top of page
  • Writer's pictureAndrew Russell

The Devil is in the Detail: The Precarious Task of Washing and Soaking Rice

Updated: Apr 29, 2021

There is a very famous saying amongst sake brewers that goes: 1st koji, 2nd moto, 3rd moromi. This is a list of three key stages in the sake brewing process in order of their importance. Repeated like mantra within the walls of the numerous kura throughout Japan, to dispute these words of wisdom is to argue against centuries of proven success in brewing one of the world's most sophisticated and refined alcoholic beverages. However, there is one other crucial stage that is so often overlooked, one that is intrinsically linked with all of the others and has the biggest influence in determining the success or otherwise of each and every one them. Yes, the techniques used when washing (senmai) and soaking (shinseki) polished rice are the glue that binds all of the numerous different stages of sake brewing together. Furthermore, the accuracy during these stages will permeate into every other step of the process and can make or break any batch for better or worse.

However, despite the rather unassuming descriptions, senmai and shinseki, often just collectively referred to as "rice washing" (kome-arai), are actually far more interesting and complex than one might think. These crucial stages, always carried out consecutively, are as difficult to master as any of the others, often requiring good old fashioned experience and intuition in order to complete successfully. It is perhaps for this reason that most young brewers start learning the required techniques early in their careers, and often their first responsibility is in helping a more senior member with this time consuming and often stressful daily task.

This was certainly the case with myself when I first started brewing, initially assisting the toji, and then later my most senior co-worker (kashira). Having first assumed it was just as straight forward as it sounded, it was early in my first season when I unfortunately came to understand the hard way just how difficult a task it can be, and the full extent of how it can negatively affect the latter stages of brewing if not done accurately. I will use this experience throughout in order to explain this process more thoroughly. But before I go into any details of where I went wrong, and ultimately what the consequences were, lets first look at what this particular process involves, and why it is so important to carry out with the upmost diligence and as free from error as is possible.

Rice Washing 洗米

Before polished rice can be used in the various different stages of the brewing process it must first be thoroughly washed. This is done primarily to remove unwanted particles of rice leftover from its often lengthy time in the polishing machine (seimai-ki). Without doing this, these particles would cause the rice to clump together the following day during steaming, generally causing no end of headache for the brewers. In fact, there is even a saying for this in Japanese, with brewers often heard muttering "sabake ga yokunai", a phrase that refers to rice that clumps together and is difficult to handle. Furthermore, not cleaning the rice thoroughly before continuing to the later stages can cause off flavours in the final sake. Therefore, the first step is always to wash away all of these undesirable elements which helps ensure a smooth transition into the later stages.

Nowadays, as is the case in my workplace, the job of washing is often done with the help of a special machine known as a Woodson ウードソン. The rice, usually separated into 10kg batches, is poured into the body of the machine as water is pumped in with considerable pressure from one side to create a sort of whirlpool effect. As the rice is being swirled around inside, the unwanted rice particles are being collected in the water and pushed out through the top and drained away. After a few minutes, a narrow sluice is opened at the bottom and the washed rice is collected in a net bag. It is then rinsed again with a large shower before finally being ready for the next stage.

Rice Soaking 浸漬

The objective of this next stage is to give the rice the best possible moisture content for the later stages of brewing. The making of koji (seikiku), the yeast starter (moto-tsukuri), and the main mash (moromi-tsukuri) all require steamed rice with different moisture levels to be successful, and therefore must be adjusted accordingly. Of the three, the production of koji is the most sensitive to variations in moisture and also typically requires a slightly higher level than the other two batches. Ultimately, the main objective is to prepare the rice in the optimum condition for steaming the next day. Good steamed rice is dependent on this stage, and if done correctly will lead to a non-sticky, elastic kernel with a slightly hard surface and contrastingly soft centre.

There are numerous different techniques and schools of thought applied in order to achieve this, and the best method differs from place to place. However, they all involve soaking, or maybe more apt to say steeping, the rice in water for a specific time in order to achieve a target weight. In the case of the kura where I work, we place 10kg bags of rice directly into a large metal container filled with water. Using a stopwatch or timer, from the moment it is submerged a countdown begins until it reaches a pre-calculated time that, in theory, should allow the rice to absorb just the right amount of water and achieve its target moisture rate (kyusui).

As noted previously, rice intended for the production of koji tends to require a slightly higher rate than that destined for adding directly to the main mash (kake-mai). However, this very much depends on the style of sake being produced. On average though, kake-mai can have a target of anywhere in the region of 28-35% moisture content, whereas koji typically starts around 33% or higher. Once the estimated time has been reached, the bag of rice is pulled from the water, quickly drained or vacuumed, and placed on a simple set of scales to verify the final weight by which the moisture content can then be calculated.


The reason that I have used such non-committal vocabulary as "target", "in theory" and "estimated" is that there is a formidable array of factors that affect how the rice behaves during steeping. A far-cry from just making a few rough estimations, even a slight variation in weight can have huge implications during the later stages. For example, an increase in moisture before the koji spores are sprinkled on the steamed rice will drastically change how the koji will grow during its development stage. Therefore, brewers must be extremely precise, first factoring in the temperature of the water, the cultivar of the rice they are using, its polishing ratio, and, if possible, even the moisture content inside the kura. All of these factors will alter the time it takes to reach its target weight, and in the case of some cultivars, the timing required will be vastly different depending on the conditions present on that day.

Of the four mentioned above, water temperature is perhaps the most predictable to deal with. However, as even slight variations in temperature will lead to the rice absorbing water at completely different rates, care must be taken to check the correct temperature before soaking can begin. For example, warmer temperatures will result in the rice absorbing at a much quicker rate than with colder temperatures. These warmer conditions also increases the chance that the rice kernels will split, which causes them to melt faster in the main mash. As the opposite is true for colder water, lower temperatures are preferable, particularly with higher polishing ratios (seimaibuiai) and certain more difficult rice cultivars. In general, the ideal water temperature lies somewhere between 10-15° Celsius, and most breweries will keep detailed data on how much time it took for each year's rice to reach their desired moisture content.

Factoring in the unique characteristics of the numerous different rice cultivars is unfortunately not as straight forward to manage as the previous variable. Each year's harvest produces rice that will behave differently, and the only method brewers have at their disposal in order to anticipate these changes is to check the previous year's data for a rough indication then simply observe the rice in the water and see how it progresses. If the kernels start cracking and turn completely white, then they know that they have already absorbed too quickly. Therefore, it is crucial that the first batch of each variety used in a season is monitored carefully to see how it behaves. An experienced toji will know roughly by sheer intuition how each cultivar is likely to behave, but nobody really knows for sure until it is tested.

To give an example, at the beginning of this season (BY30) I remember washing two batches of rice in sequence that were the same cultivar and had identical polishing percentages. The only difference was that they were from different rice fields, albeit fairly close to one another. One might then assume that they would behave pretty similarly during soaking. However, this turned out the opposite, with one taking a further two full minutes to reach the same weight. When you consider that most rice varieties average anywhere from seven to seventeen minutes to reach their desired moisture content, this time difference is both substantial and highlights just how big a variation there is even from within the same cultivars.

The rice in question was no less than the crown jewel of the Okayama sake industry, the highly prized Omachi-mai. Much loved by consumers for its wonderful nuanced aromas and flavours, it is also the source of much stress and anguish for the brewers who have to work with it. One of the main reasons for this love and hate relationship is specifically down to how the rice behaves during washing and soaking. This is due to the fact that Omachi-mai has a very large shinpaku, the lovely white centre of the rice that can produce such wonderfully active koji . Although having a big shinpaku is effectively what separates expensive brewers rice from simple table rice, it does also mean that the rice kernels are much more susceptible to cracking during shinseki.

Ironically then, rice that is considered of lesser quality, for example those with smaller shinpaku, are actually often much easier for the brewers to work with during washing and soaking. Even the mighty Yamada-nishiki, differing to Omachi in that it is a hard rice as appose to a soft one, is still a bit of a tricky customer due to its nutrient rich centre. This is especially the case when the rice has been heavily polished, as the hard outer protective layer that slows water absorption is reduced, or in some cases removed. This results in some incredibly fast absorbing rice that if not soaked under the strictest of time management will result in insufficient or surplus moisture, leading to undesirable traits in both the koji and the rice added directly to the mash.

Learning from Mistakes

Nobody ever said that sake brewing was easy though, and figuring out how to get the optimum moisture content into each grain before steaming is just one of the many challenges that have to be faced each season. Accuracy in the timings are paramount, and the aim is to be as close to the target weight as possible. During the days when we are washing large quantities of rice, sometimes up to 550kg in one afternoon, you cannot afford to have a lapse in concentration for even a few seconds or you run the risk of messing up the timings. Unfortunately, when doing such a precarious task, tiredness does set in and mistakes do inevitably happen. And as I mentioned previously, I have had such an experience.

Unfortunately for me, my mistake occurred during the soaking of rice destined for use in making the much more unforgiving koji-mai. Moreover, it just so happened to be on a day when we were using top-grade Yamada-nishiki rice from one of Okayama's most prized regions for rice cultivation. Just the slightest lapse in concentration led to an error that went on to have implications for days to come. As I mentioned previously, most breweries employ slightly different techniques, and at my workplace we use a fairly high intensity, but ultimately time efficient, method whereby the bags of washed rice are placed in order of when the soaking begins. We stagger the timings over two minute intervals and use a simple clothes peg to mark the order in which they are submerged into the water. Once the desired time is reached, each bag is pulled from the water in sequence and the peg is simply attached to the next bag in line.

Trying to be clever, I switched the peg to the next bag in line a full thirty or forty seconds before it was due to be pulled from the water. I had maybe pulled around thirty bags from the water by this point, and as I stood there waiting I lost concentration. Without thinking, I pulled the bag with the peg attached which was of course a full two minutes ahead of schedule. As we have already ascertained, two minutes when using highly polished rice is significant, and it was immediately noticeable on the scales. As the work cannot simply be paused once we start, it led to a hectic scramble to get back in sequence and try to minimise any further damage.

The next day, still not fully appreciating the extent of my mistake, we carried out the task of tane-kire whereby we sprinkle a mould (koji-kin) over the rice to encourage it to grow into koji. However, the steamed rice was woefully lacking in moisture, despite the increase that it received as a result of steaming. The koji-kin seeks out moisture before it grows, and as the rice's composition was hard and lacking in moisture, the spores did not penetrate the rice kernels properly resulting in koji lacking in enzymes, the crucial element that transforms starch into sugar and makes it possible to produce alcohol from rice.

However, as my very forgiving toji said a few days later, you have to make these mistakes in order to understand the importance of accuracy in sake brewing. Mistakes are of course better avoided, but when they do occur they should be used as an opportunity to learn what the various consequences are and how they affect the later stages in the process. Ultimately, they should be used to try and further streamline the processes so that they are not repeated in the future.

The Devil is in the Detail

Having witnessed firsthand the implications of having insufficient moisture during the making of koji it has made me fully aware of just how crucial a stage it is. Not only does the moisture content after steaming affect the composition of the rice, it also fundamentally determines the finished nature of the koji. As the final composition of the koji is said to be the most important factor in determining the final flavours and aromas of the sake, brewers must ensure that the steamed rice is just as it should be before the task of cultivating koji can begin.

However, although not quite to the same extent, the same applies for kake-mai, or the rice being added directly to the mash. It is during the last of these three important stages mentioned in the opening paragraph when all the accuracy during the previous steps will pay dividend. In short, it all comes down to how quickly or slowly the rice melts in the main mash (moromi). Of course this is a topic worthy of its own discussion, and certainly goes beyond this one. However, rice that melts too quickly during the main fermentation stage will often result in sake with harsher, sometimes astringent flavours. Conversely, if the rice is too hard then it will not impart all the wonderful flavours and amino acids that make sake so appealing.

In summary, every stage on that previously mentioned list of koji, moto, and moromi all require the rice to be of a certain specification. As the brewing process gathers momentum, the final stage of main mashing (sandan-jikomi) is a careful balancing act with how the rice dissolves. Balance requires precision, and therefore all of these stages can either strengthen the brewing chain or weaken it depending on how accurately kome-arai is completed. Like a crescendo that grows and grows, each batch of sake is build up in gradual stages increasing in size and scale. With each stage dependent on the quality of the last, it stands to reason that the desired outcome of any given batch of sake is achieved though their continuous accuracy. At each and every point, the results of kome-arai permeate into all of the other stages, and as it always seems to be with sake brewing, it is these little details that make all the difference.

If you found this article interesting or useful, please consider making a small donation via the link below. Just a small contribution is greatly appreciated and helps me create more content in the future.

Thank you.

467 views0 comments

Recent Posts

See All
bottom of page