How a Professor Turned Tamil Nadu’s Traditional Kolams Into a Computer Language
Kolam became a computer language when Gift and Rani Siromoney described its drawing patterns with precise rules. Their approach treated each design as a sentence in a pictorial language. The idea mattered because a computer could run those rules, turning a doorstep art into something that could be studied and generated computationally. Working with computer scientist Kamala Krithivasan, the Siromoneys published papers on picture languages and array grammars in 1973 and 1974. Their rules rewrote small arrays of symbols into larger arrays, much as grammar rules build sentences. Computers could then generate kolams of a given type mechanically. The rules reflected how practitioners built designs in stages and remembered chains of smaller units. The women did not need to know any grammar for their skill to have grammatical properties. Later researchers continued using Siromoney array grammars, and a 2018 study generated kolams using membrane computing, which could produce designs earlier grammars could not handle.
What did it mean to turn a kolam into a computer language?
Kolam became a computer language when Gift and Rani Siromoney described its drawing patterns with precise rules. Their approach treated each design as a sentence in a pictorial language. The idea mattered because a computer could run those rules, turning a doorstep art into something that could be studied and generated computationally.
Working with computer scientist Kamala Krithivasan, the Siromoneys published papers on picture languages and array grammars in 1973 and 1974. Their rules rewrote small arrays of symbols into larger arrays, much as grammar rules build sentences. Computers could then generate kolams of a given type mechanically.
The rules reflected how practitioners built designs in stages and remembered chains of smaller units. The women did not need to know any grammar for their skill to have grammatical properties. Later researchers continued using Siromoney array grammars, and a 2018 study generated kolams using membrane computing, which could produce designs earlier grammars could not handle.
What is a kolam, and what are the pulli dots used to draw one?
A kolam is a patterned drawing made at a home’s threshold, a practice described in the article as common across Tamil Nadu. Women sweep the doorstep, sprinkle it with water and cow dung, then use powder to create symmetrical designs. The practice brings together careful drawing and a daily household tradition.
A kolam begins with a grid of dots called pulli. Practitioners draw the dots first, then build the pattern around them. Siromoney’s observations and experiments found that skilled practitioners developed designs stage by stage, in natural segments, rather than treating each kolam as one whole picture.
Tradition holds that rice flour used for the drawings feeds ants and insects, making the morning practice an act of kindness. Siromoney traced the earliest mentions he found in Tamil literature to the 16th-century Madurai Meenakshiammai Kuram. The article notes that older word lists, paintings and travellers’ accounts do not mention the art, so its documented literary history begins in the 16th century.
How did Siromoney record the stages practitioners used to build a kolam?
Siromoney wanted to see not just a finished kolam but how practitioners built one. He devised a way to preserve the drawing process in stages. The resulting record helped him study the sequence of actions and the way skilled women constructed their designs.
He stacked seven sheets of paper, placing carbon paper between them, and asked practitioners to draw with a red ballpoint pen. At the end of each stage, he removed the top sheet and its carbon paper. The sheets below retained the drawing, with the earlier strokes in black and the newest stage in red.
This created a frame-by-frame record of the developing pattern. The observations showed that practitioners drew the pulli first and then built the kolam in natural segments. Siromoney concluded that they remembered designs as chains of smaller units rather than as whole pictures. The staged record helped make the drawing process visible for analysis.
How many sheets of paper did he use, and what did the different layers show?
Siromoney used seven sheets of paper to capture how a kolam developed. He placed carbon paper between the sheets and asked practitioners to draw with a red ballpoint pen. This unusual setup let him record the process, rather than relying only on the finished design.
At the end of each stage, he removed the top sheet and its carbon paper. The remaining sheets carried the strokes made so far. In the record, earlier strokes appeared in black, while the latest stage appeared in red. That color difference separated the new marks from the drawing already made.
Together, the layers gave Siromoney a frame-by-frame view of the kolam’s construction. The records showed that practitioners started with the pulli dots and added the design in natural segments. These observations supported the idea that skilled artists remembered a kolam as a chain of smaller units, not simply as a complete picture.
How do array grammars use rules to grow small arrays of symbols into kolam patterns?
Array grammars describe how a small arrangement of symbols can develop into a larger one. In the Siromoneys’ work, the rules acted on arrays, or organized groups of symbols, rather than on words alone. The approach mattered because it gave a computer a formal way to produce picture patterns.
The Siromoneys treated each kolam as a sentence in a pictorial language. A rule rewrites a small array into a larger array, much as a grammar rule can grow a sentence. By applying such rules, computer programs could generate kolam patterns of a given type mechanically. The researchers published work on picture languages and kolam in 1973 and 1974.
Later researchers called these models Siromoney array grammars. The article also describes related work on interlaced sikku kolams, where cut and join operations can turn a single-strand pattern into a multi-strand one and back. In 2018, researchers used membrane computing to generate designs that earlier array grammars could not handle.
What could computers do once kolams were described using formal rules?
Once kolams were described with formal rules, a computer program could generate patterns of a given type mechanically. That was a striking shift: an art drawn by hand at doorsteps could also serve as a precise example of a picture language. It gave computer scientists patterns they could study using formal methods.
With Kamala Krithivasan, Gift and Rani Siromoney developed rules that rewrote small arrays of symbols into larger ones. The computer could apply those rules to construct a kolam pattern. For interlaced sikku kolams, Siromoney also introduced cut and join operations to convert between single-strand and multi-strand patterns.
Kolams supplied ready-made examples of existing picture languages and inspired new types. The article says this helped shape how machines draw and recognise images. Research continued: a 2018 paper generated kolams using membrane computing, a model inspired by living cells, and produced designs that earlier array grammars could not handle.
What is a formal grammar, and how can rules describe a picture as well as a sentence?
In general language theory, a formal grammar is a set of explicit rules for building valid structures from smaller parts. Grammars are often used to describe sentences, but the rules can also describe patterns made from symbols arranged in a picture. This general definition goes beyond the article’s brief account of grammar.
The Siromoneys treated each kolam as a sentence in a pictorial language. Their array rules rewrote a small array of symbols into a larger one. The analogy is that a sentence grows through grammar rules, while a kolam pattern grows through rules for its visual parts. A computer program could follow the rules to generate designs of a given type.
This made the drawings useful to computer scientists studying picture languages, and it showed how mathematical rules could describe a craft without changing how practitioners made it. The women did not need to know the grammar. Later researchers continued to use array grammars, while a 2018 study explored membrane computing to generate patterns beyond the earlier models’ reach.
This brief was written by AI from the original reporting and checked by other models. Names, figures and quotes come from the source; read it for full context.
Read more in the JupiteX app
Pulse is free. New stories every 4 hours, each one broken into the questions that explain it.
Or read more news on the web