The great wine filtering debate
The question first came up, at least for me, due in no small part to a book written by Kermit Lynch back in the late 80s called "Adventures on the Wine Route."
Kermit, who is a wine importer, and since the book, a very successful one, made his name by importing wines that had not been filtered. He went out of his way to have the wineries make special unfiltered releases just for him.
This captured the imagination of the public, and a new marketing ploy was born. Unfiltered/Unfined wines. These wines were presumed to be of better quality, because they had not had some part of them stripped out.
I have always had a problem with this theory. My skepticism came from a series of blind tastings that I conducted with wines that were available both from Kermit, and in a filtered form.
There was a huge difference in the wines, although not the difference that Mr. Lynch was no doubt hoping for. The panel concluded that the unfiltered wines were "dirtier" and exhibited more flaws.
I have never been a huge fan of Kermit Lynch imported wines for this very reason. No question it is a matter of taste. There are those that appreciate the minor flaws of some wines, saying that it adds complexity.
I tend to find that the people that say this are either making flawed wines, or grew up drinking them. Technology has cleaned up wine, even since the 80's, but the unfiltered debate still rages.
I recently was fortunate enough to draw Greg La Follette of Tandem Winery into a round of discussion. Some of his wines (notably the top tier) are unfiltered, others are not.
Greg is considerably more schooled in the science of winemaking that I could ever hope to be, and so his answers to my questions tend towards the very technical.
Here are my original questions:
My opening question is:
No less an authority than Emile Peynaud, the renown Bordeaux oenologist, states in his "Knowing and Making Wine" - "It may be stated that that the mechanical action of filtering has never had a negative influence on quality. To suggest the contrary would mean conceding that the foreign substances in suspension and their impurities that form the lees, which filtration is precisely designed to remove, have a favorable taste function."
Is it your contention that indeed these foreign substances that would normally be removed by careful filtering add positive tastes to the wine? And if so, what are these substances and what size are they that they could not be allowed with the proper filter system?
If it is not your contention that there are flavors that are lost due to filtering, what is lost?
My second, and related question is:
Filtering and fining are tools. Like any tool they can be misused, or used with great skill. I realize you do not eschew filtering and fining as a matter of course, but when you do elect to avoid these steps is it because you feel that no matter how carefully used, fining and or filtering have sensory consequences for the wine? If so, what consequences?
In those cases where you elect not to fine, does this lead to additional racking? If so how do you balance the sensory consequences of inherent oxidation during racking vs. any consequence of fining? If there is no increase in racking when you do not fine, what method of clarification are you using? Are you worried about creating wines that are not brilliantly clear (as is the modern expectation)?
What is your stance on sterility of wine? If you believe a wine should be sterile when bottled, what are you doing to achieve this in those instances you do not filter. If you are not concerned with sterility, why not?
Greg's reply was:
There has been a huge body of research done since Emile Peynaud's time on macromolecular level of influence of fining and filtration. Fining is frequently based on charge association; filtration also has a charge association with it as well as direct elimination of a large number of macromolecules that could not have been forseen [sic] in Emile's time.
Rose-Marie Llaubers (Canals) has done a lot of work on this out of the University of Bordeaux II on this area and the upshot is this: filtration does remove some macromolecular components that, in and of themselves, may not have a direct mouthfeel component, but these macromolecules help to stabilize flavor and texture and do in fact contribute to this important area of wine organoleptics. This is only one area of the contribution of textural and flavor components affected by filtration. Other researchers are Ferrari, Feuillat, the University of Dijon the AWRI.
While this response sent me into a researching frenzy, and taught me a lot in the process, it didn't really answer my questions. I responded with the email below, but Mr. La Follette's attention seems to have been diverted elsewhere, and I never got a follow up.
Here is the email I never got an answer to (if Greg does ever follow up again, I will post his response).
-------- My response --------
First off, you didn't answer my questions about racking or whatever techniques you are using to clarify. Specifically, is there any down side to your clarification methods, and is the risk so minimal as to make them worth the risk vs. fining/filtering. Please review the other original questions below [now above - ed.], and take a stab at them when you can.
These related questions are important because there is a risk assessment issue, which I suspect you take into consideration, since you do filter and fine some of your wines.
Makes perfect sense what you are saying about Charge Association. Is it possible, and or has it already been done, to create a fining system that changes polarity of the charge, or other technical approaches to dealing with the charge association?
When you say that there are some macromolecular components that have a positive organoleptic effect, what are the components specifically, and what is the effect for a trained taster?
Is it possible to have positive effect on the wine by filtering above the threshold of the macromolecular components, or are they larger than the other (e.g. colloidal matter) components you may want to remove?
Fining, as you point out is frequently based on charge association, what about this, if anything do you object to? Is it the lack of precision in being able to target specific particles? Is there something inherently destructive to the mouthfeel or other organoleptic sensations?
You mention stabilizing flavors, since this is a test over time, how long has the research into this been going on? You mention that they do not have a direct mouthfeel component, so it is not a haptic sensation, is there indeed a direct organoleptic trait, or is it something more nebulous, such as preserving flavors over time (and if so how is this effected by the TA)?
I still have some pretty good questions that have gone unanswered, if anyone wants to take up the gauntlet and return to the debate, you know where to find me.
With many thanks to Greg La Follette for his time in answering my original volley.