A Tale of Two Jams: Ube Halaya, Rasavalli Kizhangu Kali

I’ve a theory that halvas are grandparents to what we know as jams. Maybe that’s too daring? Then this: what’s a halva in the East is a jam or a jelly in the West. I’ve not much to substantiate either theory except a hunch. Hear me out?

Two things got me wondering. The first was the ube halaya, the second vilampazham jam–both are called “jams,” but neither is a jam at all really. Two posts on this, one on each “jam.” This is the first.

Ube Halaya is that beloved Filippino “jam” made of purple yams — ube, Dioscorea alata, ratalu kand in Gujarat and one of the signature ingredients of undhiyo. Indian ube tends, however, to be more white than purple; the stunning purple-ness of the purple yam, while variable even in Filippino soil, is undoubtedly that region’s distinction. Ube is not so common in Tamil cookery, but known locally as Rasavalli kizhangu/இராசவள்ளி கிழங்கு, possibly also as Kaachikizhangu/ காச்சி கிழங்கு, Perumvalli or Peruvalli kizhangu and sometimes (confusingly) as the tuber that has betel-like leaves–Vettrilai valli kizhangu/வெற்றிலை வள்ளி கிழங்கு (but then that can refer to any yams with heart shaped leaves, and that’s about all of them! There are possible links between halaya and rasavalli kizhangu kali, which I’ll come to later.

Halaya is really the on-loan-from-Spanish Tagalog name for jalea, which means clear jelly (properly) or jam (loosely; the proper name for jam in Spanish would be mermelada).

The word jalea itself is thought to be derived from gelata, gelatine, or the “jelled result of boiling pigs’ feet and chicken carcasses or ham bones!” [that’s my former professor Julie Taylor exclaiming from Buenos Aires on what jalea most readily called to mind]. It was savory, not sweet. Could the word jalea have also or alternatively derived somehow from halva, I got to wondering? The words are practically homonyms. Early halvas themselves used dates instead of sugars, much as early fruit preserves relied on honey, though halva-like confections such as turrón/ toronne, a gift from the Moors much like sugar itself, became popular and eventually, like halvas, relied on sugar for sweetness. There’s a huge literature on the travel of sugar cane into Europe and beyond [see David Abulafia’s The Great Sea: A Human History of the Mediterranean, for just one example], and much less on the travels of halvas into southern Europe in the 12th-15th centuries, so it’s hard to tell if the one piggypbacked on the other, halva carrying sugars deep into Europe and becoming something of a template for the sweeter jellies and eventually jams and fruit preserves (and preservation, more generally). I’m speculating, however, that at the very least, jalea might have become a local interpretation of the ‘imported’ halva, or a translation of the word, though not likely a derivative from it. In other words, and especially for those halva types that resembled clear jellies, such as this corn starch-based Karachi halva (or Bombay halva), the categories might well have been similar enough or might have overlapped.

For all those older folks in the Philippines cooking up ube before the Good Shepherd Convent nuns in Baguio City took over and made “ube jam” their speciality–ube halaya was traditionally just simple boiled yam with milk and sugar, chunkier or smoother depending on the tastes of the cook, with dayap (Citrus aurantifolia) juice and rind added in. It’s unclear if it was ever cooked down into a thick pudding, though ube thickens by nature, and a pudding it might always have been. In process, halaya was perhaps close to what is known now as nilupak na ube, which is made by pounding cooked ube with thickened milk, sugar, and a fat like butter or margarine slowly incorporated. “Nilupak” is a type of “mashed/pounded starchy food with coconut milk and sugar”; the term derives from the Tagalog verb lupak [to pound into a pulp with mortar and pestle]. These days, only confections made of mashed cassava or saba bananas are called nilupak. The sweet potato, ube, and taro variants cooked in coconut milk with no pounding are known generally as “halaya.” The distinction is a fine one, because “nilupak” can mean mashing as well as pounding, and mashing is inherent to cooking halaya as much as pounding in thickened milk etc. To the Spanish in the Philippines and maybe thanks to the nuns, nilupak na ube gave way to the now-more-common jalea.

But why “jalea”? Did the Spanish have clear, gelatinous jalea in mind really or was it just the closest approximation, applied to the local ube preparation?

For clear jelly the halaya is most certainly not! In fact this Filipino author suggests that “‘jam’ may not be entirely the right word to describe what it truly is” and that “pudding” describes it better, really. Indeed, halaya is almost the twin of what Jaffna Tamils will know as rasavalli kizhangu “pudding” (in English) or kanji/கஞ்சி/porridge, or even a kali, I assume for its soft-to-mouldable texture. The halaya production process, too, the addition of milks and sugars and fats, the long and consuming stirring, even the use of the yam as key ingredient–all these things make it so very akin to all the “halwas” of the near-to-far-east, that it’s difficult to think of it as a “jam” in the same breath as other fruit preserves at all. The halaya is a lot more halva than jam, but the fact that it was even interpreted as a jalea suggests at the very least much overlap in these categories.

Jams & Halvas

The word “jam” itself, says Alan Davidson in the Oxford Companion to Food (1999) “is a newcomer. Its derivation seems an abiding mystery.” The OED [Oxford English Dictionary] suggests its source is the verb ‘to jam’, to bruise or crush by pressure, which makes its first appearance in 1706. Yet the historian David Potter has found a plum jam recipe in a manuscript belonging to the family of Ann, Lady Fanshawe that must date ‘some years before 1708’, “although the word does not enter the printed record until the first edition of Mrs. Mary Eales’ Receipts in 1718.”

What Davidson doesn’t mention is that jams aren’t just the accompaniments-to-toast-and-scones that we know them to be now, but often served as sweets in their own right. Maguelonne Toussaint-Samat tells us in her exhaustive History of Food, that medieval jams were like fruit pastes and that sugar was regarded as a spice, offered “as company arose from the table,” both to deodorize breath and aid digestion, its quality and variety defining the renown of the table from which it was consumed (489). And though we’ve stopped thinking of sugar as a “spice,” spices and flavorings distinguish most halvas and serve similar purposes, once again suggesting some historical overlap.

Halvas are of course far older than jams, and didn’t always rely on sugar. Alan Davidson again:

HALVA is the “name of a hugely varied range of confections made in the Middle East, C. Asia, and India, derived from the Arabic root hulw, sweet. In 7th-century Arabia, the word meant a paste of dates kneaded with milk. By the 9th century, possibly by assimilating the ancient Persian sweetmeat afroshag, it had acquired the meaning of wheat flour or semolina, cooked by frying or toasting and worked into a more or less stiff paste with a sweetening agent such as sugar syrup, date syrup, grape syrup, or honey by stirring the mass together over a gentle heat. Usually a flavouring was added such as nuts, rosewater, or puréed cooked carrots (still a popular flavouring). The finished sweetmeat could be cut into bars or moulded into fanciful shapes such as fish.”

You see from this how halva could have been the sister of the turrón, but Davidson’s more interesting point might be that really the vast diversification of halva happened not in today’s Middle-East but in India where it’s not just semolina that’s used as a base, but several other grains and lentils (Karachi halva uses corn flour, Tirunelveli halva uses samba godhumai, mung halva is a Rajasthani speciality or a Thiruvaiyaru special as you please, etc.) and, more importantly at the moment, vegetables and fruits, including carrots, potatoes and yams.

Then there’s this entry on HALVA/HALVAH in Gil Marks’ Encyclopedia of Jewish Food (2010):

“When the conquering Arabs arrived in Persia in 642, they discovered sugar as well as a highly sophisticated and diversified cuisine, including a vast range of confections, puddings, and pastries, which they adopted and spread westward. The Arabs referred to certain confections as halwa from the Arabic root hilwa (sweet), which the Persians pronounced halva. Later the Ottoman Turks introduced these confections to their domain. Consequently, halva, evolving into an array of versions, became popular from India to the Balkans and North Africa.

“Over the course of centuries, various terms and identifications changed as well. When the word halwa first appeared in the seventh century, it referred to a mixture of mashed dates with milk. By the time of the Kitab al-Tabikh (Book of Dishes)—compiled in Baghdad in 1226 but based on a collection of ninth-century Persian-inspired recipes—there were so many versions of the confection that the chapter on “halwa and its varieties” entailed nine recipes, including barad (Arabic and Hebrew meaning “hail,” a confection of yeast fritters encased in a boiled honey and rose water candy), mukaffan (Arabic for “shrouded,” a confection wrapped in thin pastry), makshüfa (Arabic for “uncovered,” a boiled confection of ground nuts, sesame oil, and saffron), and several types of almond paste sweets. The book also contained an entire chapter devoted to judhab/gudab, sweetened grain dishes dating back to at least the ninth century.

“An anonymous thirteenth-century cookbook from Moorish Spain contained a recipe for halwa al (excellent confection), consisting of sheets of boiled sugar, honey, sesame oil, and flour that were rolled out; spread with ground pistachios, sugar, and rose water; topped with a second candy sheet; and cut into triangles. Sephardim enjoyed similar confections, including turron (a nougat lightened with egg whites) and azuqaques (from “sugar,” an almond confection).“

Marks isn’t concerned with South/South-East Asian halvas, but his account points to many halva types–including ones that have deep fried or pastry-like variants–whereas really in South Asia and especially India the diversity is of ingredients. Truly, both Indian home kitchens and halvais can prepare halvas out of just about anything, but there’s always as Davidson says a base ingredient, a sweetener, a fat, flavorings, and possibly nuts or other textural additions. Milks could be subcontinental innovations, serving at once as sweetener, fat, and flavoring, but they are not always essential. The process is long, slow, mixing and cooking until a thick paste-like confection emerges.

So, then, to my original theory–what we loosely call a “jam” or the Filipinos call a “halaya” is really a halva by another name and although I’m struggling to establish etymological connections, historical ones and the similarity of constituent ingredients, production processes, and even uses could well point to the idea of “halva” as a base for both.

Ube Halaya and Rasavalli Kizhangu Kali

The process of making both kali and halaya are markedly similar: sugar, coconut milk, and a lot of stirring, a lot of elbow grease, as my Filipino friends would often say. Condensed milk is likely a much later addition, possibly flavor-adding, a short-cut in an otherwise arduous cooking process surely, and possibly the distinction of the nuns’ halaya at Good Shepherd Convent. The Jaffna Tamil rasavalli kizhangu kanji indicates a simpler, more porridge-like dessert, whereas taking it to the form of a kali as I’ve done makes it much more obviously akin to halaya. Bree Hutchins, Hidden Kitchens of Sri Lanka, indicates a recipe that uses sago/javvarisi, possibly for binding or texture, but I’m not at all sure that is needed. Far more intriguing is the final addition of roasted urad dal, which would be the only major distinguishing element in the two recipes.

Since I started my explorations with ube in Manila, carting back the purplest of purple yams to grow in the garden–like the man in the old Jewish story retold by Uri Shulevitz, who learns that “Sometimes one must travel far to discover what is near”–I’m offering the halaya recipe as it’s known today but with the following possible variations:

• You can skip the condensed milk entirely and replace with more thick coconut milk and sugar. This gives you something close to what we imagine is the “original” halaya and rasavalli kizhangu kali, pared down to their minimalist bests.
• You can use dayap if you have access (or some other equally flavorful lemon like the Meyer) to impart a more uniquely Filippino taste.
• Or you can sprinkle in a small amount of roasted, powdered ulundu/urad dal to give it a more distinctly Tamil fragrance.
• You can leave the pudding runnier and serve it in a bowl, or cook it down further and set it in moulds in the spirit of the old jaleas or just like a good warm halva with cold ice cream (noteworthy here that ube halaya is a key component in the famous halo halo, alongside ice cream!)
• You could even steam the puddings wrapped in pandan (or, simpler, in steamers lined with pandan leaves) for a wonderful extra layer of fragrance and taste. [I’ve used the same method of wrapping as is used for chicken pandan, and served with a spoon to eat right out of the top opening, or cut the pandan wrapping open with a scissors — and spoon some fresh coconut milk on top]

Whatever your choices, consider also that there is considerable evidence for Tamil and Sanskrit, Hindu and Buddhist influences traveling to what we know today as the Philippines from the 2nd to the 14th centuries possibly via Indonesia/Malaya, though a lot of that was wiped out by missionizing. So it’s not at all unthinkable that an idea of a confection made by cooking a locally beloved, so-flavorful, and visually stunning yam could have emerged from just such a confluence. What is truly amazing, considering that theory, is that the confection has remained almost the same in both regions, down to this very day.

The process, in pictures: