Students of the veda often question each other on trivia to test their knowledge of the shruti. We have always believed that graphically depicting some of these produce a clear effect on the student. This presentation of trivia might be viewed in conjunction with an earlier presentation that we had produced when we were in the mouth of kR^ityA.
This is a basic summary of the RV by maNDala. The height of the bar is the number of sUkta-s while width is scaled as per the median sUkta length in terms of number of R^ik-s. One can note that shAkala (shAkalya), the last redactor of the surviving text aimed at symmetry of maNDala 1 and 10 between which the maNDala follow a stepwise pattern in terms of length by number of sUkta-s. But what is striking is that the median length of a sUkta in terms of number of R^ik-s is noticeably different between different maNDala-s. We look at this below in greater detail.
This is a standard box-plot of the length distribution in terms of number of R^ik-s of sUkta-s in each of the 10 maNDala-s of the RV. The median length is the point in the box, while the lower and upper quartiles bound the box and the whiskers give the complete range. The horizontal line is the median sUkta length across the 10 maNDala-s.
Some interesting points:
-The two dramatic differences from the general trend are seen in the kANva maNDala(RV 8) and the pavmAna soma maNDala (RV 9). The former has the greatest median sUkta length and the latter the least median sUkta length. Both are the top two most-densely connected maNDala-s in the R^iShi-sharing graph, yet they are antipodes in terms of sUkta length. Clearly, the 8th dominated by a kANva core is different from all other maNDala-s in with respect to length, including their A~Ngirasa cousins like the gotama-s and the bharadvAja-s. The soma maNDala is particularly striking because it contains the longest sUkta and a few other particularly long sUkta-s, yet the median length of a sUkta in it is the shortest.
-The gR^isamada-s, vishvAmitra-s, vAmadeva gotama and atri-s never composed too long sUkta-s (>25) though the first 3 had sUkta-s of greater than or equal to overall median length. In contrast, the bharadvAja-s and vasiShTha-s composed sUkta-s some sUkta-s of great length thought their median sUkta length tended to be less than the overall median.
-The maNDala 1 and 10 despite having the same number of sUkta-s differ in median sUkta length, with RV 1 clearly exceeding the overall median.
The shortest sUkta: RV 1.99 (1 R^ik) This is supposed to be the locus of the insertion of the secret shredhI-vidyA (or mysterious expansion R^ik-s in the form of an arithmetic series).
Longest sUkta= RV 9.97(58 R^ik-s)
Median sUkta length for RV across the 10 maNDala-s=9
These features have implications for the nature of the old ritual and the tradition of redaction of the RV text. Clearly the tendency reasonable clustering of most books close to the overall median distribution suggest that the old Indo-Aryan ritual was centered around the recitation of sUkta-s of about 7-9 R^iks along with fire oblations. This might have also defined the preferred length that an extempore kavi could compose at a rite. The longer hymns in family books other than the kANva-s are always related to specialized issues: e.g. the ashvamedha ritual hymns in RV1, the astronomical and philosophical hymn of dIrghatamas in RV 1, the medical toxin hymn of agastya in RV 1, the philosophical vaishvadeva hymns of prajApati vishvAmitra in RV 3, the fire kindling and installation hymns of the bharadvAja-s in RV 6 or vasiShTha’s hymn of yatudhAna-s etc in RV 7 are examples of such. This again supports the issue that the observed median length is a reflection of the extempore composition at the more archaic form of the shrauta ritual, whereas the longer hymns are for specialized purposes. Again, the notably shorter median length of the soma maNDala indicates that the liturgy specifically used in the archaic soma rite typically featured shorter compositions centered on a size of 6 R^ik-s. Thus there appears to have been a distinction between the compositions used in the fire ritual and those in the soma ritual. However, the soma maNDala contains the longest sUkta and some other long sUkta-s that are clearly outliers. These sUkta-s upon examination are clearly later redaction efforts that combine sUkta-s or their fragments from different authors into a single long sUkta. For example the longest sUkta is a compilation of the sUkta-s of short length by 11 vasiShTha-s and a single A~Ngirasa. These later redactions appear to mark the beginning of the move from the archaic short soma-specific composition to the classic shrauta liturgy of a draw-out soma recitation.
The kANva-s who are exception, are also characterized by unusual meters and guru/laghu distributions in common meters like gAyatrI. This observation, taken together with the multiple features suggesting a closeness of the kANva-s to the Iranian zone, implies that the kANva-s probably had a distinct system of ritual performance. They organized compositions into longer sUkta-s, which were possibly an archaic version of the longer shastra-recitation seen in the classical shrauta ritual, where multiple shorter sUkta-s are strung together. Finally the difference between maNDala 1 and 10 is reflected in fact that while maNDala 1 is largely a collection of short family books, the terminal part of RV 10 is a collection of several short sUkta-s of unclear provenance that probably formed a floating mass of tradition.
The frequency distribution of sUkta-s by length in number of R^ik-s across all 10 maNDala-s. Note across all maNDala-s a length of 5 R^ik-s appears to be most common. The blue vertical line is the median sUkta length across the entire RV saMhitA.
Click here to magnify
The frequency distribution of sUkta-s by length in number of R^ik-s in each of the 10 maNDala-s.The the most frequent sUkta size and the median sUkta size are indicated for each maNDala. A most frequent sUkta length of 5 R^ik-s appears to be most common individual maNDala-s too. However, note the dramatic difference in the vAmadeva(RV 4) and kANva (RV 8) maNDala-s.
mAnasa-taraMgiNI 
