to the other, so that vb(Perfumes~sweet) vb(colours~sweet) has no measurable help from vb(Perfumes~ sweet) vb(sounds~sweet) because of the repetition of (Perfumes~sweet). On every other point the principles governing glosses with gradients obtained through multiplication by two are largely applicable. For example vb(Nature~Perfumes), vb(Perfumes~temple) directed to deal with rb(Nature~temple) must attenuate it in a calculable way like vb(Nature~colours), vb(colours~temple) which reinforce them. The two pairs can neither oppose the text nor undermine each other. Finally for glosses considered in twos, here vb(Nature~Perfumes) vb(Nature~colours) on the one hand and vb(Perfumes~temple) vb(colours~temple) on the other, a reinforcement must have taken place on an intuitive level and there must be a common trace. More generally we will use "x" as a symbol of “b or d, always the same one”, and then "y" for “b or d, whether or not it is identical to x, as long as it remains constant once in use”. Then on both sides with vx(A~E) vy(F~L) and vx(A ~H) vy(F~R) a meaning is necessary which allows vx(A~E) and vx(A~H) to reinforce each other while vy(F~L) vy(F~R) do the same.

Let us try to calculate two reinforcements for the glosses rb(rich~amber), rb(triumphant~amber). The gradients 1/qepfzgj and 1/q’e’p’f’z’g’j’ lead to the carousels (1/qepfzgj)+(1/q’e’p’f’z’kg’j’) and (1/q’e’p’f’z’g’j’)+ (1/qepfzkgj). Fortunately the two gradients have the same form with q=q’=e=e’=1 owing to the terms; p=p’=2 because the text is silent on these problems; f=f’=z=z’=1 as a result of the strong grammatical links; finally g=g’=j=j’=2 as no problem appears to be decisively raised, «triumphant» and «rich» easily passing as figurative terms from the creator’s point of view. The components (1), (1), (2), (1), (1), (2), (2) provide the necessary for the same gradient to be given to both glosses: 1/(1)(1)(2)(1)(1)(2)(2)=⅛=0.125. The exterior spacing (k) equals 1 since there appears to be a complete continuity of intention linking the four terms. The identical carousels are as follows: 1/(1)(1)(2)(1)(1)(2)(2)+1/(1)(1)(2)(1)(1)(1)(2)(2)=⅛+⅛=¼.

Let us examine the reinforcements of ob(N~Nature) and ob(N~temple). Their gradients 1/q¹e¹p¹f¹z¹g¹ j¹ and 1/q¹’e¹’p¹’f¹’z¹’g¹’j¹’ are identical: 1/(2)(1)(2)(1)(1)(2)(2). However since the rivet means these two values are negligible, we will not discuss them very long. We can see that q¹=q¹’=2 from the pivot; e¹=e¹’=1 from the term; p¹=p¹’=2 from lack of mention; f¹=f¹’=z¹=z¹’=1 from the textual link; g¹=g¹’=j¹=j¹’=2 from the possibility of substantial (v) glosses. Each of the carousels will obtain a plausibility of ((1/16)+(1/(16) (1)))=2(1/16)=⅛ with the quantity k¹=1. Let us modify the text, in order to increase the values, by imagining it begins with “Reality or Nature is the absolute comprising even, if one considers it well, every temple through which man passes…” The term “absolute” justifies the N and in no way belongs to an identifiable collision, giving us 1 for the components q² and q²’. The result is to increase the gradient common to both expressions to ⅛ by multiplying 1/16 by 2. The carousels are not bothered by (k²) since the requirements of grammar make it 1 and thus they are ⅛+⅛=¼. If we expand this we have (1/q²e²p²f²z²g²j²) +(1/q²’e²’p²’f²’z²’k²g²’j²’) and (1/q²’e²’p²’f²’z²’g²’j²’)+(1/q²e²p²f²z²k²g²j²) with (1/(1)(1)(2)(1)(1)(2) (2))+(1/(1)(1)(2) (1)(1)(1)(2)(2))=¼ here for both.

The reinforcement of vb(Nature~parfums¹) vb(parfums¹~temple) by vb(Nature~couleurs) vb(couleurs ~temple) in view of the collision rb(Nature~temple) which is to be attenuated, is composed of that provided for vb(Nature~parfums¹) from vb(Nature~couleurs), plus the help given to vb(parfums¹~temple) from vb(couleurs~temple). (The reinforcement of vb(Nature~Perfumes) vb(Perfumes~temple) by vb(Nature~ colours) vb(colours~temple) in view of the collision rb(Nature~temple) which is to be attenuated, is composed of that provided for vb(Nature~Perfumes) from vb(Nature~colours), plus the help given to vb(Perfumes~temple) from vb(colours~temple).) The plausibility of vb(Nature~parfums¹) (Nature-Perfumes) is 1/qepfzgj=1/(1)(1)(2)(2)(5.7)(1)(1)=0.043 with q=e=1 because of the terms; p=2 since the text makes absolutely no declaration on the commentary; f=2 as the link between the terms is weak; z=5.7 from the 37 fronts between «Nature» and “parfums¹” (Perfumes); g=j=1 owing to «répondent» (answer) which suggests the faithful for the perfumes and so guarantees the clarification given by the link “parfums¹-répondent- Nature-temple” (Perfumes-answer-Nature-tempel). Related justifications present us with an amount of 1/(1) (1)(2)(2)(5.8)(1)(1)=0.043 for vb(Nature~couleurs) (Nature-colours). The result for vb(parfums¹~temple) (Perfumes-temple) is very close to this at 1/(1)(1)(2)(2)(5.5)(1)(1)=0.045. In the same way vb(couleurs~ temple) (colours-temple) gives 1/(1)(1)(2)(2)(5.6)(1)(1) or 0.044. As we have sum gradients here, to find out what they are we must add the values from vb(Nature~parfums¹) and vb(parfums¹~temple) (Nature- Perfumes, Perfumes-temple) for example. The two gradients are thus (0.043+0.045) and (0.043+0.044), that is respectively 0.088 and 0.087. However, when the reinforcement is to be measured, the two exterior spacings are essential. The calculation is thus ((0.043+0.045)+(0.043/k)+ (0.044/k’)) with for (k) the value separating vb(Nature~couleurs) from vb(Nature~parfums¹) (Nature-colours, Nature-Perfumes), that is 5.8 and for k’ that between vb(couleurs~temple) and vb(parfums¹~temple) (colours-temple, Perfumes-temple), 5.6. Thus the values are: ((0.043+0.045)+(0.043/5.8)+(0.044/5.6))=((0.043+0.045)+(0.007+0.007))=(0.088+ 0.014)=0.102.

Let us continue the previous calculation to find the ell of the plateau rb(Nature~temple) (Nature- temple), I, vb(Nature~parfums¹) (Nature-Perfumes), II, vb(parfums¹~temple) (Perfumes-temple), III, vb(Nature~couleurs) (Nature-colours), IV, vb(couleurs~temple) (colours-temple), V, with glosses II and III reinforced by IV and V. The various attenuation exterior spacings resemble very closely the interior spacings since the distances are based on the same traces: «Nature», «temple», “parfums¹”, «couleurs» (Nature, temple, Perfumes, colours). On the other hand they contrast strongly with the reinforcement exterior spacings. On principle the ell is 1/tsmw((1/qepfzkgj+1/q’e’p’f’z’k’g’j’)+(1/q’’e’’p’’f’’z’’k’’k°g’’j’’+1/q’’’e’’’ p’’’f’’’z’’’k’’’k¹g’’’j’’’)), and, by giving the effective values (1/(1)(1)(1)(1))((1/(1)(1)(2)(2)(5.7)(5.7)(1)(1))+(1/ (1)(1)(2)(2)(5.5)(5.7)(1)(1))+(1/(1)(1)(2)(2)(5.8)(1)(5.8)(1)(1))+(1/(1)(1)(2)(2)(5.6)(1)(5.8)(1)(1)))=((1/129.96) +(1/125.4)+(1/134.56)+(1/129.92))=(0.007+0.007+0.007+0.007) or 0.03 when rounded up. The result may seem low but is acceptable. If we add a further reinforcement, from «sons» (sounds), we would not be far from 0.04.

It proves difficult to obtain a complete view of the work analysed in a plateau as, if we suppose that vb(Nature~Perfumes) is reinforced by vb(Nature~colours), the converse is not easily applicable in the same plateau. In fact (h)+(h’/k) and then (h’)+(h/k) would first have to be used. The whole would make (((h)+(h’/k)) ((h’)+(h/k)))=hh’+((h squared)/k)+((h’ squared)/k)+(h’h/(k squared)) which would often greatly weaken the number obtained for a slight gain in observation.

In order to increase the plausibility of a plateau, some opportunities for reinforcement should be seized if they add anything to the meaning. In this way the overall result which is usually rather low unless dealing with banalities, is brought up to 1. Thus, even if rather cumbersome, we should opt for a plateau with rb(answer~Perfumes), vb(Perfumes~cool) vb(cool~colours), vb(Perfumes~sweet) vb(sweet~sounds), vb(colours~triumphant) vb(triumphant~sounds) for example.

Since it must be possible to use the analysis whatever viewpoint is adopted on Baudelaire, so long as it is an admissible one, to see how adaptable it is let us imagine a clarification contrasting with those most often chosen, relating to rb(Nature~temple). Let us look at the attenuation vb(corrupt~temple). According to this commentary, the author, perceiving a negative nature in temples, considers it appropriate to use the religious edifice to symbolize the evil in the world, and though he conceals these thoughts, he provides clues for the most able to find them. It is very doubtful that the poet conceived „Correspondences“ in this way, but the description rb(Nature~temple) vb(corrupt~temple) makes a plateau.

Let us measure the plausibility of the plateau rb(Nature~temple) vb(corrompus~temple) which is extremely limited but contains a problem with an attenuation and so has internal coherence. The ell is 1/tsmw(2(1/qepfzgj)), that is (1/(1)(1)(1)(1))(2(1/(1)(1)(2)(2)(7.3)(7.5)(1)(1)))=1/109.5=0.009. The terms justify q=e=1; the lack of textual clarification gives the key to understanding p=2; the loose relation means f=2; z=7.3 from the 53 fronts between «corrompus» (corrupt) and «temple»; k=7.5 because «est» (is) and «temple» must be added to reach «Nature»; g=j=1 because of the lines referring to the corruption of incense.

The calculation of the plateau rb(répondent~parfums¹) vb(parfums¹~frais) vb(frais~couleurs) (N.B. frais: cool) gives 1/tsmw(1/qepfzkgj+1/q’e’p’f’z’k’g’j’) or 1/(1)(1)(2)(1)((1/(1)(1)(2)(1)(1)(1)(1)(1))+(1/(1)(1)(2) (2)(2.5)(2.6)(1)(1)))=½(½+1/26)=0.269. Reinforcements from vb(parfums¹~Doux) vb(couleurs~Doux) (N.B. Doux: soft) or vb(parfums¹~triomphants) vb(sons~triomphants) (N.B. sons: sounds) improve the result. The other sort of exterior spacing that is used for the reinforcement should then be counted. The presence of vb(parfums¹~frais) vb(parfums¹~Doux) vb(parfums¹~riches) vb(parfums¹~triomphants) can nevertheless be noted; the exterior spacing between these glosses seems to be limited by strong grammatical links and so, instead of doing the whole calculation, it is enough to see these points. Amongst the four last glosses it is true that there is a discrepancy between the last pair and the first one, but the perfumes of the incense type are still perfumes in the sense of the second quatrain and so the exterior spacing of vb(parfums¹~riches) for vb(parfums¹~frais) and rb(répondent~parfums¹) is 1. Together rb(répondent~parfums¹) vb(parfums¹~frais) vb(parfums¹~Doux) vb(parfums¹~riches) vb(parfums¹~triomphants) would as such reach (1/(1)(1)(2)(1))(1/ (1)(1)(2)(1)(1)(1)(1)(1)+1/(1)(1)(2)(1)(1)(1)(1)(1)(1)+1/(1)(1)(2)(1)(1)(1)(1)(1)(1)+1/(1)(1)(2)(1)(1)(1)(1)(1)(1)) =½(½+½+½+½)=(½)(2) but with the leveller, this becomes (½)(1)=½. It is not necessary to establish the other quantities since the overall result will be limited in any case to this amount by convention.

The ells of rb(vivants~piliers) vb(piliers~forêts) (living-pillars, pillars-forests) and of rb(forêts~ symboles) vb(symboles~piliers) are close to each other. The first gives us (1/(1)(1)(1)(2))(2(1/(1)(1)(2)(2) (2.9)(1)(1)(1)))=½(2(1/11.6))=0.086 with «La Nature est un temple…» justifying the slides, while for the second we get 0.083=(1/(1)(1)(2)(1))(2(1/(1)(1)(2)(2)(3)(1)(1)(1))) with the same stratum. The «y» (there) in the third line ensures the continuity of meaning which justifies k=1 in both calculations.

As the rivet is employed in the case of not very substantial judgements, it is frequently used since the extreme facility of the constitution of glosses means that they mostly present nothing accurate. For this reason it is very difficult to find many significant ones. Let us examine the shelving of the spit (corrompus~ sens) (corrupt-senses) to find how many glosses can be counted in it as plausible. The problem rb(corrompus~sens) has a gradient of 1/qepfzgj=1/(1)(1)(2)(2)(3.5)(2)(2); the terms give q=e=1; since the text says nothing concerning a difficulty, we are lead to p=2; the clear link between «corrompus» and «transports» does not in any way allow us to speak of an absolute connection between «sens», and «corrompus» and so f=2; z=3.5 because of the 15 fronts between the two traces; the corruption comes from a badly directed sensuality according to a very ancient moral heritage and therefore g=j=2. The gradient is then negligible at 0.017. To say that rb(corrompus~sens) is a collision proves impossible since there is no tension between the terms. For the other calculations q, e, p, f and z do not alter so we can concentrate on g and j. With rd(corrompus~sens) the slides remain 2 since the so-called problem appears to be invented. There is no problem for example in considering the body innocent: the moral or theological debate on the body as originally holy but corrupted by sin remains possible without any provocation concerning this difficult point. Thus the rivet comes into play again. The gloss vb(corrompus~sens) gives a gradient of 2(1/ (1)(1)(2)(2)(3.5)(1)(1))=0.142 clarifying rb(corrompus~encens); the stratum here comes from «transports» which suggests drunkenness and so sensual pleasures. Again for vd(corrompus~sens) the negligible is exceeded, in relation to rb(Nature~temple). This approach is reasonable but unconvincing: to argue that the sensitive natural world keeps its innocence because sensitivity is not guilty hardly seems to be what Baudelaire would frequently have heard. The two terms though avoid the value of 2 by this view which the author may have known although it was not common at that time in the milieu in which he found the bases of his thinking, the stratum coming then from the first lines of the first tercet which celebrate simplicity. As for vb(corrompus~sens) the gradient is 0.142 with qepfzgj=(1)(1)(2)(2)(3.5)(1)(1). Finally the rivet is applicable to ob(corrompus~sens) and od(corrompus~sens) since passable attenuation glosses exist and so g=j=2 in both cases. The total number of gradients that can be counted, far from being 1, is 2(0.142)=0.284.

In order to further appreciate the parallel between the intuitions and the numerical values found for them, let us consider some variations of the components and of (k). First, (q) will be reduced for vb[in~ incense] with a porterage giving the ending “…like amber, musk, benzoin and incense, which with nasal voice sing of the transports of the mind and the senses.” While in the hoard vb[in~incense] has a gradient of 2(1/(2)(1)(2)(1)(1)(1)(1)) to clarify rb[corrupt~incense], the dispenser for its part gives 2(1/(1)(1)(2)(1)(1)(1) (1)) since “nasal” is in no collision and provides an appropriate meaning for “in”. In the hoard q=2 because of the absence of any term allowing an indubitable commentary on “in”; e=1 by the term; p is 2 because of the complete silence concerning any attenuation; f=1 through the internal nature of “in” for «incense» which makes the relationship a certainty; z=1 as a result; g=j=1 since “in” provides an indication about smell, with an equally vague one concerning sensitivity, sensuality and finally corruption. The porterage thus doubles the result 2(¼)=½ giving a gradient of 2(½)=1.

Some components of vb[incense~senses] attenuating rb[corrupt~incense] can be reduced with a porterage ending thus: “…like amber, musk, benzoin and incense, which sing of the transports of the mind and the senses, clarifying the paradox.” The hoard for vb[incense~senses] gives q=e=1 from the terms; p=2 because of the silence regarding the attenuation; f=2 owing to the weak link between the meaningful words; z=2.4 as, with an unclear connection, 4 fronts separate «incense» from «senses»; g=j=1 with a stratum in the last line concerning drunkenness. The dispenser changes p, f and z which become 1 so that instead of 2(1/ (1)(1)(2)(2)(2.4)(1)(1))=0.208, we get 2(1/(1)(1)(1)(1)(1)(1)(1))=2 which goes back down to 1 with the leveller. If apart from the modification of the pole (p), the peg (f) and the interior spacing (z) are also changed here, this is because the presence of an argument, shown by the last words of the porterage, replaces the lack of clarity in the coordination of meanings by a strong link. Everything becomes dense, the new specification providing the key to the process. Whereas beauty is lost, rigour increases, clarifying the allusion by the new connection between “corrupt”, “incense” and “senses”; thus z=1 and even f=1.

Let us give high value components for the problem rb[Nature~temple], which is a collision since the rapprochement of ideas has an obvious framework plus a broad tension. Even if the question is favoured by the context, it also uses properly the internal opposition which is so effective for many oddities. Let us choose a porterage starting thus: “L'absolu est un temple/////où (de) vivants piliers laissent parfois sortir (de) confuses paroles. (La)/////Nature fait passer l'homme au travers de ses apparences en des forêts de symboles…” (The absolute is a temple where living pillars let forth at times confused words. Nature lets man pass through its appearances in forests of symbols…) For rb[Nature~temple] the hoard gives 1/tsmw= 1/(1)(1)(1)(1)=1 while the interruption in thought makes the dispenser gradient 1/qepfzgj=1/(1)(1)(2)(2)(2.8) (1)(2)=0.044, which is negligible. We have q=e=1 from the trace-terms; p=2 as the text is discreet. For the channel t=1 because of the linking «est» (is) in the original and f=2 for the porterage in which two levels of reality are supposed; s=1 and z=3, the (s) coming from the unity, the (z) from the division and from the 10 fronts separating the terms; finally m=1, g=1, w=1, j=2 because in both situations the ideas are difficult to understand.

Concerning vb[incense~senses] in the attenuation of rb[corrupt~incense], the reduction in the peg and the spacing must go with increased intuitive perception. A porterage will be used, ending: “…amber, musk, benzoin and incense which sing of the senses and of the mind in its transports.” There is a strict link then between “senses” and “incense” while “transports” is slightly effaced. A gradient of 2(1/(1)(1)(2)(1)(1) (1)(1))=1 replaces 2(1/qepfzgj)=2(1/(1)(1)(2)(2)(2.4)(1)(1))=0.208. Once the hindrance to the meaning constituted by «transports» is removed, it becomes practically impossible to deduce that the organs or the relations of sensitivity are linked by the ideas of the poem to the negative judgement on the appearance of the corrupt perfumes, only by their transports and not in their ordinary functions.

The peg and the spacing for vb[piliers~forêts] when rb[Nature~temple] is considered, can be increased by transferring the second part of the first quatrain to the end of the text, after the tercets, this time retaining the “y” (there) after “L'homme” (man). The gradient goes from 2(1/qepfzgj)=2(1/(1)(1)(2)(1)(1) (1)(1))=1 to 2(1/q’e’p’f’z’g’j’)=2(1/(1)(1)(2)(2)(7.9)(1)(1))=0.063 as there are now 59 fronts between “forêts” and “piliers” and 2+(1(59/10))=2+5.9=7.9.

By using a porterage beginning “La Nature est un temple; parfums, couleurs et sons se répondent…” (Nature is a temple. Perfumes, colours and sounds answer each other…), it is possible to weaken the spacing for an attenuation of rb[Nature~temple] such as vb[Nature~parfums¹] vb[parfums¹~ temple] (Nature-Perfumes, Perfumes-temple) and for its reinforcement which is identical in type and also has a sum gradient, vb[Nature~couleurs] vb[couleurs~temple] (Nature-colours, colours-temple). The partial plausibilities go from 1/(1)(1)(2)(2)(5.7)(1)(1)=0.043 to 1/(1)(1)(2)(2)(2.2)(1)(1)=0.113 for vb[Nature~ parfums¹] (Nature-Perfumes); from 1/(1)(1)(2)(2)(5.5)(1)(1) to 1/(1)(1)(2)(2)(2)(1)(1) or from 0.045 to 0.125, for vb[parfums¹~temple] (Perfumes-temple); overall from (0.043+0.045)=0.088 to (0.113+0.125)=0.238 for vb[Nature~parfums¹] vb[parfums¹~temple] (Nature-Perfumes, Perfumes-temple). With the other attenuation, the reinforcement exterior spacing must be counted if the calculations are to be accurate; apart from this the changes are similar. To obtain the ell of the smallest plateau containing the attenuation with reinforcement, the exterior spacing of the attenuated gloss rb[Nature~temple] must be used. However, the passage from one gradient to another, from 0.088 to 0.238 gives, for the attenuation itself without the reinforcement or the attenuation exterior spacing, a good idea of the intuitive advantage of a lesser distance between the terms.

Let us imagine a porterage beginning “The foundation of beings, Nature, is something where real entities let forth at times activities seen by the absolute man through their appearances as forests of symbols which observe him with familiar eyes.” This will give for the gloss rb[Nature~man], the gradient 1/ (1)(1)(2)(1)(1)(2)(2) instead of 1/qepfzgj=1/(1)(1)(2)(1)(1)(1)(2) with g’=2 as a result when g=1. For the hoard, the idea that perhaps man passes like a stranger through the "temple-Nature" means that the difficulty has a certain strength without being in any way a paradox, an oddity, a collision. In the dispenser the relations seem to have been sorted out so that it is debatable whether any problem is raised. The passage from ¼ to ⅛ therefore clearly describes the intuitive difference.

The slides of the neutral gloss ob[echoes~mingle] can reach 1 if répondent-confondent, the bridge for the meaning, is severed by a porterage with a second stanza: “Certainly long echoes mingle in the distance, but, with difficulty, since they overstep the barriers of perception, in a dark and profound unity, vast as the night and as the light, perfumes, colours and sounds answer each other.” In this case the use of vb[echoes~mingle] for rb[sounds~answer], rb[Perfumes~answer] and rb[colours~answer] is rendered less effective and so the gradient for ob[echoes~mingle] in the hoard, 1/qepfzgj=1/(1)(1)(2)(1)(1)(2)(2)=⅛, becomes 1/q’e’p’f’z’g’j’=1/(1)(1)(2)(1)(1)(1)(1)=½ in the dispenser. With the porterage the intuition changes as much as the measurement and we almost reach the level of a statement, with no description of any clarification or problem. Some difficulties remain in the background but can hardly be seen as hindering the understanding of the poem: thus an example of the minor difficulties are the long echoes. Should they not be taken as reflections rather than in the literal sense?

The variations in carousels and ells must also be tested to determine the value of the new measurements since the gradients might carry out their functions while the other degrees of plausibility were not regularly following the intuitive aspects of the text. We will try first with the reinforcement of vb[N~ paroles] (N-words) attenuating the problem rb[piliers~paroles] (pillars-words): changing the carver (q) should modify the intuition as well as giving a different carousel. To start with we have 2(1/(2)(1)(2)(2)(2.9) (1)(1)) for vb[N~paroles] at the same time as imagining that vb[N~Nature], which is capable of reinforcing the first gloss vb[N~paroles], is similarly modified in its carver relating to N. So 2(1/(2)(1)(2)(2)(2.9)(1)(1)) and 2(1/(2)(1)(2)(1)(1)(1)(1)) are the respective gradients of vb[N~paroles], vb[N~Nature] because in this case N gives us q=q’=2; the term e=e’=1; the silence p=p’=2; the insufficient relationship f=2, z=2.9; and the satisfactory one f’=1=z’. Finally the reference to the symbols observing man provides the stratum. With a porterage “L'être absolu, la Nature, est une immensité dont le réel entier fait le contenu, donc aussi tel temple où l'homme passe…” (The absolute being, Nature, is an immensity of which the whole of reality is the content, comprising therefore any temple where man passes…) In this case “absolu” represents the meaning of N without belonging to a collision and the first carvers can be valued as q=q’=1. The channel for rb[piliers~paroles] is 1/(1)(1)(1)(1) and the exterior spacings of the collision and the two attenuation glosses are k=2.9 and k’=1. The reinforcement distance is k’’=2.9. The calculation of both ells is straightforward, 1/ (1)(1)(1)(1)((2(1/(2)(1)(2)(2)(2.9)(2.9)(1)(1)))+(2(1/(2)(1)(2)(1)(1)(1)(2.9)(1)(1)))), in the hoard, but 1/(1)(1)(1) (1)((2(1/(1)(1)(2)(2)(2.9)(2.9)(1)(1)))+(2(1/(1)(1)(2)(1)(1)(1)(2.9)(1)(1)))) for the dispenser. So (1/33.6)+(1/5.8)=0.201 in the hoard has been doubled in the dispenser, giving (1/16.8)+ (1/2.9)=0.403 which reflects the value of the reference to the very depths of things, if the business of the words needs attenuating.

To weaken a pole and other components, we can invent two covers in which the quatrains come after the tercets with, in the second, a considerable modification at the point of the link between the newly disposed units: “(La) question pour (d')autres parfums ayant (l')expansion (des) choses infinies, comme (l')ambre, (le) musc, (le) benjoin (et) (l')encens qui chantent (les) transports (de) (l')esprit (et) (des) sens est (de) savoir s'ils (sont) corrompus, riches (et) triomphants. Or (la) Nature est (un) temple…” (The question for other perfumes having the expansion of infinite things, like amber, musk, benzoin and incense which sing of the transports of the mind and the senses is to know whether they are corrupt, rich and triumphant. Now Nature is a temple…) Let us examine an interpretation, which hardly conforms to tradition, with vb[corrompus~Nature] (corrupt-Nature) reinforcing vb[corrompus~sens] (corrupt-senses) in the attenuation of rb[corrompus~encens] (corrupt-incense). The channel of the collision to be clarified for the second cover alone is reduced to 1/(2)(1)(1)(1) since because of the question, as the degree of relationship falls, it pushes the rank up to 2. The gradients, 1/(1)(1)(2)(2)(3.5)(1)(1) and 1/(1)(1)(2)(2)(2.4)(1)(1) with the stratum of “transports” for the two covers, concerning vb[corrompus~sens] are appropriate. The other attenuation takes the gradient 1/(1)(1)(2)(2)(3.6)(1)(1) for the first cover with “confuses paroles” (confused words) as stratum. In the second cover the question and apparent reply give the gradient 1/(1)(1)(1)(1)(1) (1)(1) for the same attenuation vb[corrompus~Nature] as the two terms now seem firmly attached to each other. The exterior spacing between the attenuations hardly changes: 3.6 for the first cover, 2.8 for the second. The exterior spacings regarding the problem differ a little more: for the first cover from vb[corrompus~sens] to rb[corrompus~encens] we get 3.5 and for the second 2.9 counting two fronts for “s'ils” but not counting “sont” (are); the relation between vb[corrompus~Nature] and rb[corrompus~encens] gives 3.6 without the question and 1 with. Overall the first cover produces a carousel of 2(1/(1)(1)(2)(2)(3.5) (1)(1))+2(1/(1)(1)(2)(2)(3.6)(3.6)(1)(1)) and the second 2(1/(1)(1)(2)(2)(2.4)(1)(1))+2(1/(1)(1)(1)(1)(1)(2.8) (1)(1)) which is (1/7)+(1/25.92)=0.18 without the interrogation, and (1/4.8)+(1/1.4)=0.922 with it. This would appear to represent the intuitive advantage of the question-answer link. However, when the ell has to be calculated, a weakening occurs because the channel of rb[corrompus~encens] has been somewhat whittled down in the second cover. The whole picture is on the one hand 1/(1)(1)(1)(1) ((2(1/(1)(1)(2)(2)(3.5) (3.5)(1)(1)))+(2(1/(1)(1)(2)(2)(3.6)(3.5)(3.6)(1)(1))))=1((1/24.5)+(1/90.72))=0.051 with the first made-up text and 1/(2)(1)(1)(1)((2(1/(1)(1)(2)(2)(2.4)(2.9)(1)(1)))+(2(1/(1)(1)(1)(1)(1)(1)(2.8)(1)(1))))=½((1/13.92)+(1/1.4)) =½(0.785)=0.392 with the other, the one with the question.

A peg and a spacing for ob[Vaste~unité] (Vast-unity) reinforcing ob[ténébreuse~unité] (dark-unity) can be increased by transforming the second quatrain: “Dans la ténébreuse/////immensité de longs échos, (Les) parfums, (les) couleurs (et) (les) sons se répondent. Vaste/////unité, nuit et clarté se confondent!” (In the dark immensity of long echoes, Perfumes, colours and sounds answer each other. Vast unity, night and light mingle!) Thus the links between “ténébreuse” (dark) and “unité” (unity) are more tenuous than in the original text but those between “Vaste” (Vast) and “unité” (unity) remain powerful. The result for ob[Vaste~unité] (Vast-unity) is a gradient of 1/(1)(1)(2)(1)(1)(1)(1)=0.5, the same in both dispenser and hoard. This view is so imprecise as to pose no substantial problem and clarifies nothing. The other description of a neutral relationship, ob[ténébreuse~unité] (dark-unity), gives 0.5 from 1/(1)(1)(2)(1)(1)(1)(1) in the hoard, as it is shielded from bad slides by its excessive vagueness, and in the dispenser it is only 1/ (1)(1)(2)(2)(3)(1)(1)=0.083. The exterior spacing is 1 for the actual text but reaches 2+(1(10/10))=3 with the porterage and so the carousels are as follows: (1/(1)(1)(2)(1)(1)(1)(1))+(1/(1)(1)(2)(1)(1)(1)(1)(1))=(0.5+0.5) =1 and (1/(1)(1)(2)(1)(1)(1)(1))+(1/(1)(1)(2)(2)(3)(3)(1)(1))=½+1/36=0.527. Failure to recall terms also causes the level of intuition to be greatly reduced and a halving of the initial result seems acceptable.

For vb[répondent~parfums¹] (answer-Perfumes) we will reduce one peg and the distance when this gloss reinforces vb[chantent~parfums¹] (sing-Perfumes) in the attenuation of rb[corrompus~encens] (corrupt-incense). The porterage will use “parlent des” (speak) instead of «chantent» (sing) in the last line and will use “chantent” alongside “répondent”: “…toutes les activités du réel spontané chantent, ainsi les parfums¹, les couleurs et les sons se répondent.” (…all the activities of spontaneous reality sing, thus Perfumes, colours and sounds answer each other.) The hoard gradients, with «répondent» (answer) and «chantent» acting as strata for the glosses in which they do not occur, are 2(1/(1)(1)(2)(1)(1)(1)(1)) for vb[répondent~parfums¹] (answer-Perfumes) and 2(1/(1)(1)(2)(2)(5)(1)(1)) for vb[chantent ~parfums¹] (sing- Perfumes), taking «parfums» (Perfumes) in the eighth line. The exterior spacing toward rb[corrompus~ encens] is 4.8 in each of the attenuations and the one between the two clarification glosses is 5. In this way the hoard ell is 1/(1)(1)(1)(1)((2(1/(1)(1)(2)(2)(5)(4.8)(1)(1)))+(2(1/(1)(1)(2)(1)(1)(4.8)(5)(1)(1))))=0.02+0.041 =0.061. In the dispenser, vb[répondent~parfums¹] gives an unchanged gradient of 1/(1)(1)(2)(1)(1)(1)(1) while vb[chantent~parfums¹] reaches the same amount. The exterior spacing between attenuations is limited to 1, and from rb[corrompus~encens] is 4.8 for the gloss with “répondent” and 5 for the other gloss because “chantent” is further from “encens” than “parfums” in the porterage. Overall the dispenser ell is 1/(1)(1)(1)(1)((2(1/(1)(1)(2)(1)(1)(5)(1)(1)))+(2(1/(1)(1)(2)(1)(1)(4.8)(1)(1)(1))))=1/5+1/4.8=0.2+0.208=0.408, which is almost seven times the ell in the hoard, showing numerically the intuitive advantage of the porterage.

The increase of a peg and certain distances in the case of ob[loin~échos] (far-echoes) when this gloss reinforces ob[longs~échos] is possible with a porterage having a second stanza as follows: “Pour qui se trouve loin/////(d')un paysage, (les) parfums¹, (les) couleurs (et) (les) sons se répondent. Dans (une) ténébreuse (et) profonde unité, Vaste comme (la) nuit et (comme) (la) clarté, (de) longs/////échos se confondent.” (For he who finds himself far from a landscape, Perfumes, colours and sounds answer each other. In a dark and deep unity, vast as the night and as the light, long echoes mingle.) The gloss reinforced with “longs” keeps its original gradient of 1/(1)(1)(2)(1)(1)(1)(1), but the other with “loin”, which reinforces the first, becomes 1/(1)(1)(2)(2)(3.7)(1)(1) in the dispenser instead of 1/(1)(1)(2)(1)(1)(1)(1) with the non modified text. The hoard carousel for an exterior spacing of 1 is 1/(1)(1)(2)(1)(1)(1)(1)+1/(1)(1)(2)(1)(1)(1)(1) (1)=1 but the exterior spacing becomes 3.7 in the dispenser and the gradient falls to 1/(1)(1)(2)(1)(1)(1)(1)+ 1/(1)(1)(2)(2)(3.7)(3.7)(1)(1)=½+1/54.76=0.5+0.018=0.518. The lower number, 0.518 instead of 1, matches the recent intuitive difficulty.

It is easy to reduce the spacings without changing any other components with a porterage beginning its first tercet: “Existe-t-il des parfums frais? Comme les prairies? Il en est de frais comme des chairs d'enfants…” (Are there cool perfumes? Like meadows? There are cool ones like the flesh of children…) In the hoard we find the carousel 1/(1)(1)(2)(1)(1)(1)(1)+1/(1)(1)(2)(2)(2.8)(2.8)(1)(1)=½+1/31.36=0.5+0.031= 0.531 for the reinforcement of ob[frais~enfants] (cool-children) by ob[frais~prairies] (cool-meadows). The dispenser gives a higher value: 1/(1)(1)(2)(1)(1)(1)(1)+1/(1)(1)(2)(2)(1)(1)(1)(1)=½+¼=¾=0.75. Our intuition is apparently backed up by the calculation since the porterage links more strongly than the original the traces “prairies” and “frais”.

We will keep the «y» (there) of the present first quatrain in a porterage while placing the third and fourth lines just before the tercets in order to increase some spacings concerning the reinforcement of vb[Nature~symboles] by vb[piliers~symboles], with a view to an attenuation of rb[forêts~symboles]. The collision retains its gradient 1/tsmw=1/(1)(1)(2)(1)=½. The gloss vb[Nature~symboles] goes from 2(1/qepfzgj)=2(1/(1)(1)(2)(2)(3.5)(1)(1)) to 2(1/(1)(1)(2)(2)(5.5)(1)(1)) since to the fronts giving 3.5 must be added the twenty from the second quatrain; so, 15+20=35 and 2+(1(15/10))=3.5 while 2+(1(35/10))=5.5. The other attenuation gloss vb[piliers~symboles] which reinforces the previous one gives a modest gradient, for the calculation of the ell: 2(1/(1)(1)(2)(2)(3)(1)(1)) in the hoard, then 2(1/(1)(1)(2)(2)(5)(1)(1)) for the dispenser with the twenty extra fronts resulting in (z) going from 3 to 5. In both cases the stratum comes from the mention of the natural temple since in a sanctuary symbols are not surprising. The three exterior spacings in the hoard, one for the reinforcement, and two for the attenuation relationships, are 3.5 with vb[Nature~symboles] rb[forêts~symboles], 3 for vb[piliers~symboles] rb[forêts~symboles], and again 3.5 with vb[Nature~symboles] vb[piliers~symboles]. Thus the dispenser for the counterparts of these relationships gives respectively 5.5, 5 and 5.5. The hoard ell is 1/(1)(1)(2)(1)((2(1/(1)(1)(2)(2)(3.5)(3.5)(1) (1)))+(2(1/(1)(1)(2)(2)(3)(3)(3.5)(1)(1)))) which is ½((1/24.5)+(1/63))=0.027. The dispenser is a third of this: 1/(1)(1)(2)(1)((2(1/(1)(1)(2)(2)(5.5)(5.5)(1)(1)))+(2(1/(1)(1)(2)(2)(5)(5)(5.5)(1)(1)))) or ½((1/60.5)+(1/275))= 0.009. This calculation transcribes the weakening of links resulting from the slightly more tenuous recollection of the terms used in the first line when understanding the others.

Let us use different spacings in two covers for ob[Nature~clarté] reinforcing ob[Nature~nuit] (Nature/ light, night). In one of the versions the beginning will be: “La Nature/////est (un) temple où (de) vivants piliers, dans (une) ténébreuse (et) profonde unité vaste comme (la) nuit et (comme) (la)/////clarté, laissent parfois sortir de confuses paroles…” (Nature is a temple where living pillars, in a dark and profound unity, vast as the night and as the light, let forth at times confused words…) The carousel for the first cover will be (1/qepfzgj)+(1/q’e’p’f’z’kg’j’) or ((1/(1)(1)(2)(2)(3.1)(1)(1))+(1/(1)(1)(2)(2)(3.3)(3.3)(1)(1)))=((1/12.4)+ (1/43.56))=0.102. In the other cover we will put “La Nature,/////dans (une) clarté profonde où (les) vivants piliers (d')(un) temple font (une) ténébreuse unité vaste comme (la)/////nuit, laisse parfois sortir de confuses paroles…” (Nature, in a profound light where the living pillars of a temple make a dark unity vast as the night, lets forth at times confused words…) The carousel is then ((1/(1)(1)(2)(2)(3.2)(1)(1))+(1/(1)(1)(2)(1)(1) (3.2)(1)(1)))=((1/12.8)+(1/6.4))=0.234. Our calculation is backed by the intuition owing to the lesser obstacle to the recollection of the terms in the case of the second cover.

A porterage is necessary to break any evidence of a link when increasing two slides in the reinforcement of ob[long~echoes] by ob[distance~echoes], for example with a second quatrain as follows: “In a dark and profound unity, vast as the night and as the light, like long echoes by their proximity make a chorus, perfumes, colours and sounds answer each other, and in the distance mingle like them.” The hoard carousel is (1/(1)(1)(2)(1)(1)(1)(1))+(1/(1)(1)(2)(1)(1)(1)(1)(1))=1 with the author being so discreet as to give neither any convincing clarification nor any problem from this. For its part the dispenser carousel is (1/(1)(1) (2)(1)(1)(1)(1))+(1/(1)(1)(2)(1)(1)(1)(2)(2))=½+⅛=0.625 as problems from all sides are raised about the meaning of ob[distance~echoes] from the indistinctness concerning the chorus and the confusion of voices. The loss of the value 0.375 thus describes the obscuring of links prejudicial to the lines in question.

It is easy to lower a slide in the calculation of the plausibility of rb[rich~amber] helping rb[rich~musk] by adding a porterage like “…and others, corrupt, covered in splendour, rich and triumphant as amber…” At present «corrupt» is not sufficient, even accompanied by «triumphant», to conclusively push into the background the idea that the scent of the perfumes in question is composed of many elements and so contains abundance not like a man with a huge fortune but like a product. In fact «rich» here is the point of meeting of the two meanings, the literal and the figurative, and so accentuation is needed to change the direction, while the other words «triumphant» and «corrupt» need no assistance in leading to the idea of comparing with human things. In these conditions the hoard carousel has the appropriate level since neither the first nor the second gloss, rb[rich~amber] and rb[rich~musk], are tensions: (1/(1)(1)(2)(1)(1)(2) (1))+(1/(1)(1)(2)(1)(1)(1)(2)(1))=¼+¼=½. The change creates further disruption thus further justifying the idea of a difficulty, giving: (1/(1)(1)(2)(1)(1)(1)(1))+(1/(1)(1)(2)(1)(1)(1)(1)(1))=½+½=1.

Let us now consider the attenuations to the same extent as the gloss reinforcements. First the value of 1 will be given to the carver of N in the gradient belonging to vb[N~piliers] which is used to attenuate the problem rb[piliers~paroles] (pillars-words). In a cover “l'être absolu” (the absolute being) will be placed after “Nature” followed by “…offre de vivants piliers en un temple où l'homme passe…” (…offers living pillars in a temple through which man passes…) to represent the N by an acceptable idea quite apart from any collision. The resultant attenuation gradient is 2(1/qepfzgj)=2(1/(1)(1)(2)(1)(1)(1)(1)). In the same cover as well as in another the reinforcement can be favoured by putting a capital T in “Temple”. Thus we have rb[piliers~paroles] clarified by vb[N~piliers], which gloss is reinforced by vb[T~piliers]. With the first cover containing two changes, overall the ell is 0.106=(1/tsmw)((2(1/qepfzkgj))+(2(1/q’e’p’f’z’k’k’’g’j’)))=(1/(1)(1)(1) (1))((2(1/(1)(1)(2)(2)(2.4)(2.9)(1)(1)))+(2(1/(1)(1)(2)(2)(2.2)(2.7)(2.4)(1)(1)))) while the second cover with only one modification is less than half the other ell, being (1/(1)(1)(1)(1))((2(1/(2)(1)(2)(2)(2.4)(2.9)(1)(1)))+ (2(1/(2)(1)(2)(2)(2.2)(2.7)(2.4)(1)(1))))=0.052. The absence of the expression “l'être absolu” and of the accompanying amendments makes vb[T~piliers] less plausible than its equivalent in the first cover since both T and N are clarified by this addition. The stratum for the attenuations comes from the immediate context being in a position to suggest the idea that the characteristics conceivable for “Nature” or “Temple” influence “piliers”.

To show the various possibilities of these calculations, let us re-examine an unusual interpretation, in spite of the risk of error. However two covers are needed since the implausibility here means that the gradients are negligible when the real poem is used. In the two imitations the tercets will come before the quatrains and in only one of them the end will be: “…comme l'ambre, le musc, le benjoin et l'encens, amis de ces colonnes formant l'âme du sanctuaire, qui chantent les transports de l'esprit et des sens.” (…like amber, musk, benzoin and incense, friends of these columns forming the soul of the sanctuary, which sing of the transports of the mind and the senses.) In this way we can try to reduce the pole of one of the gradients used to calculate the ell of rb[vivants~piliers] attenuated by vb[transports~piliers], a gloss reinforced by vb[transports~sens]. The first cover, the one with only one amendment, gives an ell of 0.102 being 0.102=(1/tsmw)((2(1/qepfzkgj))+(2(1/q’e’p’f’z’k’k’’g’j’)))=(1/(1)(1)(1)(2))((2(1/(1)(1)(2)(2)(2.7)(2.7)(1) (1)))+(2(1/(1)(1)(2)(1)(1)(2.7)(2.7)(1)(1)))). In the channel, w=2 comes from “forêts” which points to "tree- pillars", taking the edge off the clash with “vivants” (living). The stratum for all the slides can easily be taken from “corrompus” which goes along with the audacious meaning given to “piliers”, no doubt inappropriately. The full-stop after the last term in the tercets breaks the train of thought, giving k’=k’’=2.7 in spite of s=z’=1. The second cover, with more amendments and with the enigma of the pillars whose identity is not expressly indicated, produces an ell which is almost double the other, of 0.205=(1/(1)(1)(1)(2))((2(1/(1)(1)(1)(2)(2.7) (2.7)(1)(1)))+(2(1/(1)(1)(1)(1)(1)(2.7)(2.7)(1)(1)))) as the poles have gone from 2 to 1. The contrasts, numerical and intuitive, remain very moderate because of the allusive nature of the ideas introduced in the second cover.

Within a problem channel, increasing the rank as well as the distance should have a negative effect on the ell of a reinforced attenuation. The porterage “Il existe une Nature/////où (de) vivants piliers laissent parfois sortir (de) confuses paroles. (L')homme passe (à) travers (le)/////temple…” (There exists a Nature where living pillars let forth at times confused words. Man passes through the temple…) allows rb[Nature~ temple] to have the channel 1/(2)(3.1)(1)(1) as against 1/(1)(1)(1)(1) at the start. As a porterage only changes those aspects necessary in the tests used in the calculation, the distances from “Nature” and “temple” to “sons” (sounds), “couleurs” (colours) and “parfums¹” (Perfumes) can be assumed to be the same in the dispenser as in the hoard. We will therefore examine how the ell for rb[Nature~temple] attenuated by vb[Nature~parfums¹] vb[parfums¹~temple] (Nature-Perfumes, Perfumes-temple) is modified, these glosses with sum gradients being reinforced by vb[Nature~couleurs] vb[couleurs~temple] (Nature- colours, colours-temple) and again by vb[Nature~sons] vb[sons~temple] (Nature-sounds, sounds-temple). The total for the hoard is (1/(1)(1)(1)(1))(((1/(1)(1)(2)(2)(5.7)(5.7)(1)(1))+(1/(1)(1)(2)(2)(5.5)(5.7)(1)(1)))+ ((1/(1)(1)(2)(2)(5.8)(5.8)(5.8)(1)(1))+(1/(1)(1)(2)(2)(5.6)(5.8)(5.6)(1)(1)))+((1/(1)(1)(2)(2)(5.9)(5.9)(5.9)(1)(1)) +(1/(1)(1)(2)(2)(5.7)(5.9)(5.7)(1)(1))))=1((0.007+0.007)+(0.001+0.001)+(0.001+0.001))=0.018. As the calculation is only up to three decimal places, certain spacings vary very little in spite of the different distances between the terms used, as from “Nature” to “sons” (Nature-sounds), and from “temple” to “sons” (temple-sounds). It should also be noted that the reinforcement distance with vb[sons~temple] (sounds- temple) is counted with reference to vb[parfums¹~temple] (Perfumes-temple) while [Nature~parfums¹] (Nature-Perfumes) is used for that of vb[Nature~sons] (Nature-sounds). In the dispenser the ell is six times smaller, above all because of the channel: ((1/(2)(3.1)(1)(1))(0.02))=0.003. Indeed, as the problem to be clarified may just have been invented by the interpreter, it is perfectly consistent for the plausibility of the whole plateau to be reduced even if the ideas for the attenuation remain as good as before in themselves.

It is easy to increase certain distances for vb[répondent~parfums¹] (answer-Perfumes) attenuating rb[Nature~temple] and reinforced by vb[chantent~parfums¹] (sing-Perfumes) when a porterage is used with “Des éléments corrompus” (Corrupt elements) in the eleventh line. The dispenser, with “répondent” and “chantent” as strata for the glosses where they are absent, produces an ell of (1/(1)(1)(1)(1))((2(1/(1) (1)(2)(1)(1)(6.1)(1)(1)))+(2(1/(1)(1)(2)(2)(5)(8.8)(5)(1)(1))))=(1/1)((1/6.1)+(1/(2)(5)(8.8)(5)))=(1)(0.163+0.002) =0.165 the point of view given being that if the perfumes, colours and sounds look like worshippers then “La Nature est un temple…” The hoard is (1/(1)(1)(1)(1))((2(1/(1)(1)(2)(1)(1)(6.1)(1)(1)))+(2(1/(1)(1)(2)(1)(1)(8.8) (1)(1)(1))))=(1)(0.163+0.113)=0.276.

A shorter distance can lower a peg, as in the case of rb[corrompus~piliers] clarified by the gloss vb[corrompus~transports] reinforced by vb[corrompus~sens] (corrupt/pillars, transports, senses). This distance can be obtained through the removal of «Laissent…autres». In the hoard we find (1/qepfzgj) ((2(1/q’e’p’f’z’kg’j’))+(2(1/q’’e’’p’’f’’z’’k’k’’g’’j’’)))=(1/(1)(1)(2)(2)(7)(1)(1))((2(1/(1)(1)(2)(2)(3.3)(8.4)(1)(1))) +(2(1/(1)(1)(2)(2)(3.5)(8.6)(3.5)(1)(1))))=((1/28)((1/55.44)+(1/210.7))), the stratum coming from «musc» for the two attenuations. In the dispenser the ell is very different: (1/(1)(1)(1)(1)(1)(1)(1))((2(1/(1)(1)(2)(2)(3.3) (3.4)(1)(1)))+(2(1/(1)(1)(2)(2)(3.5)(3.6)(3.5)(1)(1))))=((1/22.44)+(1/88.2))=0.055 and this is greater than the other since the obstacle in question is much more easily seen.

Let us reduce two exterior spacings of the plateau in which rb[répondent~parfums¹] is attenuated by vb[N~Nature], a gloss reinforced by vb[Nature~temple]. A porterage will be used beginning: “La Nature est un temple où parfums, couleurs et sons se répondent.” (Nature is a temple where Perfumes, colours and sounds answer each other.) The hoard gives an ell of (1/tsmw)((2(1/qepfzkgj))+(2(1/q’e’p’f’z’k’k’’g’j’)))=(1/(1) (1)(2)(1))((2(1/(2)(1)(2)(1)(1)(6.1)(1)(1)))+(2(1/(1)(1)(2)(1)(1)(6.1)(1)(1)(1))))=½((1/(2)(6.1))+(1/6.1))=0.122 with the passage on «paroles» (words) as the stratum. The dispenser produces an ell matching the improved intuitive understanding, (1/(1)(1)(2)(1))((2(1/(2)(1)(2)(1)(1)(1)(1)(1)))+(2(1/(1)(1)(2)(1)(1)(1)(1)(1) (1)))), levelled to ½.

For rb[Nature~temple], attenuated by vb[parfums¹~symboles] reinforced by vb[couleurs~symboles], two spacings will be increased. The porterage will give initially the first two lines followed by the tercets and finally the remaining lines of the quatrains. The hoard gives an ell of (1/tsmw)((2(1/qepfzkgj))+(2(1/q’ e’p’f’z’k’k’’g’j’))) = (1/(1)(1)(1)(1))((2(1/(1)(1)(2)(2)(4.1)(5.7)(1)(1))) + (2(1/(1)(1)(2)(2)(4.2)(5.8)(4.2)(1)(1)))) = 1((1/46.74)+(1/204.624))=0.025. In the dispenser the two attenuation exterior spacings become 8.7 and 8.8 from the 30 fronts in the tercets, added to the 37 and 38 of the original text between «Nature» and «parfums», and «Nature» and «couleurs»: 2+(1(37/10))=5.7 and 2+(1(38/10))=5.8 giving, after modification, 2+(1((37+30)/10))=8.7 and 2+(1((38+30)/10))=8.8. The dispenser ell is therefore smaller than that of the hoard: (1/(1)(1)(1)(1))((2(1/(1)(1)(2)(2)(4.1)(8.7)(1)(1)))+(2(1/(1)(1)(2)(2)(4.2)(8.8)(4.2)(1)(1))))=1((1/71.34)+ (1/310.464))=0.017. The reduction observed is thus consistent with the greatly increased difficulty of recall with the interposition of the tercets.

Let us lower two exterior spacings in a case with a pivot. The hoard ell (1/tsmw)((2(1/qepfzkgj))+ (2(1/q’e’p’f’z’k’k’’g’j’))) will give very high values for (k’) and (k’’) while their equivalents in the dispenser will fall. We will use a porterage with instead of the present ending “…et d'autres, parfums du temple, l'encens, l'ambre, le benjoin, le musc, corrompus, riches et triomphants, ayant l'expansion des choses infinies…” (…and others, perfumes of the temple, incense, amber, benzoin, musk, corrupt, rich and triumphant, having the expansion of infinite things…) The plateau of rb[corrompus~encens] vb[en~encens], the latter gloss being reinforced by vb[en~temple], should have a hoard ell of (1/(1)(1)(1)(1))((2(1/(2)(1)(2)(1)(1)(1)(1) (1))) +(2(1/(2)(1)(2)(2)(8.4)(8.4)(8.4)(1)(1)))) which is unfortunately impossible with a negligible gradient of 2(1/ (2)(1)(2)(2)(8.4)(1)(1)) from vb(en~temple) with “en” taken from «encens» (in/incense). For the two attenuations the stratum comes from «sens» which suggests the sense of smell, but without being able to justify q=q’=1. The dispenser ell is (1/(1)(1)(1)(1))((2(1/(2)(1)(2)(1)(1)(1)(1)(1)))+(2(1/(2)(1)(2)(1)(1)(1)(1)(1) (1))))=1 and the intuitive contrast is hardly surprising in these circumstances.

The reduction in slides and spacings that comes from giving a stratum to an idea that is difficult to defend with the actual text can prevent its measurement of plausibility from being negligible. For the plateau of rb[corrompus~encens] vb[esprit~encens] (mind-incense), this latter gloss being reinforced by vb[sens~ encens] (senses-incense), a porterage will give a very different end to the text: “…l'ambre, le musc, le benjoin et l'encens sont-ils corrompus, riches et triomphants à cause de l'esprit et des sens qui détourneraient les effets de ce qui dépend trop d'eux?” (…are the amber, the musk, the benzoin and the incense corrupt, rich and triumphant because of the mind and the senses diverting the effects which depend too much on them?) The hoard ell uses 2(1/(1)(1)(2)(2)(2.3)(2)(2)) and 2(1/(1)(1)(2)(2)(2.4)(2)(2)) in (1/(1)(1)(1)(1))((2(1/(1)(1)(2)(2)(2.3)(3.4)(2)(2)))+(2(1/(1)(1)(2)(2)(2.4)(3.5)(2.4)(2)(2)))). The gradients are less than 0.062 as the lack of stratum means g=g’=j=j’=2 for (1/tsmw)((2(1/qepfzkgj))+(2(1/q’e’p’f’z’k’k’’g’j’))) and therefore the whole is unusable. The dispenser increases the plausibilities: (1/(2)(1)(1)(1))((2(1/(1)(1) (2)(2)(1)(1)(1)(1)))+(2(1/(1)(1)(2)(2)(1)(1)(1)(1)(1))))=½.

Imagining a porterage which favours the intuition of an idea in one respect but makes it difficult in another should bring us to a situation which is roughly equivalent to the one we started with. Weakening the notions of corrupt incense and perfumes which answer or sing is a way of getting pegs of 2 for the ell of rb[corrompus~encens] vb[répondent~parfums¹] with the attenuation gloss reinforced by vb[chantent~ parfums¹]. The same porterage will be responsible for bringing together all those ideas: “…les parfums¹, les couleurs, les sons, se répondent-ils et chantent-ils, en particulier l'ambre, le musc, le benjoin et l'encens, corrompus, riches et triomphants?” (…do the Perfumes, colours and sounds answer each other and sing, in particular, the amber, musk, benzoin and incense, corrupt, rich and triumphant?) This device will lower the three exterior spacings, and the overall result, through both its lowered characteristics and its equally important increased ones, should show hardly dissimilar intuitions and numerical values. The hoard gives (1/tsmw)((2(1/qepfzkgj))+(2(1/q’e’p’f’z’k’k’’g’j’)))=(1/(1)(1)(1)(1))((2(1/(1)(1)(2)(1)(1)(4.8)(1)(1)))+(2(1/(1)(1) (2)(1)(1)(4.8)(1)(1)(1))))=1((1/(4.8))+(1/(4.8)))=(2/4.8) or (1/2.4)=0.416 with the first line as a stratum. The dispenser gives a similar result: (1/(2)(1)(1)(1))((2(1/(1)(1)(2)(2)(1)(1)(1)(1)))+(2(1/(1)(1)(2)(2)(1)(1)(1)(1) (1)))) or ½(½+½)=½=0.5.

In order to constitute a wide plateau we will try to assemble some ideas on attenuations with sum gradients and the problems clarified by them, first vb(Nature~Perfumes) vb(Perfumes~temple), reinforced by vb(Nature~colours) vb(colours~temple), vb(Nature~sounds) vb(sounds~temple). Even though they appear too vague to clarify rb(Nature~temple), the attenuation glosses together describe enough meanings to be considered of interest. However, any misgivings about them on an intuitive level are accompanied by the formidable perspective of the wide spacings that mark their measurements: paragraph 151 showed that a plateau with such attenuations, plus the difficulty they deal with rb(Nature~temple), has an ell of about 0.02. In a general ell, for a sufficiently numerous plateau, many values for which no cut exists must be multiplied and so the overall result would be lowered owing to the weight of 0.02, which explains why in the end this unit will have to be avoided, even though its orderliness makes it amusing. On the contrary, paragraph 128 gave us a glimpse of how the series rb(answer~Perfumes) vb(Perfumes~cool) vb(cool ~colours), vb(Perfumes~sweet) vb(sweet~colours), vb(Perfumes~triumphant) vb(triumphant~colours), vb(Perfumes~rich) vb(rich~colours) easily comes to ½ only from rb(answer~Perfumes) vb(Perfumes~cool) vb(Perfumes~sweet) vb(Perfumes~triumphant) vb(Perfumes~rich). Glosses of this type can come into a plateau without serious consequences along with those more captivating ones which may have a lower numerical value.

Let us use in the same plateau rb(answer~Perfumes) vb(Perfumes~cool) vb(cool~colours), vb(Perfumes~sweet) vb(sweet~sounds), vb(colours~triumphant) vb(triumphant~sounds); and then rb(pillars~words) vb(N~Nature); rb(Nature~temple) vb(N~Nature); rb(eyes~symbols) vb(N~Nature); rb(observe~symbols) vb(N~Nature); rb(corrupt~incense) vb(transports~senses); rb(sing~Perfumes) vb(Perfumes~oboes). In order to avoid descending to very low values, a powerful attenuation, such as vb(N~Nature), must be used more than once, and each time a very modest spacing must be sought. The first glosses, which number seven, have a value of ½ as a result of the collision rb(answer~Perfumes) which has a channel of ½. The two attenuations vb(Perfumes~cool), vb(Perfumes~sweet) which, by reinforcing each other, are 1, the highest figure possible, keep this value. This gives ½(1)=½. The four collisions mentioned later have a channel of 1 which cannot harm the plateau, and the same attenuation vb(N~Nature) is used everywhere but is only counted once for its gradient of ½, using the cutting device. This amount which is constantly 2(1/(2)(1)(2)(1)(1)(1)(1)) has slides from a single stratum: the eighth line. On the other hand the exterior spacings in the cases studied remain 1, in particular from the «There» in the third line which ensures continuity of meaning in the first quatrain. The product up to here is therefore ((½) (½))=¼. Next comes rb(corrupt~incense) vb(transports~senses) which is a less attractive pair than rb(corrupt~incense) vb(incense~senses) but has the advantage of an interior spacing of 1 for vb(transports~senses) and an exterior spacing of 1 in relation to the collision, giving (1/(1)(1)(1)(1)) (2(1/(1) (1)(2)(1)(1)(1)(1)(1)))=1 as regards this small plateau made use of within the larger one. The stratum comes easily from «senses», a term here very useful for us although ousted from the main area of the present clarification. The general product remains at ¼ since ¼(1)=¼. There remains therefore rb(sing~Perfumes) vb(Perfumes~oboes). The collision has a channel of 1/(1)(1)(2)(1) from «sing» which can be thought of in the figurative sense. The attenuation gloss has a gradient of 2(1/(1)(1)(2)(1)(1)(1)(1)), the stratum from the eighth line coming into play again; the exterior spacing cannot fall below 1 since thoughts provoked by the contrasts within the tercets unite the terms, and the two amounts relating to internal distances conserve the value of 1. The total for the ell amounts to (¼)((1/(1)(1)(2)(1))(2(1/(1)(1)(2)(1)(1)(1)(1))))=(¼)(½)(1)=⅛= 0.125.