Do It Yourself Tawny

The past three days Rupert Symington has been entertaining a group of American sommeliers and wine writers at Malvedos, showing them where our wines come from – the Douro, the quintas and vineyards, the wineries, and finally, last night, the blending process.

10 Year Old Tawny and its 4 component winesRupert set out bottles of Graham’s 10 Year Old and 20 Year Old Tawnies, and alongside each lined up four sample wine bottles marked A, B, C and D, then poured out a tasting glass of each.  He then announced, “Right, Writers against Sommeliers:  duplicate that finished wine using the component wines to blend your own sample.”  He told them they should be able to do it on smell alone, taste was actually irrelevant.  Funny, no one seemed to pay attention to that last bit.

Precision blending to get just the right balance of qualities

As the two teams set to work, it was interesting to observe that the Writers first sniffed and tasted, and began to work out the blend based on tasting notes – one component was decidedly raisinny, another very floral and fruity, a third was citrus, tangerine – and tried to assess to what degree the aroma qualities prevailed in the finished wines.  The Writers were also focussing on the mouth feel, deciding a proposed blend was not creamy enough, and trying to identify which component wine would increase that quality in the finished blend.

20 Year Tawny on the right, could do better on the left

Meanwhile, the Sommeliers, on the front line of restaurant service, first took a very commercial approach, assessing the likely age of the component wines and figuring that on a cost basis the finished blend might be weighted toward the younger and implicitly less expensive lotes.  The Sommeliers also focussed on colour early on, deciding which component wines were driving the the colour of the finished blend, whereas the Writers only assessed colour after they had assembled a blend devised on aroma, taste and mouth-feel qualities.

The Sommeliers also observed that one wine of the four constituents for the 20 Year Old was distinctly sweeter and more concentrated, and perhaps a bit musty, “but we did enjoy it!”  Rupert later explained that particular wine was what is known as a rancio, a very sweet old sticky wine, which is used to touch up older blends, in fact that quality is an identifier of age in any style of port.

When the Writers were happy with their 10 Year Old blend and the Sommeliers their 20 Year old, they noted the percentage of each component wine in their finished blends and were just feeling confident and enjoying the results when Rupert told them to switch wines and do it again, but this round in only 5 minutes’ time.

They then set to it again, with more confidence and speed.  One team suggested dropping an Alka-Seltzer tablet in a sample to throw off the other team, but no one had any handy, thankfully.  Amongst the comments wafting about:

Trial blend in the beaker, ready to compare and test

This one’s missing some red…
Well, whatever it is, it tastes good!
We need some citrus…
There’s that genius moment when you make a mistake, and it’s right…
I need some body…
Yessssss…. Colour!

Before discussing the blends, Rupert tasted the samples, and as he put down one glass he made a very satisfied noise and commented, “Tell you what, that’s a pretty good 20 Year Old.”

Magic formula number two, then...

Ultimately, the Writers prevailed, coming closest to the finished Graham’s product in both their 10 Year Old and 20 Year Old blends.  The winning strategy was to focus on the texture, body and aroma qualities of the finished wine.

The exercise was an eye opener for the group, who not only learned more about the blending process, but were eager to be able to replicate the experience for their customers as a way of emphasising the importance, power and sheer magic of blending in Port wines.  It also seemed probable that Charles will shortly receive a half-dozen applications to work in our blending and tasting room.

Share this post

Leave a Reply

Your email address will not be published. Required fields are marked *